gsoc2013-evolution - Google Summer of Code 2013 - GNOME

diff options

author	Christopher James Lahey <clahey@helixcode.com>	2000-11-02 23:38:32 +0800
committer	Chris Lahey <clahey@src.gnome.org>	2000-11-02 23:38:32 +0800
commit	8e47e60edaca6ef49b04a276f0993d6691b62b4d (patch)
tree	1dee5e600a5d20363e607389c68ee71ae84816f7 /po/fi.po
parent	9df50a569cc7ef527799dc050feeb6a8e54439bd (diff)
download	gsoc2013-evolution-8e47e60edaca6ef49b04a276f0993d6691b62b4d.tar.gz gsoc2013-evolution-8e47e60edaca6ef49b04a276f0993d6691b62b4d.tar.zst gsoc2013-evolution-8e47e60edaca6ef49b04a276f0993d6691b62b4d.zip

Fixed a missing ".

2000-11-02 Christopher James Lahey <clahey@helixcode.com> * fi.po: Fixed a missing ". svn path=/trunk/; revision=6350

Diffstat (limited to 'po/fi.po')

-rw-r--r--

po/fi.po

1 files changed, 1 insertions, 1 deletions

2002-11-15 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (imap_connect_online): If the namespace is at/below INBOX, check for the INBOX explicitly (since it obviously won't show up in a LSUB INBOX.*). If either INBOX is not returned in the response or if the folder flags contain \NoSelect, subscribe to INBOX and then try LSUB again. 2002-11-11 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-tohtml.c (html_convert): Rewritten. Much much much cleaner implementation now, though uses malloc/free more often than I'd like. 2002-11-11 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (get_message_simple): Use g_strerror when setting an exception string (we need it to be in UTF-8). * providers/pop3/camel-pop3-store.c (pop3_try_authenticate): Use g_strerror when setting an exception string (we need it to be in UTF-8). * providers/pop3/camel-pop3-folder.c (pop3_refresh_info): Use g_strerror when setting an exception string (we need it to be in UTF-8). (pop3_get_message): Same. * providers/local/camel-spool-summary.c (spool_summary_sync_full): Use g_strerror when setting an exception string (we need it to be in UTF-8). (spool_summary_check): Here too. * providers/local/camel-spool-store.c (construct): Use g_strerror when setting an exception string (we need it to be in UTF-8). (get_folder): Same. (scan_dir): Here too. * providers/local/camel-spool-folder.c (spool_lock): Use g_strerror when setting an exception string (we need it to be in UTF-8). * providers/local/camel-mh-summary.c (mh_summary_check): Use g_strerror when setting an exception string (we need it to be in UTF-8). * providers/local/camel-mh-store.c (delete_folder): Use g_strerror when setting an exception string (we need it to be in UTF-8). * providers/local/camel-mbox-summary.c (summary_update): Use g_strerror when setting an exception string (we need it to be in UTF-8). (mbox_summary_sync_full): Here too. (mbox_summary_sync_quick): Same. (mbox_summary_sync): Also here. (camel_mbox_summary_sync_mbox): Again here. * providers/local/camel-mbox-folder.c (mbox_lock): Use g_strerror when setting an exception string (we need it to be in UTF-8). (mbox_append_message): Same. (mbox_get_message): Here too. * providers/local/camel-maildir-summary.c (maildir_summary_load): Use g_strerror when setting an exception string (we need it to be in UTF-8). (maildir_summary_check): Same. * providers/local/camel-maildir-store.c (get_folder): Use g_strerror when setting an exception string (we need it to be in UTF-8). (delete_folder): Same. (delete_folder): Here too. * providers/local/camel-local-summary.c (local_summary_sync): Use g_strerror when setting an exception string (we need it to be in UTF-8). * providers/local/camel-local-store.c (get_folder): Use g_strerror when setting an exception string (we need it to be in UTF-8). (create_folder): Same. (xrename): Here too. (rename_folder): And here. (delete_folder): Also here. * camel-provider.c (camel_provider_init): For debugging printfs, we want to use normal strerror (we want locale charset, not UTF-8). * camel-movemail.c (camel_movemail): Use g_strerror when setting an exception string (we need it to be in UTF-8). (movemail_external): Same. (camel_movemail_copy_file): Here too. (camel_movemail_solaris): Also here. * camel-mime-utils.c (rfc2047_decode_word): For debugging printfs, we want to use normal strerror (we want locale charset, not UTF-8). (header_encode_param): Same. * camel-mime-part-utils.c (convert_buffer): For debugging printfs, we want to use normal strerror (we want locale charset, not UTF-8). * camel-lock-client.c (camel_lock_helper_init): Use g_strerror when setting an exception string (we need it to be in UTF-8). * camel-data-cache.c (camel_data_cache_remove): Use g_strerror when setting an exception string (we need it to be in UTF-8). * camel-tcp-stream-raw.c (flaky_tcp_write): For debugging printfs, we want to use normal strerror (we want locale charset, not UTF-8). (flaky_tcp_read): Same. * camel-gpg-context.c (gpg_ctx_op_step): For debugging printfs, we want to use normal strerror (we want locale charset, not UTF-8). * camel-service.c (camel_gethostbyname): Use g_strerror when setting an exception string (we need it to be in UTF-8). * camel-lock.c (camel_lock_dot): Use g_strerror when setting an exception string (we need it to be in UTF-8). (camel_lock_fcntl): Same. 2002-11-07 Not Zed <NotZed@Ximian.com> * camel-mime-part.c (camel_mime_part_set_content_type): constify the content_type. 2002-11-07 Jeffrey Stedfast <fejj@ximian.com> * camel-i18n.h: Added. * camel-object.h: #include camel-i18n.h instead of gnome-i18n.h * camel-lock.c: Same. 2002-11-07 Rodrigo Moya <rodrigo@ximian.com> * camel-object.h: removed gnome-defs.h, it does not exist. 2002-11-05 Not Zed <NotZed@Ximian.com> * Makefile.am (camel_lock_helper_LDADD): Created temporary link list, we need to link with something for gettext. * camel.h: Remove gstring-util.h and hash-table-utils.h. * camel-text-index.c: (text_index_normalise): Changed for g_utf8_strdown api change. * camel-search-private.c: * camel-mime-utils.c: * camel-mime-part-utils.c: * camel-html-parser.c: * camel-charset-map.c: Include glib/gunicode.h from glib instead of gal. * camel-filter-driver.c: Remove include of gtk/gtk.h, should never have been there. 2002-11-01 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (g_string_append_len): Removed. Glib2 has this function. * providers/local/camel-local-store.c (delete_folder): Use g_path_get_basename instead of g_strdup (g_basename (filename)). * camel-uid-cache.c (camel_uid_cache_new): Use g_path_get_dirname since g_dirname has been deprecated. 2002-10-31 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (camel_gpg_context_class_init): Set the virtual method pointers to the import/export methods. (camel_gpg_context_init): Set the key_protocol string. (gpg_hash_to_id): Handle 2 more hash types. (gpg_id_to_hash): Same. (gpg_ctx_op_step): Slight fixes to support import/export. (gpg_ctx_parse_status): Fix to hack around the fact that importing keys doesn't write to stdout. (gpg_import_keys): Implemented. (gpg_export_keys): Implemented. * camel-cipher-context.c (camel_cipher_context_class_init): Hook up default virtual methods for import/export. (camel_cipher_import_keys): Implemented. (camel_cipher_export_keys): Implemented. 2002-10-31 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Removed hash-table-utils.[c,h] from the build. * hash-table-utils.[c,h]: Removed. * string-util.c: Imported g_strcase[hash,equal] into here so we can remove hash-table-utils.[c,h]. * camel-medium.c: Removed #include "hash-table-utils.h" * camel-mime-message.c: Same here. * camel-mime-part.c: And here. * camel-session.c: Here too. * providers/imap/camel-imap-store-summary.c: #include string-utils.h instead of hash-table-utils.h * camel-charset-map.c: Same. * camel-folder-summary.c: Here too. * camel-provider.c: Again here. * camel-store-summary.c: And again... 2002-10-31 Jeffrey Stedfast <fejj@ximian.com> Remove a ton of useless snot. * Makefile.am: Remove gstring-util.[c,h] from the build. * gstring-util.[c,h]: Removed. * string-utils.c (string_equal_for_glist): Removed. (string_split): Removed. (string_trim): Removed. (string_prefix): Removed. (string_unquote): Removed. (strip): Removed. * hash-table-utils.c (g_hash_table_generic_free): Removed. g_str[n]casecmp functions are deprecated in glib2. * string-utils.c (strstrcase): Use strncasecmp instead of g_strncasecmp. * hash-table-utils.c (g_strcase_equal): Use strcasecmp instead of g_strcasecmp. * camel-smime-utils.c (camel_smime_is_smime_v3_signed): Same. (camel_smime_is_smime_v3_encrypted): Here too. * camel-sasl-digest-md5.c (decode_data_type): And here. (parse_server_challenge): Again here. * camel-pgp-mime.c (camel_pgp_mime_is_rfc2015_signed): Same. (camel_pgp_mime_is_rfc2015_encrypted): Same * camel-mime-part-utils.c (check_html_charset): Here too. * camel-folder-summary.c (camel_system_flag): Same. 2002-10-25 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (gpg_ctx_parse_status): Don't prematurely set gpg->complete to TRUE. Let gpg_ctx_op_step () set it to TRUE when gpg closes the status-fd instead. 2002-10-24 Not Zed <NotZed@Ximian.com> ** For bug #31647 and bug #31456. * camel-store-summary.c (store_info_string): for STORE_INFO_NAME, skip the leading /. * providers/imap/camel-imap-store.c (parse_list_response_as_folder_info): Remove jeff's last patch, and use the store summary to create the name and path of the folderinfo so it manages namespace issues. (get_folder_info_online): Just pass @top directly to build_folder_info always, since namespace is mapped to 1 tree level. (imap_build_folder_info): Remove jeff's last patch, dont strip leading /'s, they shouldn't exist. (imap_connect_online): Remove adding the INBOX here, we add it later. (get_subscribed_folders): Make sure INBOX is always in the list. some imap servers dont seem to let you subscribe to it(?), so always have it act as subscribed. * camel-store.c (camel_folder_info_build): back out the last 2 patches from Jeff (for #31456) to get the original behaviour. (camel_folder_info_build): When creating a fake parent, dont strip the namespace from the full_name. malloc keys in hash since we dont have them anymore. (free_name): Helper to free names. * providers/imap/camel-imap-store-summary.c (camel_imap_store_summary_namespace_new): Canonicalise the namespace (strip trailing dir_sep), and change the path to remove any /'s. (camel_imap_store_summary_namespace_find_path): (camel_imap_store_summary_namespace_find_full): new, find namespace by path/full name. (camel_imap_store_summary_full_from_path): Changed to a simple wrapper around path_to_full, after checking namespace. (camel_imap_store_summary_add_from_full): map the namespace if present. (camel_imap_store_summary_path_to_full): If namespace exists, unmap it. 2002-10-18 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_folder): Canonicalise the source_uri to not have a path. Fixes bug #32268. 2002-10-17 Jeffrey Stedfast <fejj@ximian.com> Possible fix for bug #32270 * providers/pop3/camel-pop3-store.c (try_sasl): If we get an I/O error, we should not be setting the CANT_AUTH exception but should instead be setting the SYSTEM exception. Also check for EINTR which signifies a USER_CANCEL exception. (pop3_try_authenticate): If the auth mechanism isn't supported, don't set the CANT_AUTH exception since then we will loop and try again with the same data which will just cause an infinite loop. (pop3_connect): Simplified a bit. 2002-10-17 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (camel_certdb_nss_cert_set): If ~/.camel_certs doesn't exist, create it. 2002-10-15 Not Zed <NotZed@Ximian.com> * camel-tcp-stream-ssl.c (cert_fingerprint): helper, build fingerprint. (camel_certdb_nss_cert_get): Helper for nss certs. Lookup cert. As well as fingerprint, the whole raw cert is checked for validity. (camel_certdb_nss_cert_add): Add an nss cert to a certdb, also saves the cert by fingerprint in ~/.camel_certs/. (ssl_bad_cert): Changed to use above functions to simplify logic. (ssl_bad_cert): Also added non-compiled code which mimics what mozilla does, but it doesn't work right :-/ (camel_certdb_nss_cert_set): Save the raw cert associated with a cert. * camel-certdb.c (certdb_cert_free): Free the raw cert data if set. 2002-10-14 Not Zed <NotZed@Ximian.com> * camel-file-utils.c (camel_file_util_encode_string): Encode a length of 0 as 1, not 0, to match the decode code. 2002-10-03 Not Zed <NotZed@Ximian.com> * camel-folder.c (filter_filter): Do the filtering ourselves. This is so a problem like a missing uid (which can happen legitemitely) doesn't prematurely abort filtering. Stop on all other errors. See #31667. 2002-10-16 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (camel_gpg_context_new): No longer takes a path argument. (camel_gpg_context_fianlise): No need to free ctx->path. (gpg_ctx_new): No longer takes a path argument either. (gpg_ctx_free): No need to free ctx->path, it's no longer used. (gpg_ctx_op_start): Don't stat the path anymore and change the first arg to execvp to "gpg" so that we use the shell's environment to find gpg. (gpg_sign): Updated. (gpg_verify): Updated. (gpg_encrypt): Updated. (gpg_decrypt): Updated. (gpg_ctx_get_argv): Removed some debug printfs. 2002-10-15 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (gpg_id_to_hash): If the hash id is NULL, return CAMEL_CIPHER_HASH_DEFAULT. Fixes bug #32229. 2002-10-08 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-command.c (imap_read_response): If the response from the IMAP server is "No", don't set the SERVICE_UNAVAILABLE exception, this makes error reporting in the UI for deleting IMAP folders that cannot be deleted inaccurate (ie, it reports "Cannot delete in offline mode" which is not the problem). 2002-10-07 Jeffrey Stedfast <fejj@ximian.com> Fixes bug #31752 * providers/smtp/camel-smtp-transport.c (connect_to_server): Don't forget to send another EHLO command to the server once we toggle into STARTTLS mode. (smtp_helo): Reset any flags set using the EHLO response and also any authtypes. 2002-10-06 Jeffrey Stedfast <fejj@ximian.com> Fixes bug #31681 * camel-mime-utils.c: Fix all mailing list regex patterns to allow any number of spaces *or* tabs as pre-padding for the header values. 2002-10-03 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part.c (write_to_stream): Use the content-object's rawtext flag to decide if it needs to be filtered through the charset filter instead of using the mime part's rawtext flag since this will never be set. Partial fix for bug #31655. 2002-10-03 Not Zed <NotZed@Ximian.com> * camel-folder.c (folder_changed): Dont get the filter driver inside the change_lock. Its not necessary, and can cause deadlock with the way the mailer gets the filter driver via g_mainloop. Should fix #31572. 2002-10-02 Jeffrey Stedfast <fejj@ximian.com> * camel-charset-map.c (camel_charset_iso_to_windows): Map us-ascii to windows-1252 also. 2002-10-02 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-folder.c (get_matching): Set *set to NULL if we dont get any matches. (imap_sync_online): If we get no matches, skip any work, also reorder some code to make it easier to skip. See #31031. * providers/imap/camel-imap-store.c (imap_check_folder_still_extant): Default to "TRUE", if the list command failed, it probably means a server problem, assume the worst. This makes imap_refresh_info not clear the exception and crash. Fixes crash of #31000. 2002-10-01 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-folder.c (imap_refresh_info): Make sure we pass the exception to imap_folder_selected(), otherwise failures can be lost. See bug #31000. 2002-10-01 Not Zed <NotZed@Ximian.com> * camel-folder.c (thaw): Add an assertion that the frozen count>0. (freeze): Same for >= 0. * camel-vee-folder.c (camel_vee_folder_remove_folder): Use the unmatched freeze_count when thawing folders removed from unmatched, rather than the folder's freeze_count. Might be related to #27391. 2002-10-01 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (decode_internaldate): Use strtol when decoding the timezone (since it can be negative) and don't forget to increment inptr to the start of the time (ie, don't leave inptr pointing to the year when decoding the hour:min:sec). 2002-09-30 Jeffrey Stedfast <fejj@ximian.com> Fixes bug #31456. * providers/imap/camel-imap-store.c (imap_connect_online): Don't LSUB "" "*", instead get both an LSUB containing the subfolders of the namespace and an LSUB of INBOX (assuming namespace was non-empty). This fix really has nothing to do with bug #31456 but is what should have been done in the first place. (parse_list_response_as_folder_info): Simplify a tad and strip extra leading /'s from fi->path. (imap_build_folder_info): Strip extra leading /'s from fi->path. * camel-store.c (camel_folder_info_build): Don't strip the namespace from the fi->full_name when hashing or creating fake parent folder-infos. Fixes a bug I found while trying to reproduce bug #31456. (camel_folder_info_build_path): Strip off extra leading dir_sep chars from the path. 2002-09-30 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-folder.c (imap_get_message): Remove debug content_info_dump, could cause a crash, and not really needed anymore. * camel-folder-summary.c (camel_content_info_dump): Check ci->type != NULL before dereferencing it. Should fix crash #31331. 2002-09-27 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (decode_internaldate): New function to decode the INTERNALDATE response from an IMAP server so we don't have to use my broken-date-parser routines. 2002-09-27 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (connect_to_server): NULL-check the streams before unreffing them in the case of a failure during ssl negotiations. * camel-tcp-stream-ssl.c (camel_tcp_stream_ssl_enable_ssl): Check SSL_ResetHandshake() for errors. Also force a handshake after we've reset the handshake state on the socket. 2002-09-27 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-folder.c (imap_update_summary): Use the summary's last uid as the one to fetch from, ignoring the cache. Use strtoul instead of atoi as well. * providers/imap/camel-imap-store.c (get_folder_counts): If we have the folder open, and the unread count has changed, refresh it. Should fix #30399 enough. Also in non-check-all mode, if we have the folder open, use it anyway. 2002-09-26 Not Zed <NotZed@Ximian.com> * tests/folder/test2.c (main): Treat spool as a local folder, so the ref checks work right. * providers/local/camel-spool-store.c (get_folder): Implement FOLDER_CREATE flag. (scan_dir): Dont free name on exception, its alloca'd. (scan_dir): If we start scanning from a file, just add that directly. (scan_dir): Allow empty files to also show up in folder list, as well as files starting with "From ". (get_folder): Use creat() instead of open() to create file. * providers/local/camel-spool-folder.c (camel_spool_folder_new): Check folder != NULL before writing to it. * providers/local/camel-local-store.c (create_folder): Handle a parent of NULL for creating top-level dirs. Part of #31186. * providers/imap/camel-imap-store.c (parse_list_response_as_folder_info): Store the folder flags in the store summary. (get_one_folder_offline): Set the noselect url parameter if this is a noselect folder, from the stored summary flags. #30877. 2002-09-23 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-charset.c (complete): Handle EILSEQ by eating bytes in the input buffer until conversion is successful. The old behaviour aborted when it got EILSEQ and just dumped the raw data. (filter): Same. Fixes bug #30873. 2002-09-24 Not Zed <NotZed@Ximian.com> * camel-mime-utils.c (mail_list_magic[]): Changed the Mailing-List rule to get the complete domain (* in wrong spot). #31087. * camel-text-index.c (text_index_compress_nosync): If we have no data written to a key, dont add it, or its partition header. 2002-09-23 Not Zed <NotZed@Ximian.com> * camel-block-file.c (sync_nolock): Mark root block dirty if we're going to sync it. (camel_key_file_finalise): de-count the active list if we close it. * camel-text-index.c (text_index_delete): Implement. (text_index_compress_nosync): Use index_delete when we're finished with the old one, so it is properly deactivated. * camel-index.c (camel_index_delete): Remove api call. Mark index deleted in new state variable. (camel_index_*): Wrap all calls in check for deleted state. 2002-09-23 Jeffrey Stedfast <fejj@ximian.com> * camel-folder.c (transfer_message_to): If the original message is deleted, reset the deleted flag when we're done. Fixes bug #30876. * providers/imap/camel-imap-folder.c (add_message_from_data): Decode the INTERNALDATE if we've got one. (imap_update_summary): Instead of requesting a list of specific headers, request HEADER.FIELDS.NOT (RECEIVED) to reduce bandwidth usage even more. (parse_fetch_response): Change slightly to allow HEADER.FIELDS or HEADER.FIELDS.NOT so that we can make changes in imap_update_summary and not need to keep changing this function to match. Also parse an INTERNALDATE response. 2002-09-22 Dan Winship <danw@ximian.com> * camel-mime-parser.c (folder_scan_drop_step): fix non-ANSI switch statement. * camel-service.c (service_getv): Likewise. * providers/imap/camel-imap-store.c (imap_getv): Likewise. * providers/pop3/camel-pop3-folder.c (cmd_builduid): Likewise. 2002-09-20 Not Zed <NotZed@Ximian.com> * tests/misc/split.c: New regression test for word splitting/search input parsing code. * tests/folder/test9.c (main): Fix for filter_driver api change. * camel-search-private.c (camel_search_words_split): Handle "'s and \'s to escape characters. For fat, lazy, slobs who dont like anything changing. 2002-09-19 Not Zed <NotZed@Ximian.com> * camel-filter-search.c (camel_filter_search_match): Take new session arg. Update callers. (symbols[]): changed get-source to header-source (header_source): Chagned from get_source. Now a boolean, returns true if source matches one of the arguments. Uses the provider url comparison directly for accurate checking. Makes 'filter on source' work. * camel-filter-driver.c (camel_filter_driver_new): Take new arg, session. (camel_filter_driver_finalise): Free session. 2002-09-19 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (parse_fetch_response): Don't increment the response pointer when deciding what type of header response we got, this screws up the string that we strdup into the part_spec string later. Also, instead of using the entire blob of HEADER.FIELDS that we get in the response, shorten it to just HEADER.FIELDS to use as the part_spec key. 2002-09-18 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (CAMEL_MESSAGE_INFO_HEADERS): Include MIME-VERSION and CONTENT-TYPE. 2002-09-18 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-command.c (camel_imap_command_response): If we get an ALERT, pass it to the user. An rfc2060 MUST, and bug #22496. 2002-09-17 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (parse_fetch_response): Handle when the response is from a HEADER.FIELDS (if so, we don't want to cache the result). (imap_update_summary): Request specific headers when building the CamelMessageInfo's rather than fetching the full headers. 2002-09-13 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part.c (process_header): If header_msgid_decode fails for the content-id field, just grab whatever is between the <>'s (yes, this means that the content-id is invalid - but what can ya do?). Addresses bug #21027. 2002-09-11 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_encode_string): Revert a patch that I had not meant to commit that was aimed at fixing bug #29234. 2002-09-10 Jeffrey Stedfast <fejj@ximian.com> * camel-store-summary.c (camel_store_summary_load): Same as below. Also save errno before closing the file so our caller can figure out why we failed. (camel_store_summary_save): Same here, but also flush the output stream to disk. * camel-folder-summary.c (camel_folder_summary_load): s/fclose(in) == -1/fclose(in) != 0/ since fclose doesn't necessarily return -1 on fail, all we really know is that it returns 0 on success and non-zero on fail. * camel-certdb.c (camel_certdb_save): fsync() the stream after we fflush it. * camel-folder-summary.c (perform_content_info_save): Do proper error checking and return -1 on fail. (camel_folder_summary_save): Check the return of perform_content_info_save and a few other output calls within the message_info_save loop. If any of them fail, save errno, close the file, and return -1. If we finish the loop without fail, fflush the stream and then fsync (fflush only flushes user-space buffers, you still need to fsync afterward to flush the data to disk). If either fail, treat it as an exception by saving errno, closing the stream, and returning -1. I suspect that this also fixes bug #30150 because the old code would fclose if fflush or fclose failed in the check after the loop (man fclose(3) states that any further calls using the stream (even another call to fclose) will have undefined behaviour no matter what the first fclose call returned). * providers/local/camel-local-summary.c (camel_local_summary_init): Don't malloc a private struct of 0 size. * providers/imap/camel-imap-folder.c (camel_imap_folder_fetch_data): Clear the exception even if we failed to get the message (part) from the imap-message-cache if we are online and able to try fetching it from the IMAP server. 2002-09-10 Not Zed <NotZed@Ximian.com> * camel-data-cache.c (data_cache_expire): Remove the hash table entry using the right path. Related to #29731. (camel_data_cache_remove): " 2002-09-09 Jeffrey Stedfast <fejj@ximian.com> Fixes bug #4224 * providers/imap/camel-imap-folder.c (camel_imap_folder_fetch_data): Pass ex into camel_imap_message_cache_get(). * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_get): Now takes an exception and sets it on fail. (camel_imap_message_cache_copy): Updated to properly handle cache_get error conditions. 2002-09-04 Dan Winship <danw@ximian.com> * camel-mime-utils.c (header_decode_date): Use e_mktime_utc. * camel-mime-message.c (camel_mime_message_set_date): Use e_localtime_with_offset. * broken-date-parser.c (decode_broken_date): Use e_mktime_utc. 2002-09-04 Not Zed <NotZed@Ximian.com> * tests/folder/test4.c (main): clear nonfatal stuff. * tests/folder/test6.c (main): check inbox and 'another folder' because some servers do different things w/ inbox. * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_get): Dont try to open the directory instead of an empty "" uid. * providers/imap/camel-imap-folder.c (imap_get_message): we dont want to g_return_if_fail, we need to set an exception and always handle the case. (get_message_simple): Set an exception if we get a construct failure. (imap_refresh_info): if we're refreshing inbox, force a reselect. this is required for at least cyrus. CHECK doesnt work either :( * tests/lib/folders.c (test_folder_basic): for non-local stores, the folder will have an extra ref for selection, take this into account when checking ref leaks. (test_folder_message_ops): Dont try to delete folder with messages in it, it works generally with imap. also, change params so we can test different mailbox types. (test_folder_message_ops): disconnect remote services before finishing off. doesn't need to stricly but makes ref count checking more accurate. (test_folder_message_ops): removed explicit remote sync, imap does it itself now ... * providers/imap/camel-imap-store.c: (camel_imap_store_finalize): call service_disconnect, so it isn't called later in the finalise chain, to properly cleanup on exit. * camel-mime-parser.c (folder_scan_content): If we fake the from eof file boundary, also make sure we say we matched nothing. Also make the end case a little more robust to make sure we expired all hope of finding a sub-boundary. * providers/imap/camel-imap-store.c, providers/imap/camel-imap-folder.c: Lots of changes, too numerous to list. Changed to use camel-imap-store-summary to cache list requests. Changed to use a canonicalised url path with / instead of per-store directory separator. Indirects folder name so invalid folder names can still be accessed. Summary now stored in a new expandable format in .ev-store-summary. 2002-08-29 Not Zed <NotZed@Ximian.com> * tests/folder/test3.c: Change the case sensitive search of subject to expect case insensitive results, as the behaviour has changed. (main): Made the search content before sync nonfatal. its something that needs to be fixed but not practical concern in evolution. * camel-block-file.c (block_file_validate_root): Only spit out the invalid root warnings if the file isn't empty. * camel-text-index.c (text_index_compress_nosync): Swap the path as well when we compress. * camel-mime-parser.c (folder_scan_content): Treat the end of file as a boundary if we're scanning From lines, and drop the last \n. (folder_scan_init_with_fd): Dont pre-read from the fd, and init eof. (folder_scan_init_with_stream): Similar. (folder_read): Handle eof, and set eof on 0 read. (folder_seek): Dont pre-read after a seek, and reset eof flag. (camel_mime_parser_init_with_fd): Fix doco, no pre-read occurs anymore. (camel_mime_parser_init_with_stream): Same. * providers/local/camel-mbox-summary.c (camel_mbox_summary_sync_mbox): Add a \n to end of content of each message, not at start. * providers/local/camel-mbox-folder.c (mbox_append_message): Instead of appending "\nFrom " to a mailbox, start with "From ", and append a \n after the message always. For better mutt/elm compatability. Also, unlock after we've stat'd. (mbox_append_message): Set the message's from_pos exactly as the mbox size. 2002-08-28 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-openssl.c (stream_setsockopt): Fixed `set' to hold the correct value when setting the flags. * camel-tcp-stream-raw.c (stream_setsockopt): Fixed `set' to hold the correct value when setting the flags. * camel-mime-filter-tohtml.c (is_protocol): Match file: urls too. 2002-08-28 Not Zed <NotZed@Ximian.com> * tests/folder/test8.c (test_add_message): Change message number to use hex, so its properly unique. (worker): Same here. * tests/misc/utf7.c: some utf7/utf8 tests. * providers/imap/camel-imap-utils.c (imap_mailbox_encode): Chagned to use camel_utf8_utf7 code. (imap_mailbox_decode): As above, using camel_utf8_utf7. 'UTF-7' isn't a widely support iconv() codeset, and besides the new code is simpler. * camel-utf8.[ch]: robust utilities for working with utf8 and utf7. 2002-08-27 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-thread.c (camel_folder_thread_messages_new): Now takes a third argument to specify whether or not to group messages by subject when threading. It seems some people don't want this feature (see bug #22791 for details). (thread_summary): If thread_by_subject is enabled, group un-threaded messages by subject otherwise don't bother. 2002-08-26 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-command.c (camel_imap_response_free_without_processing): If the response is NULL, just return. (now works the same way as imap_response_free which allowed a NULL response argument). Fixes bug #25491. 2002-08-26 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c: * providers/imap/camel-imap-folder.c: * providers/imap/camel-imap-command.c: Lock the CamelService->priv->connect_lock instead of using our own command_lock. This fixes bug #28177. 2002-08-23 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_helo): If we fail to send the EHLO/HELO command successfully, set transport->connected to FALSE when we close the tcp connection. Also updated to work on an IPv6 network. (smtp_connect): transport->authtypes can be NULL, so NULL-protect the call to g_hash_table_size() - this should fix a warning that was reported on the evolution@ximian.com mailing list. (smtp_set_exception): If the status message is multi-line, add a \n between lines. (connect_to_server): If we are going to ignore the EHLO/HELO error as if it were non-fatal, then we should clear the exception. 2002-08-22 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (imap_noop): If current_folder is not NULL and the summary for that folder is dirty, sync the flags rather than sending a NOOP. * providers/imap/camel-imap-utils.c (imap_parse_list_response): Workaround for Courier imap's brokeness when LSUBing the INBOX folder when it isn't subscribed to. Fixes bug #28929. 2002-08-21 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (connect_to_server): If imap_get_capability() fails, after unlocking the command_lock, disconnect the service. 2002-08-21 Jeffrey Stedfast <fejj@ximian.com> * camel-session.c (get_service): Don't register the noop timeout here anymore, we're gonna handle all this in the mailer now. (camel_session_register_timeout): Removed. (camel_session_remove_timeout): Removed. 2002-08-28 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (construct): Load store summary if it exists. (can_work_offline): Just see if we have any folders to say whether we can work offline or not. Should probably always just return true. 2002-08-27 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-store-summary.[ch]: New files to handle offline definition of folders, etc. * camel-url.h: Define CamelURL to be struct _CamelURL rather than anonymous struct. * camel-store-summary.[ch]: a few api tweaks. Also, the summary header is versioned separately at each level, so that version upgrades can be handled separately. Renamed FolderInfo -> StoreInfo to avoid namespace with current FolderInfo code. This should be reversed when the FolderInfo code is rationalised to this new base. 2002-08-23 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-command.c (camel_imap_command): domt encode folder name. * providers/imap/camel-imap-folder.c (do_copy): dont encode folder name. (do_append): dont encode folder name. * providers/imap/camel-imap-store.c (get_folder_status): don encode folder name in imap request. (get_folder_online): here too for creating folder. (rename_folder): Assume the incoming 'new name' is a utf8 path, whereas the 'old name' is as from get folder info (raw). (create_folder): Dont encode parent_name, assume its the raw thing. 2002-08-22 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (get_folder_online): Select based on unconverted name. (imap_build_folder_info): New function to create a folderinfo properly based on raw name. (subscribe_folder): Use above helper. (imap_folder_effectively_unsubscribed): Same here. (imap_forget_folder): Same here. (get_one_folder_offline): " 2002-08-21 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (parse_list_response_as_folder_info): Setup path properly, as decoded path with / separator. Setup full_name as non-decoded raw name. Keep url as decoded path but with server separator ... (ick). (create_folder): Dont call build_path anymore, get_folders() does it for us. (subscribe_folder): Build the path ourself. (imap_folder_effectively_unsubscribed): Same here. (get_subscribed_folders): list using %S not %F, we're using the raw server provided name directly. (subscribe_folder): As above, for SUBSCRIBE. (unsubscribe_folder): Same here. (delete_folder): Same. (rename_folder_info): Same here for source name. (rename_folder): And here? (get_folders_online): Amd here. * providers/imap/camel-imap-utils.c: (imap_parse_list_response): Dont decode the mailbox. * camel-utf8.[ch]: some new utf8 & utf7 utilities. * providers/imap/camel-imap-utils.c (imap_mailbox_encode): (imap_mailbox_decode): use camel_utf7/8* functions instead. : Add config.h and alloca.h headers. 2002-08-20 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (imap_noop): Override the default noop implementation. * camel-store.c (camel_store_noop): New virtual method to ping a store. * camel-session.c (get_service): Register a timeout that calls camel_store_noop() every 10 minutes. 2002-08-19 Peter Williams <peterw@ximian.com> Plug some large leaks in the indexing code. * camel-mime-filter-index.c (camel_mime_filter_index_set_index): Unref the old index if necessary. * camel-text-index.c (camel_text_index_get_type): Give it the correct parent type so that finalize handlers chain. (camel_text_index_name_get_type): Here too. (camel_text_index_cursor_get_type): Here too. (camel_text_index_key_cursor_get_type): Here too. 2002-08-20 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (get_folders): Don't let the minimum recursive depth be negative. (get_folders): Pass flags & CAMEL_FOLDER_INFO_SUBSCRIBED as the `lsub' argument to get_folders_online() otherwise we'll end up doing a LIST when we really wanted to do an LSUB. * camel-session.c: Set CAMEL_URL_NEED_PATH for the vFolder provider. 2002-08-19 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c: Changed the STOREINFO_VERSION - should fix bug #28571, but if it doesn't - oh well, I don't care. I'm considering it fixed no matter what. * camel-gpg-context.c (gpg_sign): If the diagnostics are empty, just explain that gpg failed to execute. (gpg_encrypt): Same. (gpg_decrypt): And here too. * tests/lib/camel-test.c (camel_test_init): Updated to pass the needed arguments to camel_init() and to init gthreads before camel (since camel expectes it to be initialised already). * tests/lib/session.c: #include <config.h> needed for ENABLE_THREADS so that sizeof (CamelSessionClass) gives the correct size. * tests/*: Updated to work with camel API changes. 2002-08-16 Dan Winship <danw@ximian.com> * camel-session.c (get_service): If the passed-in URL contains a path, but the service doesn't allow paths, then the path part must be a folder name, and we should ignore it here. #29010 2002-08-16 Chris Toshok <toshok@ximian.com> * camel-arg.h: remove the outer parens from the expansion of camel_argv_{start,end} since at least on freebsd va_end is #defined to be nothing, so (va_end(..)) causes a syntax error. 2002-08-16 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (get_folders): Fix the logic check for working out if we add inbox or not, we do it if we listed the namespace. 2002-08-15 Not Zed <NotZed@Ximian.com> * camel-store.h (CamelFolderInfo): added a flags field, defined some flags, currently only used by imap. This and below is first step to getting incremental folder tree loading. * providers/imap/camel-imap-store.c (get_subscribed_folders): If the subscribed list is empty, always add inbox. (imap_connect_offline): If the 'storeinfo' namespace isn't the same as our own (if its set), then ignore it, assume we changed settings. Otherwise you can't change the namespace ... (get_folders): allow the env variable CAMEL_IMAP_MAX_DEPTH to override the default maximum recursion depth if 10 levels. (camel_imap_store_readline): Dont depend the log debug on d(x) being defined. (get_folder_info_online): changed to use new functions/names. 2002-08-14 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (get_subscribed_folders): Removed the by_hand sillyness. Return an array instead. (get_folders): New method to get folders recursively without having to use '*'. Uses '%', and stops if it gets too deep (current max of 10). (get_folder_counts): New method to fill out unread counts on folderinfo tree. (get_subscribed_folders): Fix some failure logic. * providers/imap/camel-imap-utils.[ch]: Changed IMAP_LIST_FLAGS to use the new CamelFolderInfoFlags directly. 2002-08-09 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (get_temp_uid): New function, generate a UID that has better chances of being unique. (imap_append_offline): Use it here. (imap_transfer_offline): And here. 2002-08-09 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-store.c (get_folder_info_online): Tweak how we build the tree based on whether a full tree or a subtree is being requested. Fixes subscribe dialog on UW servers. (get_one_folder_offline): Don't add folders that we're not subscribed to; fixes UW folders that weren't being marked as NoSelect because the storeinfo doesn't record that information. 2002-08-09 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (imap_rename): When renaming our cache, add the "/folders" to the path. 2002-08-08 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_init): Set some default permanent_flags. This is the other cause of bug #28038. (imap_refresh_info): After calling camel_imap_command() with a NULL format value (this is a shortcut to change the current_folder by SELECTing it), call camel_imap_folder_selected(). This may have been why permanent_flags never got set on the folder even after going online in bug #28038. (imap_update_summary): Same here. Also NULL-guard camel_imap_response_free(). 2002-08-07 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_encode_param): Oops - outbuf pointed to alloca'd memory but we were g_free'ing it after using it. Instead use g_malloc for this outbuf buffer since it may be kinda large. Also don't depend on a single byte to nul-terminate the outbuf buffer so as to be safe with charsets such as UCS2 and UCS4, instead keep a pointer to the end of the buffer. * providers/imap/camel-imap-folder.c (parse_fetch_response): Only add the stream to the gdatalist if it is non-NULL. 2002-08-06 Jeffrey Stedfast <fejj@ximian.com> * providers/sendmail/camel-sendmail-transport.c (sendmail_send_to): If writing to the sendmail pipe fails, wait for the sendmail process to exit before returning. Fixes bug #19636. 2002-08-06 Not Zed <NotZed@Ximian.com> * providers/pop3/camel-pop3-folder.c (cmd_list): Add messageinfo to id table, and store array index in messageinfo too. (fi_to_index): Removed, use index member of messageinfo, fixed callers. (cmd_builduid): Parse the headers of the data, and ignore any status or x-status headers. (uid_to_fi): Removed, use the hashtable instead. (id_to_fi): Same. (cmd_uidl): Add messageinfo to uid table. (pop3_refresh_info): Setup hash tables, and fill out uid table for nonuid capable servers. (pop3_finalize): Destroy hash tables. Only free resources if the uids table is not null (if finalised before we ran refresh info). (id_to_fi): Removed. Use id hash table instead, fixed caller. 2002-08-05 Jeffrey Stedfast <fejj@ximian.com> * camel.c (camel_init): Always set a g_atexit handler to flush/save the camel-certdb. * providers/imap/camel-imap-store.c (get_folder_info_online): If we are asking for the toplevel folders, use the IMAP store's namespace as the toplevel folder rather than an empty string. Fixes bug #28574. (imap_forget_folder): Rearrange the code so that name can never be used uninitialised. * providers/imap/camel-imap-utils.c (imap_namespace_concat): Removed. 2002-08-05 Dan Winship <danw@ximian.com> * camel-folder-summary.h (CamelMessageFlags): Add CAMEL_MESSAGE_ANSWERED_ALL, for folders that can distinguish a reply-to-sender from a reply-to-all. (eg, Exchange) * providers/local/camel-local-folder.c (local_init): Add ANSWERED_ALL to permanent_flags too. 2002-08-02 Jeffrey Stedfast <fejj@ximian.com> Fixes bug #26237. * providers/imap/camel-imap-command.c (imap_command_strdup_vprintf): Don't prepend any namespace shit in the %F case. * providers/imap/camel-imap-utils.c (imap_parse_list_response): Don't strip off the namespace. 2002-08-01 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-store.c (create_folder): If the folder name contains the directory separator, reject it. 2002-08-02 Not Zed <NotZed@Ximian.com> * providers/local/camel-mbox-summary.c (mbox_summary_check): Clear the summary after building the 'removed uid' list. * camel-folder-summary.c (message_info_load): revert jeff's patch below, and do it slightly differently, and stop the auto-reformatting bullshit. 2002-08-02 Not Zed <NotZed@Ximian.com> * providers/local/camel-maildir-summary.c (message_info_new): If we get a duplicate, just re-use the old info. (remove_summary): Add any removed to a change list. (maildir_summary_check): If we find new ones in 'cur' track them in the change list. Removed FIXME about it. For #18348. 2002-08-01 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.c (message_info_load): Do more error checking to prevent crashing if we fail to read a string for example. 2002-07-29 Peter Williams <peterw@ximian.com> Fix bug #28238 * providers/imap/camel-imap-store.c (imap_forget_folder): New function, breaking out the code in delete_folder() to handle when a folder is removed. (imap_folder_effectively_unsubscribed): New function, breaking out the code in unsubscribe_folder() to handle when a folder is unsubscribed. (imap_check_folder_still_extant): New function, check whether a folder exists by LIST'ing it. (imap_store_refresh_folders): Add code here to check if the folder still exists if we're unable to refresh its info. If so, pretend that it was unsubscribed (to get the mailer to remove it from the tree) and delete its cache. If somehow this goofs up, the worst that can happen is that we must redownload the headers. (get_folder_status): If we can't get the status, behave as above. (delete_folder): Call imap_forget_folder() where the bulk of this code has gone. (unsubscribe_folder): Call imap_folder_effectively_unsubscribed() where the bulk of this code has gone. 2002-07-31 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (parse_fetch_response): Disable my fix to not cache headers as it means that each time we load the message, we've got to re-fecth the headers. * camel-mime-filter-tohtml.c (is_addr_char): Don't use "isprint(c)" to mean "c >= 32 && c < 128" since it doesn't in most locales. (is_url_char): Same. (is_trailing_garbage): Same. (is_domain_name_char): New macro for dns-valid characters (email_address_extract): Use is_domain_name_char rather than is_addr_char for the part after the @. 2002-07-30 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream.c (camel_tcp_address_new): Update the comment. * camel-tcp-stream-raw.c (socket_connect): If building with IPv6 support and the address is an IPv6 address, connect using a sockaddr_in6 otherwise use the standard IPv4 sockaddr_in structure. (stream_get_local_address): Fix to work with IPv6 addresses. (stream_get_remote_address): Same. * camel-tcp-stream-openssl.c (socket_connect): Same as above. (stream_get_local_address): Fix to work with IPv6 addresses. (stream_get_remote_address): Same. * camel-tcp-stream-ssl.c (stream_connect): If building with IPv6 support and the address is an IPv6 address, initialise the PRNetAddr accordingly. (stream_get_local_address): Fix to work with IPv6 addresses. (stream_get_remote_address): Same. 2002-07-30 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-openssl.c (ssl_verify): X509_digest() needs a pointer to an int len. Also fixed some other compiler errors. * camel-certdb.h: #include <stdio.h> 2002-07-30 Jeffrey Stedfast <fejj@ximian.com> * camel-certdb.c: New source file implementing a very basic certificate database. This is mostly just here because the Mozilla NSS certdb seems to not be working for everyone's Evolution install (works fine for me and Ettore but not many other people). * camel-tcp-stream-ssl.c (ssl_bad_cert): If we have this certificate in our own CamelCertDB, then get the trust value from that and only prompt the user if the trust is unknown. * camel-tcp-stream-openssl.c (ssl_verify): Same. * camel.c (camel_init): Create our default certdb. 2002-07-30 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (imap_transfer_offline): Use the same command-counter mojo that the append code uses in case the use tries to copy the same message to the folder repeatedly. 2002-07-29 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (unset_flag): New filter action to unset a system flag. The exact opposite of set_flag. 2002-07-26 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-local-store.c (get_folder): If the path exists and is a directory, return non-NULL. Missed this before which is how we could return NULL even on success. Oops. 2002-07-26 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-summary.c (camel_imap_summary_add_offline_uncached): The flags'n'tags copy was unnecessary. Removed it. 2002-07-25 Jeffrey Stedfast <fejj@ximian.com> If we're gonna do an elite hack, lets do it right. * providers/local/camel-local-store.c (get_folder): On success, return non-NULL so that our caller doesn't have to depend on 'ex' being non-NULL in order to work correctly. * providers/local/camel-mbox-store.c (get_folder): Don't bother checking if an exception was set, just use the return value of camel-local-store's get_folder. * providers/local/camel-maildir-store.c (get_folder): Same as the mbox change. * providers/local/camel-mh-store.c (get_folder): Here too. 2002-07-25 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-summary.h (camel_imap_summary_add_offline_uncached): Prototype. * providers/imap/camel-imap-summary.c (camel_imap_summary_add_offline_uncached): New function to add an uncached message to the summary while offline. * providers/imap/camel-imap-store.c (imap_connect_online): Only refresh the folders if the disco diary is empty (ie, we're not resyncing.) * providers/imap/camel-imap-folder.c (imap_transfer_offline): Emit an event for the source if deleting its originals. Also, if we can't load the message, still process it, using the new imap summary function. 2002-07-25 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c: Remove "possibly non-fatal" exception strings, all exceptions are fatal unless their caller decides otherwise. (connect_to_server): If smtp_helo() fails due to us getting disconnected, treat the exception as fatal. (smtp_connect): Abort if smtp_helo() fails and we got disconnected. (smtp_send_to): If smtp_mail() fails, abort the send operation. (smtp_set_exception): If respbuf was NULL, then we got disconnected and so update our state accordingly. (smtp_disconnect): Only send a QUIT if we are still in the connected state (or, at least think we are). 2002-07-24 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (imap_expunge_uids_resyncing): It would help to mark the UIDs as deleted before we try to expunge them (in the case when no other UIDs were marked deleted this didn't happen.) Also fix an FMR. (imap_expunge_uids_offline): Trigger a folder_changed event with our own changeinfo because camel_imap_response_free won't be able to do that for us. 2002-07-25 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (do_shell): Renamed from shell_exec(). Also, we now want to pass argc/argv on to the CamelFilterDriverShellFunc. (camel_filter_driver_set_shell_func): Renamed a bit. * camel-filter-search.c (pipe_message): Renamed from shell_exec(). 2002-07-25 Not Zed <NotZed@Ximian.com> * camel-folder-summary.c (camel_folder_summary_save): When writing the summary, use TRUNC flag, duh. Also, write to a temp file first, and rename when closed successfully, and check ferror() and fclose() against 0 rather than -1. * providers/local/camel-mbox-summary.c (summary_update): Decrement i if we remove the summary item so we dont skip every 2nd one. * camel-mime-utils.c (header_decode_mailbox): Use rfc2047_decode_word explicitly incase we just found an encoded word. Stops us re-decoding the string twice, which fixes memory corruption in #26330 when the HUGE string is used later. 2002-07-24 Not Zed <NotZed@Ximian.com> * camel-partition-table.c (camel_key_table_next): Didn't unlock if we exited on an empty key list. * providers/imap/camel-imap-folder.c (imap_expunge_uids_online): Sync before doing an expunge if we dont have uidplus. See #25766. (imap_expunge_uids_resyncing): Same here. * providers/imap/camel-imap-store.c (get_folder_info_online): Duplicated the very much hack in imap_store_refresh_folders() to avoid a deadlock in #27959. A decent fix has to wait for a rewrite. (get_folder_online): Move folders to a subdirectory of storage_path, rather than in it. Bye bye everyones settings. See 24947. (get_folder_offline): And here too. (delete_folder): And here too. (rename_folder): and here too. (get_folder_info_offline): And here too. I love code reuse. 2002-07-23 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream.c: #include <string.h> for memcpy. 2002-07-22 Not Zed <NotZed@Ximian.com> * camel-vee-folder.c (vee_set_message_user_tag): Oops, forgot to implement this. Fixes #24604. 2002-07-21 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-crlf.c (filter): Only add a \r if the character before the \n wasn't already a \r - this covers for when people add DOS files as attachments. Fixes bug #28056. 2002-07-19 Not Zed <NotZed@Ximian.com> * camel-mime-part-utils.c (convert_buffer): Changed to use a different length calculator. 2002-07-17 Jeffrey Stedfast <fejj@ximian.com> * camel-lock.c: Add the appropriate #include for the _() macro. 2002-07-17 Not Zed <NotZed@Ximian.com> * providers/pop3/camel-pop3-store.c (pop3_try_authenticate): Got rid of the stupid "tryagain" semantics, which didn't work. (pop3_connect): Fixed here to only retry under proper conditions. 2002-07-16 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-canon.c (filter): When CRLF encoding, "empty" (after stripping trailing whitespace characters) lines need to end in CRLF also, not just lines with data in them. 2002-07-16 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (imap_sync_online): Clean up some improper usage of CamelExceptions (looking at camel_exception_is_set when ex = NULL). * providers/imap/camel-imap-store.c (get_folder_online): Here too. * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_new): Nothing sets the exception so don't bother to check it at the end of this function. Prevents a warning about checking a NULL exception's value. 2002-07-16 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_helo): Make the standard AUTH format take priority over the AUTH= priority, since sometimes servers only list a subset of the supported authtypes in the AUTH= response while they list all authtypes in the standard AUTH response. Fixes "bug" #27841. 2002-07-15 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (imap_get_message): If we're not online and the message isn't in our cache, give up. 2002-07-15 Not Zed <NotZed@Ximian.com> * camel-filter-search.c (get_label): Removed. (get_score): Removed. Labels & scores are stored in tags, we already have a function to get tags. (user_tag): Fail if we dont get valid arguments, it must be a string type. (system_flag): Same here. 2002-07-15 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_msgid_generate): Try to get the FQDN from the results of gethostname(). If that fails, then fall back to the results gotten from gethostname() or if that fails just use "localhost.localdomain". Addresses bug #17416. (header_decode_param): Protect against a NULL value. * providers/smtp/camel-smtp-transport.c (smtp_data): Get rid of the constant 'required' variable, just use the value when calling set_best_encoding. * providers/local/camel-local-provider.c: Changed the mbox and spool provider descriptions to try and be more clear. I'm not sure I like the mention of Evolution in the mbox provider description, but there was no other way to make it clear that it MOVED mail out of the mbox folder into a folder managed by Evolution without mentioning Evolution. *sigh* Anyways, to me "retrieve" means move, but apparently this isn't clear to the weak minded. 2002-07-15 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (imap_rescan): >= is not appropriate here because the sequences are 1-based; so seq = summary_len is valid. Hopefully addresses 15646, because this caused the last message in a folder to be reloaded, replacing its CamelMessageInfo, and the tree view was using the old CamelMessageInfo. 2002-07-13 Not Zed <NotZed@Ximian.com> * camel-folder-summary.c (camel_folder_summary_info_new_from_message): Only assign a uid if indexing is enabled. This stops us always assigning a uid in the imap folder and disco folder? * providers/imap/camel-imap-folder.c (imap_update_summary): Also check the uid is set at all, another bit of a fix for #15667. 2002-07-09 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-folder.c (imap_update_summary): Check for existing messages of the same uid before doing anything. If it exists, do nothing (perhaps it should merge?). A dirty hack for #15667. 2002-07-12 Jeffrey Stedfast <fejj@ximian.com> * camel-multipart-encrypted.c (camel_multipart_encrypted_decrypt): On error, do not unref the parts gotten using camel_multipart_get_part() since that function does not ref the parts. Also don't forget to reset the decrypted stream before trying to parse it ;-) 2002-07-12 Jeffrey Stedfast <fejj@ximian.com> * camel-multipart.c (camel_multipart_class_init): Updated to not use camel_type_get_global_classfuncs() since this is just a cast macro now. * providers/imap/camel-imap-folder.c (get_content): Handle multipart/encrypted parts too. * camel-mime-part-utils.c (camel_mime_part_construct_content_from_parser): Handle multipart/encrypted types too. 2002-07-11 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_data): Don't let any parts have a binary encoding. 2002-07-11 Jeffrey Stedfast <fejj@ximian.com> Fixes bug #27672 * camel-mime-filter-bestenc.c: Conditionally #include <config.h> * camel-mime-filter-linewrap.c: Same here... although we could probably just get rid of this filter? We don't seem to use it anywhere since we try to QP/Base64 encode any text parts with long lines. Besides, we couldn't use this filter for SMTP anyway since we can't risk possibly linewrapping a binary mime part. I dunno, maybe this could be useful in the composer though? *shrug* * providers/smtp/camel-smtp-transport.c (smtp_data): Always call camel_mime_message_set_best_encoding() even if the server allows 8BITMIME and even if we don't have any 8bit parts because we may have parts with long lines (>998 octets) which also need to be encoded. * camel-mime-message.c (check_8bit): Don't forget to check for the binary encoding here as well. 2002-07-11 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (parse_fetch_response): If we are fetching HEADERs, don't cache the stream - instead just create a mem-stream and return that. This reduces a significant amount of overhead due to disk i/o. * camel-multipart-encrypted.c (camel_multipart_encrypted_decrypt): Don't forget to set an appropriate exception before returning NULL. 2002-07-10 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (get_message_cb): Make sure message is non-NULL before setting the source on it. * camel-gpg-context.c: Renamed stdin to stdin_fd, stdout to stdout_fd, and stderr to stderr_fd in struct _GpgCtx since Solaris' stdin/stdout/etderr variables seem to be macros and are thus breaking the compilation. 2002-07-04 Jeffrey Stedfast <fejj@ximian.com> * providers/nntp/camel-nntp-provider.c: Set the IS_SOURCE provider flag. * providers/pop3/camel-pop3-store.c: Change all exceptions to report which POP server failed (and tried to make a few of the exceptions a bit more clear). 2002-07-03 Not Zed <NotZed@Ximian.com> * camel-vee-folder.c (vee_sync): If we get an exception from the child sync, update the exception string so we know where it really was. * camel-vee-store.c (change_folder): Util to emit created/deleted events. (vee_get_folder): If the folder has path elements, and its parents dont exist, create dummy no-select nodes. This is for bug #4246. (vee_delete_folder): Use change_folder() to save code. * providers/local/camel-mbox-folder.c (mbox_set_message_flags): Only set xevchange if we're storing status headers in pine format, so we dont do a full sync every time any flag changes. 2002-06-27 Not Zed <NotZed@Ximian.com> * camel-object.c (camel_object_hook_event): Take a void object arg. (camel_object_remove_event): " (camel_object_unhook_event): " (camel_object_trigger_event): ", also, execute events in the same order they were added, i.e. reverse hook-list order. 2002-06-28 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (gpg_ctx_get_argv): Don't use --no-auto-key-retrieve, this option has apparently been deprecated in gpg 1.0.7. 2002-06-27 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (gpg_ctx_op_step): Loop on our reads while errno is EINTR or EAGAIN. Also make sure to completely read stdout and stderr by keeping better state. (gpg_ctx_parse_status): In the case of a NODATA response from gpg, try to use the diagnostics that gpg may have written to its stderr. (gpg_verify): Check that the gpg process has not exited. (gpg_ctx_op_wait): Make sure we haven't already exited (as by gpg_ctx_op_is_exited()) and if we have, retrieve the exit status from the gpg context. 2002-06-26 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.[c,h]: Removed. * camel-pgp-mime.c (pgp_mime_part_sign_restore_part): Removed. (pgp_mime_part_sign_prepare_part): Removed. (camel_pgp_mime_part_sign): Removed. (camel_pgp_mime_part_verify): Removed. (camel_pgp_mime_part_encrypt): Removed. (camel_pgp_mime_part_decrypt): Removed. (camel_pgp_mime_is_rfc2015_signed): Documented. (camel_pgp_mime_is_rfc2015_encrypted): Same. 2002-06-26 Jeffrey Stedfast <fejj@ximian.com> * camel-multipart-encrypted.c (camel_multipart_encrypted_encrypt): Oops, this needs to take a userid argument. * camel-gpg-context.c (gpg_clearsign): Removed (never did anything anyway). * camel-pgp-context.c (pgp_clearsign): Removed. * camel-cipher-context.c (camel_cipher_clearsign): Removed unused method. * camel-multipart-encrypted.[c,h]: New class implementing the multipart/encrypted content type. Contains methods for encrypting and decrypting a multipart/encrypted MIME object. * camel-gpg-context.c (gpg_ctx_parse_status): Check for NODATA too. 2002-06-25 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-local-provider.c: Remove the #include for camel-spoold-store.h 2002-06-26 Not Zed <NotZed@Ximian.com> * camel-gpg-context.c: Include <sys/time.h> for struct timeval. * providers/local/camel-local-provider.c (camel_provider_module_init): Removed spoold provider. The spool provider does it now. (): Spoold doesn't exist anymore, remove its header. 2002-06-25 Not Zed <NotZed@Ximian.com> * providers/local/camel-spool-folder.c (camel_spool_folder_new): Support a new xstatus option - folders update/honour the Status/X-Status headers in addition to X-Evolution. * providers/local/camel-local-summary.c (camel_local_summary_write_headers): If supplied with an additional status or xstatus arg, write a Status header and/or X-Status. Also fix the case of properly terminating the headers if an xev line isn't supplied. * providers/local/Makefile.am (libcamellocalinclude_HEADERS,SOURCES): Removed spoold-store.[ch]. * providers/local/camel-local-provider.c (camel_provider_module_init): For the spoold type, just use the spool store instead. * providers/local/camel-spool-store.h: Added a type field, so the 1 store can implement different types without having to subclass. * providers/local/camel-spool-store.c (camel_spool_store_get_toplevel_dir): Removed, inherits from local store now. (construct): If we're pointing to a file, treat it as mbox mode, otherwise treat it as 'elm' mode. (get_folder): Only test for INBOX in mbox mode. (get_folder_info_elm): (get_folder_info_mbox): Two alternatives for getting folder info, depending on the type of folder we're looking at. (get_folder_info_mbox): Make the url include the protocol. (scan_dir): " * providers/local/camel-spoold-store.c (camel_spoold_store_get_toplevel_dir): Removed, inherits from local store now. * camel-folder.c (get_message_user_tag): Dont use a g_return_if_fail for info==NULL. This is not an error. (set_message_user_tag): And same here. (set_message_user_flag): Sigh, and here. (get_message_user_flag): And here. (set_message_flags): and here ... (get_message_flags): Dum de dum, de done at last. * providers/local/camel-mbox-folder.c (mbox_get_message): Check for new messages whenever we retrieve one. In the common no-update case, this is a single stat. (mbox_get_message): If we need to rescan, then force a full rescan to make sure it does the right thing. (mbox_get_message): Cleanup the exception handling a bit, if we do get an error, propagate any folder changes anyway as well. (mbox_set_message_user_flag): Argh more of these stupid g_returns taht shouldn't be. (mbox_set_message_user_tag): Here too. (mbox_set_message_flags): If the read flag is being changed, mark it as an xevchange (i.e. Status line change). * providers/local/camel-mbox-summary.c (summary_rebuild): Merged into summary_update. (summary_update): Changed to allow it to update existing lists of messages without clearing out the summary. (mbox_summary_check): Dont clear the summary, just re-scan. (message_info_new): Attempt to support the 'Status: RO' elm/pine thing. (camel_mbox_summary_encode_status): (camel_mbox_summary_decode_status): Util functions for creating/parsing the Status line. (camel_mbox_summary_sync_mbox): Write out the status line if we're going to try support it. (camel_mbox_summary_xstatus): Implement option to control read/write of (x-)status. (message_info_new): Do x-status stuff based on run-time option. (camel_mbox_summary_sync_mbox): " (mbox_summary_add): If x-status enabled, then always add status/x-status headers to message. * camel-folder-summary.c (summary_assign_uid): If the messageinfo is already in the summary, AND is the same messageinfo, dont do anything, return a value to indicate this. (camel_folder_summary_add): Do nothing if this info already in the summary, so we can perform updates. 2002-06-24 Not Zed <NotZed@Ximian.com> * providers/local/camel-local-summary.c (camel_local_summary_check_force): New method to force the next summary check to be a full check, set if a mismatch occurs. * camel-folder-summary.c (camel_folder_summary_load): If we have no summary path set, dont do any i/o, rather than abort. (camel_folder_summary_save): " (camel_folder_summary_header_load): " * providers/local/camel-spool-store.h: Inherit from camel mbox store, even if we override almost everything. * providers/local/camel-local-folder.c (camel_local_folder_construct): If the base path points to a file, use that as the folder path as well. * providers/local/camel-spool-folder.h: Inherit from camel-mbox-folder. * providers/local/camel-spool-summary.c (spool_summary_sync_full): Use camel_mbox_summary_sync_mbox to do most of the work. * providers/local/camel-spool-summary.[ch]: Make spool-summary inherit from mbox summary rather than foldersummary. * providers/local/camel-mbox-summary.c (mbox_summary_sync): Make sync_full/quick virtual methods. (camel_mbox_summary_sync_mbox): The full sync method put into a simple function that sync's from fd to fd. (mbox_summary_sync_full): Use summary_sync_mbox to do the real work. (mbox_summary_check): Create removed events if the folder gets cleared. Also, dont clear the summary before a rebuild, try to merge. 2002-06-25 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c: #include <sys/time.h> 2002-06-25 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (gpg_hash_str): Return the option as --digest-algo=<micalg> (gpg_ctx_parse_status): Decode more status information and keep track of trust and such. (gpg_ctx_op_complete): Make sure we aren't waiting for gpg to write data to stdout (such as encrypted/decrypted data or a signature). (gpg_encrypt): Remove the fejj@stampede.org hard-coded value (which had only meant to be for testing). 2002-06-24 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (gpg_ctx_get_argv): Use --command-fd rather than --passphrase-fd since what we really want is an interactive mode (besides, --passphrase-fd requires a passphrase to be sent before --status-fd will send a NEED_PASSPHRASE status message). (gpg_ctx_parse_status): When we encounter a BAD_PASSPHRASE, tell the session to uncache it. (camel_gpg_context_set_always_trust): Implemented. 2002-06-23 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (gpg_ctx_parse_status): Check for ERRSIG also. (gpg_ctx_op_step): Don't try writing a passphrase unless our operation requires it. 2002-06-22 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.c (gpg_encrypt): Return -1 on error. (gpg_decrypt): Same. (gpg_sign): Again here. (gpg_ctx_op_step): Only attenmpt to read from the status-fd if we are not already 'complete'. * camel-pgp-mime.c (camel_pgp_mime_part_encrypt): Flush the filtered stream. 2002-06-21 Jeffrey Stedfast <fejj@ximian.com> * camel-gpg-context.[c,h]: New source files implementing the CamelCipherContext class for gnupg. * camel-pgp-context.c (camel_pgp_context_new): Return a CamelCipherContext. * camel-pgp-mime.c (camel_pgp_mime_part_decrypt): Take a CamelCipherContext argument rather than a CamelPgpContext since we now have a CamelGpgContext also. (camel_pgp_mime_part_encrypt): Same. (camel_pgp_mime_part_verify): Same. (camel_pgp_mime_part_sign): Same. 2002-06-20 Jeffrey Stedfast <fejj@ximian.com> * camel-digest-folder.c: Updated to support searching as well as making it use CamelDigestSummary. * camel-digest-summary.[c,h]: New class to handle the summary for CamelDigestFolder. * camel-tcp-stream-ssl.c (set_errno): PR_IO_TIMEOUT_ERROR should map to ETIMEDOUT and not EAGAIN. (stream_connect): Reset the PR_Poll() timeout back to 2 minutes as this wasn't the problem afterall. 2002-06-19 Not Zed <NotZed@Ximian.com> * camel-mime-parser.c (folder_scan_drop_step): Drop back to initial state if we came from eof state. 2002-06-19 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (stream_connect): Set the poll timeout to be PR_INTERVAL_MIN, this one shouldn't need to be 2 minutes (plus it blocks my connection at home for far too long). * providers/imap/camel-imap-folder.c (imap_refresh_info): Make sure we hold the command_lock before accessing current_folder. (imap_append_online): Same. (imap_transfer_online): Same. * providers/imap/camel-imap-store.c (delete_folder): We need to hold the command_lock before we can set the current_folder to NULL. (rename_folder): Same. (get_folder_info_online): We need to make sure we hold the command_lock in order to use current_folder. (everywhere): Same. 2002-06-19 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_connect): Don't bother trying to authenticate (and thus don't throw up an error dialog) unless the server is an ESMTP server and supports authentication. * camel-sasl-plain.c: Change the user-friendly name to PLAIN, this is what it should be. Not "Password" because user's immediately think that they need authentication when they see "Password" even if they don't need it at all. 2002-06-17 Jeffrey Stedfast <fejj@ximian.com> * camel-vee-folder.c (camel_vee_folder_add_folder): Update the freeze state of the new source folder. (camel_vee_folder_remove_folder): Undo any freeze state that we have imposed on this source folder. 2002-06-14 Jeffrey Stedfast <fejj@ximian.com> * camel-vee-folder.c (vee_freeze): Freeze source folders. (vee_thaw): Thaw all source folders. Fixes bug #24317. * providers/imap/camel-imap-folder.c (imap_sync_online): Unlock the command_lock after each UID STORE command so that other (probably more important) threads can send their requests. * camel-arg.c: * camel-disco-diary.c: * camel-index-control.c: * camel-mime-part.c: * camel-pgp-mime.c: * camel-store.c: * camel-tcp-stream-ssl.c: * camel-text-index.c: Fixed compiler warnings. * camel-multipart-signed.c: Updated to use ctx->sign_protocol rather than ctx->protocol. * camel-cipher-context.h: Add an ecnryption protocol member. * camel-pgp-context.c (camel_pgp_context_init): Set the encryption protocol. 2002-06-11 Not Zed <NotZed@Ximian.com> * camel-vtrash-folder.c (camel_vtrash_folder_class_init): Our parent class is camel_vee_folder, not camel_folder, fix setting of parent folder (unused anyway, but removes a run-time warning). 2002-06-07 Not Zed <NotZed@Ximian.com> * camel-pgp-context.c (camel_pgp_context_init): Duh! The protocol is application/pgp-signature. TODO: need to have a signed and an encrypted protocol specified separately. 2002-06-06 Not Zed <NotZed@Ximian.com> * camel-folder-thread.c (thread_summary): Do the work here. (camel_folder_thread_messages_remove): Implement. Remove some uid's from the tree. Disalbe from build, for now. (remove_uid_node_rec): Actually do the work. (camel_folder_thread_messages_add): Implement. Add some summary items to the tree. Disable from build though. (camel_folder_thread_messages_apply): Make a thread tree built using camel_folder_thread_new() map to a new set of uids, preserving some order. Meant primarily to manage deleted + added uids in a user-friendly manner. 2002-06-09 Jeffrey Stedfast <fejj@ximian.com> * camel-multipart-signed.c (camel_multipart_signed_verify): Use CAMEL_EXCEPTION_SYSTEM rather than '1' for clarity. 2002-06-07 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (stream_connect): Up the timeout on the connect code to 2 minutes. 2002-06-05 Dan Winship <danw@ximian.com> * camel-mime-utils.c (header_decode_mailbox): Parse "From: Dan Winship <danw>" (with no domain) in the way the sender meant it, even though it's completely wrong. 2002-06-04 Not Zed <NotZed@Ximian.com> * providers/local/camel-local-provider.c: Remove the CONF_DEFAULT_PATH entry, as this is handled by the url config stuff. This made it basically impossible to configure any local folder from the gui. * providers/local/camel-local-folder.c (local_refresh_info): Force a refresh. Should work for all local folders to force them to refresh. * providers/local/camel-maildir-folder.c (maildir_refresh_info): Removed, moved into camel-local-folder. * providers/local/camel-mh-summary.c (mh_summary_next_uid_string): Once we assign or get a uid, tell the summary of this, so the next uid we get will be higher than any previously. * camel-object.c (camel_object_ref, camel_object_unref): I got sick of casting, these now take void * like they should, and perform their own run-time type checking. * providers/local/camel-mh-store.c (recursive_scan): Changed to use stat, not lstat (*shrug* maybe someone wants to softlink their maildir tree?). (inode_hash): (inode_equal): (inode_free): Copied from camel-maildir store (should put into camel-local-store or utils?). (recursive_scan): Changed to check for re-visiting inodes. Also, it builds the tree itself, rather than using folder_info_build. (add_folder): Changed to folder_info_new. (recursive_scan): Properly honour the recursive flag. Also, lookup unread count from folder. (folder_info_new): Init unread message count to -1, since we dont know yet. (folder_info_new): Take the name as an argument, and perform the merging here. (folders_update): Util func to add/remove folders from .folders file. I'm assuming its sorted. (get_folder): Add the folder to .folders if we created a new one, and if it exists. (delete_folder): Remove from .folders, etc. (folders_scan): If we have a .folders file, read and use that instead. (recursive_scan): Handle scanning from a particular directory properly. (rename_folder): Implement so we can track any changes to the .folders file if its turned on. ** Applied patch below from Greg Hudson. 2002-05-10 Greg Hudson <ghudson@mit.edu> * camel-mh-store.c (get_inbox, get_folder_info, recursive_scan, add_folder): Implement support for MH stores. * camel-mh-summary.c (mh_summary_check, sort_uid_cmp): Sort MH messages by message number (uid), like we sort maildir messages by date. * camel-local-provider.c (mh_provider): Turn on source and store flags. 2002-06-03 Not Zed <NotZed@Ximian.com> * camel-vee-folder.c (camel_vee_folder_add_folder): (folder_changed_change): (folder_changed_remove_uid): (folder_changed_add_uid): (vee_folder_build_folder): Check that any unmatched operation is not concerning any vfolder. Other vfolders should never be part of any unmatched handling. For #24615, and others probably. 2002-06-02 Not Zed <NotZed@Ximian.com> * camel-sasl.c (camel_sasl_new): (camel_sasl_finalize): setup/free the mech string. * camel-sasl.h: Added 'mech' mechanism string. 2002-06-01 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-folder.c (imap_getv): Implement. Only the object_description arg. (camel_imap_folder_get_type): Init parent_class holder. * providers/local/camel-local-folder.c (local_getv): Implement, object_description arg. * camel-folder.c (folder_getv): Implement, add a bunch of args you can get -> camel_folder_get_unread_count etc will be going RSN i hope. (camel_folder_finalize): Free cached description string. * camel-object.c (cobject_getv): Implement CAMEL_OBJECT_ARG_DESCRIPTION, just return the classname of the object. (camel_object_getv): (camel_object_get): (camel_object_setv): (camel_object_set): Take object = void *, to simplify usage. (camel_object_setv): Removed unecessary locals. (camel_object_getv): Same. (camel_object_free): New method, free an arg, upto implementations whether args are static/const or not. (cobject_free): Implement a dummy do nothing free. 2002-05-31 Not Zed <NotZed@Ximian.com> * camel-vee-folder.c (camel_vee_folder_get_location): new function to get the real location (folder) (and uid) of a vfolder object. Using the folderinfo, since we already have it, maybe it should use the uid. 2002-05-31 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (get_content): When the part is a multipart/signed, make sure to free the part_spec before returning (the other types already did this properly). * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_get): Free the path buffer when we're done with it. * providers/imap/camel-imap-folder.c (imap_update_summary): Free the final tagged response buffer. * providers/imap/camel-imap-command.c (imap_read_untagged): If we fail to read a literal, free the temporary string buffer. * providers/imap/camel-imap-folder.c (imap_rescan): Revert my fix from the other day since camel_imap_command_response() doesn't guarentee that resp will be set to NULL on error. * camel-data-cache.c (camel_data_cache_get): If we fail to be able to create a stream to insert into the cache, then free the 'real' path. 2002-05-30 Not Zed <NotZed@Ximian.com> * camel-cipher-context.h: Added a protocol field for users to use. * camel-stream-filter.c (do_read): Sigh, the pre-size is READ_PAD not READ_SIZE. Big difference. (READ_PAD): Bumped upto 128 from 64, so we can fit a typical line in full. * providers/imap/camel-imap-folder.c (get_content): Changed to load multipart/signed as an opaque block into the right kind of object. * camel-multipart.h (struct _CamelMultipart): Removed the boundary field. It wans't actually used anywhere. * camel-seekable-substream.c (camel_seekable_substream_new_with_seekable_stream_and_bounds): Shortened this stupidly long name to just :new(), its the only way its ever used. Fixed all callers. * camel-multipart-signed.[ch]: new wrapper for multipart/signed types. We need to treat the entire content as a blob of data for transport purposes. This also cleans up a lot of the sign/verify handling. * camel-mime-part-utils.c (camel_mime_part_construct_content_from_parser): Just call camel_multipart_construct_from_parser for multipart parts, also use a multipart_signed for multipart/signed types. * camel-multipart.c (camel_multipart_construct_from_parser): New virtual function for multiparts to buld themselves. (construct_from_parser): Implement for normal multiparts. (toplevel): Got rid of a warning, it'll never be an abstract class. * camel-pgp-context.c (pgp_hash_to_id): (pgp_id_to_hash): Implement. (camel_pgp_context_init): Init the pgp protocol specifier. * camel-cipher-context.c (camel_cipher_id_to_hash): (camel_cipher_hash_to_id): Util virtual methods to handle the cipher id string. * camel-mime-filter-canon.[ch]: A new filter, end/start of line canonicalisation filter. Can escape "From " and strip trailing whitespace, and canonicalise the end of line to crlf or to lf. 2002-05-29 Not Zed <NotZed@Ximian.com> * camel-multipart.h (struct _CamelMultipart): Removed the 'parent', nothing used it, anywhere. Cleaned up some formatting slightly. 2002-05-30 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (camel_imap_store_readline): On error, don't leak the GByteArray buffer. * providers/imap/camel-imap-folder.c (imap_rescan): Free the response before checking for errors. I'm not sure this'll actually fix the memory leak here, but it might? Certainly can't hurt. * camel-block-file.c (camel_block_file_finalise): Destroy the blocks hash table. 2002-05-29 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (connect_to_server): Only free the LOGOUT response if it is non-NULL. 2002-05-28 Not Zed <NotZed@Ximian.com> * providers/local/camel-maildir-folder.c (maildir_refresh_info): Implement. Run a summary check to update for any newly arrived messages. * providers/local/camel-maildir-store.c (scan_dir): If we have a folder, execute a refresh_info on it, to suck in any new stuff. 2002-05-28 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_log): Use the CamelMessageInfo rather than the CamelMimeMessage because the message may not have been loaded (thus NULL). 2002-05-27 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (connect_to_server): On connection failure, unref the tcp stream. * providers/pop3/camel-pop3-store.c (connect_to_server): On connection failure, unref the tcp stream. * providers/imap/camel-imap-store.c (connect_to_server): On connection failure, unref the tcp stream. * providers/nntp/camel-nntp-store.c (nntp_store_query_auth_types): Our parent class is no longer CamelRemoteStore. (nntp_store_init): Same. (camel_nntp_store_get_type): Here too. (nntp_connect): Rewritten to try to connect via SSL. Also remove code using CamelRemoteStore. (camel_nntp_command): Don't call camel_remote_store_connected(). 2002-05-24 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (content_info_get_part_spec): Helps if we allocate enough space here. Also, start smoking the same purple flavoured IMAP crack when counting parts (parts don't count if their parent part is a message/* part with a parent part). Fixes bug #25260. 2002-05-24 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-command.c (imap_read_untagged): Use the new readline function. * providers/imap/camel-imap-store.c (connect_to_server): Use the new camel_imap_store_readline() function which doesn't suck quite as bad as the original camel_remote_store_recv_line() function. (camel_imap_store_readline): New function to replace camel_remote_store_recv_line(). This function is at least safe with embedded nul chars. Not that any of our callers use it *sigh*. 2002-05-24 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (connect_to_server): Added some NULL protection fixes. 2002-05-22 Jeffrey Stedfast <fejj@ximian.com> * camel-remote-store.c: Removed from the build. Glory glory hallelujah. * camel-disco-store.c: Updated to inherit from CamelStore rather than CamelRemoteStore. * providers/imap/camel-imap-command.c (imap_command_start): Don't use the camel-remote-store shit to send a string. Just use camel_stream_printf for chrissakes. * providers/imap/camel-imap-store.c: Updated to not depend on CamelRemoteStore and to handle STARTTLS. (imap_disconnect_online): Unref the streams. (imap_keepalive): Removed. (camel_imap_store_connected): New function to replace camel_remote_store_connected(). (camel_imap_store_finalize): Unref the streams. (camel_imap_store_recv_line): New function to replace camel_remote_store_recv_line(). (imap_get_capability): Renamed from connect_to_server (connect_to_server): New function to try and connect to the server. (connect_to_server_wrapper): New wrapper function around connect_to_server that takes the ssl modes into consideration. (query_auth_types): Don't bother calling our parent's implementation of query_auth_types() since CamelDiscoStore doesn't have any anyway. (imap_get_name): New method to implement CamelService::get_name 2002-05-18 Not Zed <NotZed@Ximian.com> * camel-search-private.c (camel_utf8_getc): If we get an invalid start char, just ignore it and goto the next character. 2002-05-16 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (get_folder_offline): Don't call some retarded function to simply set an exception. * camel-filter-search.c (get_label): New e-sexp callback function to get a user_tag label value. 2002-05-15 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-utils.c (imap_parse_namespace_response): New function to parse a NAMESPACE response properly. (imap_namespaces_destroy): New function to destroy the returned structure from the above function. 2002-05-15 Jeffrey Stedfast <fejj@ximian.com> * camel-service.c (service_setv): Only reconnect if we are already connected, if we are in a disconnected state then no need to reconnect. * providers/imap/camel-imap-folder.c (get_content): If the part_spec is an empty string for a leaft part, use "1" since we don't actually want to get the raw message headers too. 2002-05-15 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (refresh_folder_info): removed. (imap_store_refresh_folders): Copy the folders first, then refresh them, outside of the cache_lock, which could cause deadlocks because of a workaround for crappo exchange. (imap_disconnect_online): Dont pass an exception to LOGOUT command. The required response 'BYE' always sets an exception when we call LOGOUT. This also interfered with a lot of other processing causing partial failures and messed up offline/online state. * camel-disco-folder.c (disco_prepare_for_offline): Do progress reporting. 2002-05-14 Jeffrey Stedfast <fejj@ximian.com> Fixes bug #24136. * providers/imap/camel-imap-folder.c (content_info_get_part_spec): New function to take a CamelMessageContentInfo and generate a part-specification string. (get_content): Stop passing around part_spec strings and use content_info_get_part_spec instead. * camel-folder-summary.c (camel_content_info_dump): Made this into a public debugging function. * providers/imap/camel-imap-utils.c (imap_parse_body): Make sure to set the parent of any message/rfc822 subparts. 2002-05-13 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (imap_get_message): Previous fix reverted. 2002-05-13 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (camel_imap_store_class_init): Overload the setv/getv CamelObject virtual methods. (imap_setv): Implemented. (imap_getv): Implemented. 2002-05-13 Dan Winship <danw@ximian.com> * camel-folder.c (camel_folder_append_message): Add a "char **appended_uid" argument, for the caller to optionally pass in a variable to receive the UID of the appended message (if the provider knows it). (camel_folder_transfer_messages_to): Likewise, add "GPtrArray **transferred_uids" (transfer_messages_to): Update default implementation to handle transferred_uids. * camel-disco-folder.c (disco_append_message, disco_transfer_messages_to): Update for API changes. * camel-disco-diary.c (camel_disco_diary_replay): Update the diary's temporary uid->real uid map when replaying appends and transfers. * providers/imap/camel-imap-folder.c (imap_append_offline, imap_append_online, imap_transfer_offline): Pass back the new UIDs, when requested and available. (imap_append_resyncing): Pass back the new UIDs when requested and available. Remove the diary uidmap managing code since CamelDiscoDiary can handle that itself now. (imap_transfer_online, imap_transfer_resyncing): Update for new APIs, but don't actually pass back the new UIDs yet. (It's tricky since the COPYUID response may not be in the same order as the input uids.) * providers/local/camel-maildir-folder.c (maildir_append_message): Pass back the new UID if requested. * providers/local/camel-mbox-folder.c (mbox_append_message): Likewise. * providers/local/camel-mh-folder.c (mh_append_message): Likewise. * providers/local/camel-spool-folder.c (spool_append_message): Likewise. * camel-digest-folder.c (digest_append_message, digest_transfer_messages_to): Update for API changes. * camel-filter-driver.c (camel_filter_driver_filter_message, do_copy, do_move): Update for API changes. * camel-vee-folder.c (vee_append_message, vee_transfer_messages_to): Likewise. * camel-vtrash-folder.c (vtrash_append_message, vtrash_transfer_messages_to): Likewise. 2002-05-13 Not Zed <NotZed@Ximian.com> * camel-folder-thread.c (camel_folder_thread_messages_new_summary): (camel_folder_thread_messages_new): Since the fix for #3357 uses 'order=0' to indicate tree structure nodes, make sure we dont actually set order=0 for valid messages, otherwise we silently lose duplicates of the first message (i==0). Fixes #19920. 2002-05-10 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (imap_get_message): Fetch the BODYSTRUCTURE rather than BODY since BODY seems to be lacking some of the data we need. This fixes bug #24136. * camel-transport.c (camel_transport_get_type): Might help if we called camel_transport_class_init. 2002-05-10 Dan Winship <danw@ximian.com> * camel-folder.c (camel_folder_transfer_messages_to): Replace copy_messages_to and move_messages_to with a single function that just takes a "delete_originals" flag. Also, use the vtrash implementation if *either* folder is a vtrash. (transfer_messages_to): Make this use camel_operation_progress (previously move_messages_to did but copy_messages_to didn't), and freeze/thaw the folder(s) if doing multiple messages. * camel-vtrash-folder.c (vtrash_transfer_messages_to): Update for move/copy merge. Move the "move messages into vtrash" code here from mail-ops.c. Now all of the vtrash move/copy special casing is in camel instead of half of it being here and half in mail/. (This should also make it so that "Move to Trash" will work in filter rules.) * camel-vee-folder.c (vee_transfer_messages_to): Make this just return an exception, since it will only be called when trying to move/copy messages from one vfolder to another. (vee_append_message): Add this too so we get a nicer error message than the default "unimplemented" one in camel-folder.c. * camel-digest-folder.c: Replace copy_messages_to and move_messages_to with transfer_messages_to. * camel-disco-folder.c: Likewise * camel-disco-diary.c (camel_disco_diary_log, camel_disco_diary_replay): replace MOVE/COPY with TRANSFER. * providers/imap/camel-imap-folder.c (imap_transfer_offline, imap_transfer_online, imap_transfer_resyncing): Update for changes. (This ends up being a bit more complicated than it was before for now, but later disconnected operation changes should resimplify it.) * camel-filter-driver.c (camel_filter_driver_filter_message, do_copy, do_move): Use transfer_messages_to instead of copy. 2002-05-09 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-search.c (shell_exec): New filter function to pipe a message to another program. (run_command): Fixed some bugs to make this work. 2002-05-09 Not Zed <NotZed@Ximian.com> * camel-service.c (camel_service_disconnect): Instead of testing for SERVICE_CONNECTED, we need to also handle SERVICE_CONNECTING too, as it will often have setup some details before it failed. Make it !DISCONNECTED (and !DISCONNECTING for recursive calls, which happen). Fixes #23782, and maybe also #21604 and many other random crashes. 2002-05-08 Jeffrey Stedfast <fejj@ximian.com> * camel-digest-store.c (digest_setv): Implemented. (digest_getv): Implemented. * camel-disco-store.c (disco_setv): Implemented. (disco_getv): Implemented. * camel-remote-store.c (remote_store_setv): Implemented. (remote_store_getv): Implemented. * camel-transport.c (camel_transport_class_init): Implemented. (transport_setv): Implemented. (transport_getv): Implemented. * camel-store.c (store_setv): Implemented. (store_getv): Implemented. * camel-service.c (service_setv): Implemented. (service_getv): Implemented. * providers/pop3/camel-pop3-store.c (pop3_try_authenticate): camel_pop3_engine_iterate doesn't return the state, it returns -1 on fail, 0 when finished processing request or >0 if more ops are in the queue, so don't check status against CAMEL_POP3_STATE_OK, instead check pcp->state against that. 2002-05-07 Jeffrey Stedfast <fejj@ximian.com> * camel-object.h: s/class/klass for arguments so that c++ developers don't complain later. 2002-05-07 Dan Winship <danw@ximian.com> * camel-object.c (camel_object_class_cast): Fix a crash in a g_warning. 2002-05-07 Not Zed <NotZed@Ximian.com> * camel-remote-store.c (remote_send_string): Check for LOGIN xxxx as well if debug is on, so we dont print passwords to evolution logs. * providers/imap/camel-imap-utils.c (imap_is_atom_char): This was really broken. 1. isprint() is locale dependent, and 2. it looked up an 8 bit value in a 7 bit table without truncating it. I've removed the isprint() stuff and just put it directly into the special table, which i've expanded to the right size too. * providers/imap/*: Applied patch from Preston Elder <prez@magick.tm> to make camel only use literals if it needs to for simple strings. Changed slightly to use imap_is_atom() and more consistent formatting. providers/imap/camel-imap-utils.c (imap_is_atom): Chagned from imap_needs_quoting(). ** Merged in camel-object2 branch. Simpler camelobject implementation + object args interface. * camel.c (camel_init): Call camel_object_get_type() to make sure camel_object_type is initialised. * camel-object.h (CAMEL_OBJECT_TYPE): Changed to return global camel_object_type pointer, not call camel_object_get_type. 2002-05-06 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-store.c (pop3_try_authenticate): If the pop3 command status is -1, then we probably have a TCP error (?) so set a SYSTEM exception so our caller can distinguish between a "bad password" and a "tcp error". (pop3_connect): Only uncache the password on "bad password" errors. * camel-pgp-mime.c (pgp_mime_part_sign_prepare_part): Use CamelMimeFilterBestenc to get a more appropriate encoding rather than just blindling assigning QP. * providers/imap/camel-imap-folder.c (do_append): Call camel_mime_message_encode_8bit_parts() which fixes bug #10885. * camel-tcp-stream-raw.c (socket_connect): Don't bother with counting down the timeout. 2002-05-02 Jeffrey Stedfast <fejj@ximian.com> * camel-stream-fs.c (stream_read): Same fix as the tcp stream. (stream_write): Again here. Just like tcp stream's stream_write(), also make sure to save errno before calling fcntl to restore the fd flags. * camel-tcp-stream-raw.c (stream_read): Handle EINTR errors for select(). (stream_write): Same and also preserve errno when setting the fd flags back. If w == -1, return -1. 2002-05-02 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_decode_mailbox): Fixed the fix for stupidly-broken-mailer bug #5 to work when multiple unescaped characters were in a row. Fixes bug #24140. * camel-tcp-stream-raw.c (socket_connect): Check select() for EINTR errors. * camel-pgp-context.c (crypto_exec_with_passwd): Change the order of some code so that we check if the select() failed before we check for user-cancellation. * camel-service.c (camel_gethostbyname): Check for EINTR when select()ing. (camel_gethostbyaddr): Same. 2002-04-29 Jeffrey Stedfast <fejj@ximian.com> * camel-provider.c (camel_provider_auto_detect): Now takes a CamelURL argument rather than a GHashTable *settings argument. 2002-04-29 Not Zed <NotZed@Ximian.com> * providers/local/camel-spool-store.c (get_folder_info): We want to set unread_count to get_unread_message_count, not get_message_count(). Might fix #17174. Also removed FIXME: as it was fixed. 2002-04-26 Jeffrey Stedfast <fejj@ximian.com> * providers/local/Makefile.am: Don't link to libibex.a anymore. * providers/nntp/Makefile.am: Same. * providers/imap/Makefile.am: And again here. * camel-store-summary.h: No longer want to #include <libibex/ibex.h> * camel-provider.c (camel_provider_auto_detect): New function to auto-detect configuration settings. 2002-04-26 Not Zed <NotZed@Ximian.com> * camel-block-file.c (block_file_validate_root): Remove the excessive \n's, after printfs. * camel-text-index.c (text_index_compress_nosync): @!#$@#$!@$#!. Since the rename op was fixed, this broke compression's rename, resulting in the index 'vanishing' after every compress (i.e. after every reindex). Fix this code to account for the fixed rename operation. 2002-04-25 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_rcpt): Don't put a space between the "RCPT TO:" and the "<recipient>" strings - rfc0821 was not clear on this but it seems rfc2821 defines a grammar excluding that SP. (smtp_mail): Same. 2002-04-24 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-local-provider.c (camel_provider_module_init): Configure the default paths for mh, mbox, maildir, spools, etc. * camel-provider.h: Add a CAMEL_PROVIDER_CONF_LABEL enum and define some default CamelProviderConfEntry macros. 2002-04-19 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-folder.c (pop3_refresh_info): Free the LIST pop3 command. * camel-data-cache.c (data_cache_finalise): Free the cdc->path. * camel-multipart.c (write_to_stream): Don't g_return_val_if_fail here if the boundary is an empty string. See bug #23676 for details. The way I see it, we have 2 options: 1) leave this fix the way it is, thus allowing multipart boundaries to be empty-strings; or 2) make camel_multipart_get_boundary() change the boundary to something legal if the boundary is an empty-string. Since the parser should be able to handle an empty-string boundary *and* more importantly because we want to keep the same boundaries as the original raw message so as to be able to verify multipart/signed parts, I vote for solution #1. 2002-04-19 Not Zed <NotZed@Ximian.com> * devel-docs/camel-index.txt: Start of a white-paperish document describing camel-index and older libibex. 2002-04-18 Not Zed <NotZed@Ximian.com> * providers/local/camel-local-store.c (rename_folder): If we get a failure, make sure we set an exception. * camel-text-index.c (camel_text_index_rename): If the file doesn't exist, just assume it never did, dont return failure. (text_index_rename): Add '.index' to the path name we're using, since we dont get it passed in. * camel-folder-search.c (check_header): When doing a contains match, split the words and perform an and on it. (match_words_messages): If we have an index, but were forced to do a full search, first lookup a subset of messages using the index and a simplified word set. Only do a manual search of this subset. 2002-04-17 Not Zed <NotZed@Ximian.com> * camel-folder-search.c (match_message_index): Changed to take a utf8 string not a regex pattern. (match_words_index): Matches against a camel_search_words list. (match_words_1message): Matches a single message against a camel_search_words list. (match_words_message): Same, but gets the message from the folder for you. (match_words_messages): Matches a list of messages against a words list. (search_body_contains): Rewritten to handle multiple word searches. For #23371. * providers/imap/camel-imap-search.c (sync_match): Split words when searching, to support multiple search words. Also, try searching specifying charset of utf8 if we can, if that fails, fall back to not specifying charset. TODO: It should translate the strings into the locale default charset? * providers/imap/camel-imap-store.c (connect_to_server): Added new cap - utf8_search, if set, we tell the server we're searching using utf8, otherwise we dont (incorrectly, since we always use utf8 to search). * camel-search-private.c (camel_ustrstrcase): Make this class public. (camel_search_words_split): Split a word into multiple words based on whitespace, and keep track of whether the word is simple (indexable directly), or not. (camel_search_words_free): Free 'em. 2002-04-17 Jeffrey Stedfast <fejj@ximian.com> * camel-vee-folder.c (vee_search_by_expression): If the vee-folder is the unmatched, we don't have our own expression so we cannot merge them. Instead, just use the expression passed in. This fixes a Null-Pointer-Read crash on Solaris systems at least. 2002-04-16 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_folder): Get rid of an unused variable. * providers/smtp/camel-smtp-transport.c (smtp_helo): Use camel_gethostbyaddr since gethostbyaddr is not reentrant. * camel-http-stream.c (http_connect): Updated after the rename of camel_get_host_byname. * camel-service.c (camel_gethostbyname): Renamed. (camel_gethostbyaddr): New cancellable/reentrant version of gethostbyaddr. 2002-04-14 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-spoold-store.c: Added #include <sys/types.h> for dirent.h which needs it on MacOS X. * providers/local/camel-maildir-store.c: Same. * providers/nntp/camel-nntp-store.c: Same. * providers/imap/camel-imap-message-cache.c: Same. * camel-provider.c: Same. * camel-data-cache.c: Same. 2002-04-12 Jeffrey Stedfast <fejj@ximian.com> * broken-date-parser.c (datetok): Treat commas as token delimeters. (WEEKDAY_CHARS): Use full weekday names in case the broken mailer used the full names. (MONTH_CHARS): Same as above but for months. 2002-04-15 Not Zed <NotZed@Ximian.com> * *.c: Fix callers for api changes to camel-object, mainly declare_event->add_event, and classfuncs->klass, and a couple of missing #include <config.h>'s * camel-arg.[ch], Makefile.am: New support code for camel_object_get/set arg. * *.h: Fixed all uses of get_type to return a CamelType rather than a guint (now a pointer). * camel-object.[ch]: Major cleanup of object implementation. Also added a get/set interface, and some debugging options. 2002-04-11 Not Zed <NotZed@Ximian.com> * providers/local/camel-spool-summary.c (spool_summary_sync_full): If the last message(s) were deleted, and we had any messages output, account for the lost \n of the following From line by adding an extra \n. fix for #8214. 2002-04-10 Not Zed <NotZed@Ximian.com> * camel-mime-part-utils.c (convert_buffer): If we get a 0 length input, return a 0 lenght output as valid - fixes bugs with some iconv impl and its simpler anyway. 2002-04-11 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c: Set the UID_SET_LIMIT value to 4096. I ran into an issue tonight where apparently the IMAP server changed the UIDVALIDITY and so Evo needed to re-fetch all headers and it was trying to send a uid set of some 25k (yes, I have a very large INBOX). Anyways, it was set to unlimited before. Courier IMAPd can safely handle up to ~16k per token, but UW IMAPd can only handle 8k per command-line, so I set it to 4k just to be safe. 2002-04-10 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part-utils.c (convert_buffer): Fixed a bug that would miscalculate how much data to copy into the GByteArray (negative value) thus causing a segfault. Also optimized it while I was at it. 2002-04-09 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (camel_store_init): Make the folder_lock recursive. See bug #22363 for details. Basically, get_folder_info() is requesting a diary folder which in turn connects which requests then calls get_folder() but deadlocks because get_folder_info already holds the lock. * camel-mime-message.c (camel_mime_message_set_date): Don't adjust the timezone offset if we used tm.tm_gmtoff because it is already adjusted. 2002-04-09 Not Zed <NotZed@Ximian.com> * camel-mime-part.c (construct_from_parser): If we get multiple Content-Type header values, change subsequent headers to X-Invalid-Content-Type so it doesn't wreck processing. This fixes the reported case in #18929, but i dont know if it fixes the original posters problems. 2002-04-08 Not Zed <NotZed@Ximian.com> * camel-vtrash-folder.c (vtrash_move_messages_to): If we find we're moving from the vtrash to another folder, we need to convert the uid from a vfolder uid to the source uid (+8). Fix for #20886. Also changed to batch multiple moves to different folders so they are done as efficiently as possible rather than one at a time. * camel-mime-utils.c (base64_decode_step): If we only get passed '=', we back track only if we actually output any data. Fix for #21716. (quoted_decode): Pass out size_t instead of int, and use 0 instead of -1 for error since its not signed. This will fix similar bug to above in different circumstances since the result is taken as unsigned. This is only an internal func. (quoted_encode): Return size_t just for consistency. * camel-block-file.c (block_file_validate_root): Comment out the debug and move it into a warning when the validation fails. 2002-04-08 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (uuencode_close): Don't count our filler when encoding our line-length octet. 2002-04-05 Jeffrey Stedfast <fejj@ximian.com> * camel-http-stream.c (http_get_headers): Don't get the statuscode here anymore. (http_method_invoke): Use a User-Agent header and do basic proxy authentication. (stream_read): Handle redirects. (camel_http_stream_set_user_agent): New function to allow client to set the User-Agent string. (camel_http_stream_set_proxy): New function for setting the proxy server. (camel_http_stream_set_proxy_authrealm): New function for setting the proxy auth realm. (camel_http_stream_set_proxy_authpass): New function for setting the proxy auth password. 2002-04-04 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.c (message_info_new): Simplified since we can now decode in-reply-to without getting extra cruft. Get rid of the FIXME about having to check scan->id because of the possibility of it being NULL, this can no longer happen. * camel-mime-utils.c (header_references_inreplyto_decode): New function to decode in-reply-to headers. Only grabs the first thing that looks like a message-id and then returns. (header_references_decode): Loop calling header_references_decode_single (a new internal function). 2002-04-04 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-search.c (imap_body_contains): If (body-contains) is not passed any arguments, return empty/false. Fixes a crash exposed by #15001. * camel-remote-store.c (remote_connect): Reset the keepalive timeout to 10 minutes rather than the 30 seconds I was using for testing. 2002-04-03 Dan Winship <danw@ximian.com> * camel-provider.h (CamelProvider): make service_cache be an array of CAMEL_NUM_PROVIDER_TYPES elements so you can have a single provider offer both stores and transports. (Eg, Exchange, NNTP) * providers/imap/camel-imap-provider.c: Don't initialize service_cache here. (The session code can do it itself since the url_hash and url_equal functions are stored as part of the provider.) * providers/nntp/camel-nntp-provider.c: Likewise. * providers/local/camel-local-provider.c: Likewise. * providers/pop3/camel-pop3-provider.c: Likewise. * providers/sendmail/camel-sendmail-provider.c: Likewise. * providers/smtp/camel-smtp-provider.c: Likewise. * camel-session.c (register_provider): Initialize the provider's service cache(s) here. (camel_session_class_init): Don't initialize. vee_provider.service_cache here. (camel_session_destroy_provider): Update to destroy multiple service_caches. (service_cache_remove, get_service): Tweak these a bit to deal with multiple service_caches. 2002-04-02 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (set_errno): Handle a ton more nspr i/o errno's. (stream_connect): Act as if we are doing a non-blocking connect. This is to try and work around bug #15120 where users get an EINPROGRESS error. Maybe importing a PRFileDesc into SSL mode automagically makes it non-blocking? I dunno. 2002-04-01 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.c (message_info_new): Updated the construction of the references to match JWZ's updated algorithm initialization (ie, append any In-Reply-To reference onto any References header and never take more than a single message-id from the In-Reply-To header since anything after the first will probably just be email addresses). Fixes bug #1336. 2002-04-03 Not Zed <NotZed@Ximian.com> * providers/local/camel-local-folder.c (camel_local_folder_construct): Turn indexing back on, fingers crossed ... * camel-block-file.c (sync_nolock): #!@$@$#@~#$ DF@#$!Q@$#!@$#!#%. Well it helps if we're iterating a list to iterate the node pointer ... * camel-text-index.c (text_index_sync): Sync the key tables explcitly. (text_index_sync): Debug out frag info. (camel_text_index_dump): Added a (rather large, but optional) raw dumping mode for debugging purposes. * camel-partition-table.c (camel_key_table_finalise): Sync root block when done. (camel_key_table_sync): New function, sync key table (root) explicitly. (camel_partition_table_sync): Method to explicitly sync the partition table. 2002-04-02 Not Zed <NotZed@Ximian.com> * camel-block-file.c (camel_block_file_free_block): Mark root block dirty when we change it (this function isn't used yet anyway). * camel-text-index.c (text_index_add_name_to_word): Touch the root block when we modify the counts. Also, abort processing on any errors. (text_index_sync): Fix typo in comments. Sync the block file inside the lock. (text_index_compress_nosync): Lock the old index while we're compressing. (text_index_compress_nosync): Remove the bogus while() at the end of the while() loops! Also plug a memleak - records weren't freed. (text_index_rename): Lock around rename op. (text_index_add_name): More typos. (text_index_sync): Touch root when changing it. (text_index_add_name): " (text_index_delete_name): " (camel_text_index_new): Touch root if we change it. (text_index_cursor_reset): Make sure we NULL pointers after we free them (nothing uses this yet). * camel-partition-table.c (hash_key): Remove some debug accidentally left in. (camel_partition_table_add): When linking in the next block list, set the right previous pointer. (camel_key_table_add): Simplify the 'left' calculation (it was already ok though). (camel_key_table_next): Initialise returns before processing. Broaden the lock slightly, and simplify validity calculations. * providers/imap/camel-imap-store.c (imap_keepalive): Put back in the exception setup stuff i disabled for debugging. * providers/local/camel-local-folder.c (camel_local_folder_construct): Temporarily disable indexing. 2002-03-28 Not Zed <NotZed@Ximian.com> * camel-partition-table.c (camel_key_table_lookup): Change range checking assert to a warning. * providers/pop3/camel-pop3-folder.c (pop3_finalize): Make sure we flush out all outstanding commands before finalising, stops being finalised while outsanding requests are processed by the store finalise. (pop3_get_message): Instead of pre-fetching all messages, just pre-fetch a maxiumum number at any one time, stops us running out of cache fd's. * providers/nntp/camel-nntp-folder.c (nntp_folder_init/finalise): Setup priv data + locks, & free. * providers/imap/camel-imap-folder.c (imap_rescan): Batch all message_chagned events into a single folder_changed event (otherwise updates can be >>> expensive, like >5 hours for 80K messages changing!). Alternately it could use folder freeze/unfreeze perhaps. 2002-03-27 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (imap_keepalive): Pass an exception to called code so it behaves properly since it uses the passed exception to check returns. 2002-04-01 Dan Winship <danw@ximian.com> * providers/imap/Makefile.am (libcamelimap_la_LDFLAGS): Use -avoid-version instead of -version-info 0:0:0, and specify -module. (From Max Horn <max@quendi.de>). * providers/local/Makefile.am (libcamellocal_la_LDFLAGS): Likewise. * providers/nntp/Makefile.am (libcamelnntp_la_LDFLAGS): Likewise. * providers/sendmail/Makefile.am (libcamelsendmail_la_LDFLAGS): Likewise. * providers/smtp/Makefile.am (libcamelsmtp_la_LDFLAGS): Likewise. * providers/pop3/Makefile.am (libcamelpop3_la_LDFLAGS): Likewise. Also remove $(KRB4_LDFLAGS) since KPOP is gone. (INCLUDES): and $(KRB4_CFLAGS) 2002-03-28 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (do_copy): We now have to check to make sure that p->message is non-NULL because we only load the message when we have to. (do_move): Same here. 2002-03-28 Dan Winship <danw@ximian.com> * camel-transport.c (camel_transport_send_to): Change the message arg to a CamelMimeMessage instead of a CamelMedium. Even the NNTP provider returns CamelMimeMessages, and we're never going to support anything more exotic than that. Also do a few more g_return_if_fails here instead of in the providers. (camel_transport_can_send): No longer needed. (camel_transport_send): Remove this too. It wasn't being used any more, and it doesn't behave exactly the same in sendmail and smtp. * providers/smtp/camel-smtp-transport.c (smtp_send, smtp_can_send): Gone. (smtp_send_to): Update for arg change. (smtp_data): Make this take a CamelMimeMessage too. * providers/sendmail/camel-sendmail-transport.c (sendmail_send, sendmail_can_send): Gone. (sendmail_send_to): Update for arg change, and merge in the part that used to be shared with sendmail_send. 2002-03-27 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_folder): Construct the source_url the right way. The previous way was generating urls like pop://fejj@ximian.com;keep_on_server/inbox which is wrong. 2002-03-26 Not Zed <NotZed@Ximian.com> * camel-text-index.c (text_index_normalise): Changed to use just g_utf8_strdown instead of utf8_normalise, to match the indexing code. utf8_normalise is just far too expensive (saves approx 25% total processing). 2002-03-25 Not Zed <NotZed@Ximian.com> * camel-text-index.c (text_index_add_name): When we add a new name, up all of the cache limits, because we're probably going to be adding more. (text_index_sync): Drop the cache limits back down again, we dont need them when looking words up. ** MERGE camel_index branch. * camel-text-index.[ch]: Added files i forgot to add (eep nearly lost all this work!) * camel-block-file.c (sync_nolock): Fix an infinite loop in syncing. 2002-03-21 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.c (camel_message_info_new_from_header): Use the date in the received header for the received_date. 2002-03-22 Not Zed <NotZed@Ximian.com> * providers/local/camel-local-folder.c (camel_local_folder_construct): Use the right option to remove the index file and reset the index file on creation. (camel_local_folder_construct): Remove any existing '.ibex' files - these are the old format index files. * camel-block-file.c (camel_block_file_rename): Lock io lock while renaming. (camel_key_file_rename): Lock around rename. (block_file_validate_root): Implement sync flag checking. (camel_block_file_touch_block): Turn off the sync flag if we're touching any non-root block and write it to disk. Shoudl this fsync()? (sync_nolock): sync the root block only if we need to. * providers/local/camel-local-store.c (rename_folder): Re-enable index rename code. Not sure how race-free it is though. (delete_folder): Delete the index file properly. * camel-partition-table.c (camel_key_table_lookup): Initialise output values to 0 before doing anything. (camel_key_table_add): Do some range-checking on values. * camel-text-index.c (text_index_compress): Changed to call sync here. (text_index_compress_nolock): and not here - stops a recursive sync call when sync performs a compress also. (text_index_compress_nolock): Change to _nosync, since the locking is irrelevent (recursive lock). Fixed callers. (text_index_add_name_to_word): If we get a failure with key table ops, fail immediately. (text_index_compress_nosync): Likewise. (text_index_write_name): If the nameid is 0, do nothing. (text_index_add_name): If we can't get a keyid, dont add it to the partition table. (camel_text_index_remove): Function to delete an index file. (text_index_compress_nosync): Clean up temp files when done. * camel-folder-search.c (match_messages_index): New function, split from body_contains, matches a regex against all words in an index. (match_message_index): Similar to above but matches against an individual message. (search_body_contains): Changed to use above functions for matching - substring matches should now work on indexed data. 2002-03-21 Not Zed <NotZed@Ximian.com> * camel-index.c (camel_index_words/names): New virtual methods/stubs to get a cursor of all words and names. * camel-text-index.c (text_index_compress_nolock): Split from text_index_compress, so we can call the compressor while locked also, removed lock calls. (text_index_compress): Changed to stub which calls text_index_compress_nolock. (camel_text_index_key_cursor_new): New object to iterate through a key table. (text_index_words, text_index_names): Implement virtual functions for iterating through all words or names. * camel-block-file.c: Turn off some debug. 2002-03-20 Not Zed <NotZed@Ximian.com> ** New body index implementation. * camel-index*: Code for camel index, a new class to replace ibex. * camel-block-file.[ch]: block-based and link based filesystem-in-a-file classes. * camel-partition-table.[ch]: An implementation of a partition table (automatically extending on-disk hash-table using ideal hash), and a key-table, a key<>name mapping table. Used by camelindex. * providers/local/*, camel-folder-summary.[ch]: Changed to use camel-index interface rather than ibex. 2002-03-05 Not Zed <NotZed@Ximian.com> * providers/local/camel-maildir-summary.c (maildir_summary_check): Do progress reporting of operations. (maildir_summary_sync): Same here. 2002-03-04 Not Zed <NotZed@Ximian.com> * providers/local/camel-spoold-store.c (scan_dir): Kill a warning with a cast. * providers/local/camel-*.c: Changed for ibex->camelindex. * camel-folder-search.c (camel_folder_search_set_summary): Init summary_hash to point to 'static' uid strings. (search_body_contains): Use the static uid memory to return results rather than the values from the index library. * camel-folder-search.[ch]: Changed to use camelindex object. * camel-folder-summary.c (summary_build_content_info_message): Use a stream to index content, also filter html mail first. (camel_folder_summary_info_new_from_message): Use a stream filtered to index content. (main): Removed the test main code. Added headers for open call (ibex must've had them before). * camel-folder-summary.[ch]: Changed from ibex to CamelIndex. * camel-mime-filter-index.c (camel_mime_filter_index_finalize): Unref index. * camel-mime-filter-index.[ch]: Changed from ibex to CamelIndex. 2002-03-19 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_encode_param): Fix this to work right. We need to convert the input buffer to the charset we claim in the encoded param (duh). 2002-03-18 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (connect_to_server_wrapper): Updated to use the same logic as the POP code. * providers/pop3/camel-pop3-store.c (connect_to_server): No longer takes a stls_supported argument since we no longer need it with the new logic. (connect_to_server_wrapper): New logic: First try connecting to the SSL port (995 by default), if that fails with SERVICE_UNAVAILABLE, then we attempt to connect (to port 110 by default) and try to use STARTTLS. 2002-03-15 Jeffrey Stedfast <fejj@ximian.com> * camel-folder.h: * camel-private.h: Don't allow any empty structs. If !ENABLE_THREADS, provide a gpointer dummy member. Fixes bug #6382. 2002-03-13 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_auth): Added a work-around for SMTP servers that can't read the RFCs and thus implement SASL incorrectly. Oh well, that's life in the world of mail clients I guess. 2002-03-12 Jeffrey Stedfast <fejj@ximian.com> * camel-digest-store.c (camel_digest_store_new): Now takes a url argument. * camel-digest-folder.c (digest_add_multipart): Fixed some memory corruption and also modified to use CAMEL_IS_MIME_MESSAGE() rather than comparing content-type strings. (digest_get_message): Fixed a logic blooper. * camel-folder-summary.c (camel_message_info_new_from_header): Set the date fields of the CamelMessageInfo as well. This may even fix some filter-related bugs where the user was trying to compare dates. 2002-03-11 Jeffrey Stedfast <fejj@ximian.com> * camel-digest-store.c: A pretty empty store implementation to be the parent store of a CamelDigestFolder. * camel-digest-folder.c: Updated to reference it's parent store. 2002-03-11 Jeffrey Stedfast <fejj@ximian.com> * camel-digest-folder.c (camel_digest_folder_new): Allow any leaf part to be a message/rfc822 part. (digest_get_uids): Recurse the mime structure and add all message/rfc822 parts and use a uid scheme similar to IMAP's mime part naming convention. (digest_get_message): Decode the uid to get the correct mime part. 2002-03-11 Ettore Perazzoli <ettore@ximian.com> * camel-mime-utils.c: Change the order of the mailing list magic patterns so that the more mailing-list specific ones are on the top. 2002-03-11 Jeffrey Stedfast <fejj@ximian.com> These fixes should fix bug #21737. * providers/smtp/camel-smtp-transport.c (connect_to_server_wrapper): Same as with the POP code. * providers/pop3/camel-pop3-store.c (connect_to_server_wrapper): Slight restructuring of the if-statements for the USE_SSL_ALWAYS case so that we can't possibly return TRUE unless we really did connect successfully. 2002-03-10 Dan Winship <danw@ximian.com> Rename the OpenSSL implementation of things to match the NSS implementation so that callers don't need to care which one is being used. * camel-tcp-stream-openssl.c: Implement CamelTcpStreamSSL, not CamelTcpStreamOpenSSL. Rename methods as well. Replace the camel-tcp-stream-openssl.h include with camel-tcp-stream-ssl.h. * camel-tcp-stream-openssl.h: Gone. * camel-tcp-stream-ssl.c: Add a note explaining that this implementation is only used for NSS, and that OpenSSL's implementation is in another file. (Should probably do some CVS renaming magic at some point.) * camel-http-stream.c (http_connect): Remove OpenSSL refs; the previously-NSS-specific code works for both now. * camel-remote-store.c: Likewise. * providers/smtp/camel-smtp-transport.c: Likewise. * providers/pop3/camel-pop3-store.c: Likewise. * Makefile.am (libcamelinclude_HEADERS): Remove camel-tcp-stream-openssl.h 2002-03-10 Dan Winship <danw@ximian.com> * camel-tcp-stream.c (camel_tcp_stream_get_socket): Remove this: it couldn't be generically used, because different subclasses returned entirely different types of data. (camel_tcp_stream_get_local_address, camel_tcp_stream_get_remote_address): Add these to replace what get_socket was being used for. (camel_tcp_address_new, camel_tcp_address_free): Utility functions for get_{local,remote}_address. * providers/smtp/camel-smtp-transport.c: Change localaddr to a CamelTcpAddress *. (connect_to_server): Call camel_tcp_stream_get_local_address to get the local IP address. (smtp_disconnect): free localaddr. (smtp_helo): Update for localaddr change. * camel-tcp-stream-raw.c (stream_get_socket): Remove (stream_get_local_address, stream_get_remote_address): Implement. * camel-tcp-stream-ssl.c (stream_get_socket): Remove (stream_get_local_address, stream_get_remote_address): Implement. * camel-tcp-stream-openssl.c (stream_get_socket): Remove (stream_get_local_address, stream_get_remote_address): Implement. 2002-03-08 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-provider.c (camel_provider_module_init): Don't call camel_remote_store_get_authtypes since we no longer subclass camel-remote-store. * providers/pop3/camel-pop3-engine.c: Added STARTTLS to the capabilities to look for. (camel_pop3_engine_reget_capabilities): New function to re-get capabilities. * providers/pop3/camel-pop3-store.c: Updated to not subclass CamelRemoteStore. (connect_to_server): Rewritten to not depend on CamelRemoteStore's connect implementation. Also added support for STLS (aka STARTTLS). 2002-03-07 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-mime.c (camel_pgp_mime_part_sign): Add support for hash type RIPEMD160. * camel-cipher-context.h: Add RIPEMD160 hash type. * camel-pgp-context.c (pgp_sign): Updated to consider hash function for pgp5 and pgp6. (pgp_clearsign): Same. * camel-tcp-stream-openssl.c (stream_read): Add a timeout on the select. (stream_write): Same. 2002-03-06 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (connect_to_server): Fix to work with OpenSSL. * camel-tcp-stream-openssl.c: compile fixes. (camel_tcp_stream_openssl_enable_ssl): Check to make sure that the sockfd != -1, it's not enough to check that it is non-zero. Also set the sockfd to -1 on fail (open_ssl_connection will close the sockfd on fail). 2002-03-06 Dan Winship <danw@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_construct): Make this compile. 2002-03-05 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (save_ssl_cert): Removed. Glory glory hallelujah! (ssl_bad_cert): No longer calls ssl_save_cert or ssl_cert_is_saved. 2002-03-05 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-openssl.c (camel_tcp_stream_openssl_new_raw): Start the ssl stream off in non-ssl mode (useful for STARTTLS). (camel_tcp_stream_openssl_enable_ssl): New function to toggle an ssl stream into ssl mode. (open_ssl_connection): Close the sockfd on fail so our caller doesn't have to - this also allows us to save the original errno. (stream_connect): If we want ssl mode, do our ssl stuff. (camel_tcp_stream_openssl_class_init): Init some SSL stuff here instead of in open_ssl_connection since these only ever need to be called once. (stream_read): Only use SSL_read if we are in ssl mode. (stream_write): Only use SSL_write if we are in ssl mode. * providers/smtp/camel-smtp-transport.c (smtp_helo): Check for the STARTTLS extension. (connect_to_server): Try to use STARTTLS whenever possible rather than the old way of doing things. (connect_to_server_wrapper): Wrapper around connect_to_server() to first try STARTTLS and then attempt normal SSL mode if we can't connect via STARTTLS. * camel-tcp-stream-ssl.c (camel_tcp_stream_ssl_enable_ssl): New function to toggle an ssl stream into ssl mode. (camel_tcp_stream_ssl_new_raw): Start the ssl stream off in non-ssl mode (useful for STARTTLS). (stream_connect): Only connect in SSL mode if required. 2002-03-01 Jeffrey Stedfast <fejj@ximian.com> * camel-vtrash-folder.h: * camel-vee-store.h: * camel-vee-folder.h: * camel-stream-null.h: * camel-stream-filter.h: * camel-store-summary.h: * camel-news-address.h: * camel-mime-utils.h: * camel-mime-parser.h: * camel-mime-filter-save.h: * camel-mime-filter-linewrap.h: * camel-mime-filter-index.h: * camel-mime-filter-html.h: * camel-mime-filter.h: * camel-mime-filter-from.h: * camel-mime-filter-crlf.h: * camel-mime-filter-chomp.h: * camel-mime-filter-charset.h: * camel-mime-filter-bestenc.h: * camel-mime-filter-basic.h: * camel-internet-address.h: * camel-folder-thread.h: * camel-folder-summary.h: * camel-folder-search.h: * camel-filter-driver.h: * camel-charset-map.h: * camel-address.h: Add c++ armoring. * camel-object.h: s/class/klass 2002-03-01 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part-utils.c (camel_mime_part_construct_content_from_parser): Reverted my pgp/mime fixes here too. * camel-mime-part.c (write_to_stream): Removed my pgp/mime raw stream hack, this is causing problems such as some messages to not displaying, view->source not working at all, etc. 2002-02-28 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-parser.c: Changed offset variables from int's to off_t's since the system may support large files. * camel-mime-part-utils.c (camel_mime_part_construct_content_from_parser): Rearrange the save filter stuff so that we save raw streams for all mime parts. If the mime part turns out to be a multupart, then don't bother saving the raw stream, we only need to save the raw stream for leaf parts. 2002-02-27 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.h: Don't #include camel-mime-filter-save.h, we don't use it. * camel-file-utils.c: Fixed a few 'might be used uninitialized' warnings which were real problems. * camel-mime-part-utils.c (camel_mime_part_construct_content_from_parser): Save the raw mime stream for any/all signed parts. * camel-mime-part.c (camel_mime_part_init): Initialize our raw stream to NULL. (camel_mime_part_finalize): Unref our raw stream, if we have one. (write_to_stream): If we have a raw stream, write that out instead of re-encoding. * camel-mime-filter-save.[c,h]: Rewritten to save to a stream rather than a file. 2002-02-28 Not Zed <NotZed@Ximian.com> * camel-mime-utils.c (header_fold): Use the FOLD_SIZE as a recommended folding size, but add a new FOLD_MAX_SIZE (=998, the smtp max line size) as the hard limit for any output. 2002-02-27 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-chomp.c (camel_mime_filter_chomp_new): New stream filter that chomps excess trailing whitespace from the end of the stream. This is needed to update the PGP/MIME code to comply with rfc3156. * camel-pgp-mime.c (camel_pgp_mime_part_verify): Don't attach a from filter, if it ain't from-filtered already, then we'll just be breaking stuff. To become rfc3156 compliant, add a chomp filter here. (camel_pgp_mime_part_sign): Add a chomp filter here too. 2002-02-27 Not Zed <NotZed@Ximian.com> * camel-mime-part.c (init_header_name_table): Changed header formatted table to contain a pointer to an output function, and added in-reply-to and references headers. (write_references): New function to write out references header, folded properly. It only approximates based on the last >, but it should be adequate and will also handle invalid headers. (write_fold): Function to write out headers folded. Since this is the default it isn't required. (write_raw): Write out an already formatted header, e.g. most of the rest. (write_to_stream): Lookup header output function, if we have one, use that, otherwise fold header using basic (dumb) function. This is all for #14779. A better fix is probably do have the headers always stored formatted, but that can wait. 2002-02-25 Jeffrey Stedfast <fejj@ximian.com> * camel-digest-folder.c (camel_digest_folder_new): Allow all multiparts that contain message/rfc822 attachments. (digest_get_uids): Only assign uids to message parts. 2002-02-22 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part.c (camel_mime_part_set_filename): Set the 'name' parameter on the Content-Type too. Fixes bug #20779. 2002-02-14 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_sign): Pass --always-trust to gpg (requested by users). (pgp_clearsign): Same. (pgp_encrypt): Here too. 2002-02-12 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_helo): Since the AUTH token sometimes uses '=' instead of whitespace, don't use smtp_token_next here. 2002-02-09 Not Zed <NotZed@Ximian.com> * providers/pop3/camel-pop3-engine.c (get_capabilities): Duh, when we grab the apop stamp it needs to include the <> as well, I even read the rfc, silly me. * providers/pop3/camel-pop3-store.c (pop3_try_authenticate): Doh, when iterating the authtype list, it helps to goto the next node. Found with help from miles. 2002-02-08 Not Zed <NotZed@Ximian.com> * providers/local/camel-spool-summary.c (camel_spool_summary_build_from): The day number has to be 2 chars wide, space filled to work properly with pine, etc. * providers/local/camel-spoold-store.[ch]: new type of provider 'spool directory', which lets you view external mbox dirs without adding any extra cruft. Perhaps it should use . files to store summaries? Still a bit experimental, there's a warning when you select it in the account editor. Finished off most of #1185. Can't rename or move folders. * camel-mime-utils.c (header_decode_date): If the date is 100->1900 then we actually want to use it as the year in the tm struct, not year+100. e.g. year 102 -> 2002, not 2102. 2002-02-07 Not Zed <NotZed@Ximian.com> * providers/local/camel-spool-store.c (get_folder): Pass path into spool_folder_new. * providers/local/camel-spool-folder.c (camel_spool_folder_new): (camel_spool_folder_construct): Take the full path to the folder and use that as the file path, independent of the full_name we use. 2002-02-07 Not Zed <NotZed@Ximian.com> * providers/local/camel-local-provider.c: Added new type, spoold provider, spoold: for local directories. * providers/imap/camel-imap-store.c (get_one_folder_offline): Create offline uri's in a compatible manner to online ones. 2002-02-07 Dan Winship <danw@ximian.com> * camel-sasl-ntlm.c: Implementation of NTLM (aka "Secure Password Authentication") auth, taken from soup. * Makefile.am (libcamel_la_SOURCES, libcamel_la_HEADERS): Add camel-sasl-ntlm. * camel-sasl.c: Add refs to camel-sasl-ntlm. * providers/imap/camel-imap-store.c (try_auth): Use imap_next_word() to skip over the "+ " of the continuation rather than just "resp + 2" since Exchange (incorrectly) returns "+" instead of "+ " for an empty continuation response. 2002-02-06 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.h: Remove the CAMEL_MESSAGE_NEEDS_REPLY flag, we no longer will be using this. 2002-02-04 Jeffrey Stedfast <fejj@ximian.com> * camel-http-stream.c (stream_read): Use camel_mime_parser_read to read internal parser data. (camel_http_stream_get_content_type): Implemented. (http_method_invoke): Use HTTP/1.0 instead of 1.1 * camel-mime-utils.c (header_decode_int): Made public. * camel-http-stream.[c,h]: Added. New stream for HTTP requests (currently supported are GET and HEAD). * camel-tcp-stream-ssl.c (stream_connect): Call set_errno appropriately. 2002-01-31 Not Zed <NotZed@Ximian.com> * camel-mime-utils.c (header_decode_domain): Oops, this was converting foo@[blah] to foo@[ blah ], fixed. 2002-01-30 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-provider.c: Use "pop" instead of "pop3" so current configurations continue to work. 2002-01-30 Not Zed <NotZed@Ximian.com> * camel-sasl-login.c: Changed name from "NT Login" to simply "Login". * providers/pop3/*: Entirely new pop implmentation, supporting pipelining. 2002-01-29 Not Zed <NotZed@Ximian.com> * camel-data-cache.c (free_busy): We dont want to unref the stream, instead, stop listening to the finalised events, and free the path only. 2002-01-25 Not Zed <NotZed@Ximian.com> * camel-data-cache.c (stream_finalised): Remove the object from the busy_stream hashtable, not the busy_path hashtable. 2002-01-29 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (imap_update_summary): Added more kludge to an existing Exchange IMAP 5.5 kludge to work around it returning multiple messages with the same UIDs. 2002-01-28 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-message.c (process_header): Handle Resent headers too. * camel-mime-message.h: Added Resent-* #defines. * camel-filter-driver.c (camel_filter_driver_remove_rule_by_name): Use while (node->next) instead of while (node) * providers/smtp/camel-smtp-transport.c (smtp_decode_status_code): New function to decode an enhanced status code. (smtp_set_exception): Sets an exception based on the Enhanced-Status-Code. (esmtp_get_authtypes): Don't diplicate the key in the hash since the key and value are the same. (smtp_rcpt): Include the failed recipient in the error message to be more helpful to the user. * camel-mime-utils.c (hex_decode): Make sure to allocate enough for the NUL byte. 2002-01-28 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_construct): (connect_to_server): Use flags rather than a bunch of gboolean variables. (smtp_connect): Same. (smtp_mail): Here too. Use the enhanced status codes if available. (smtp_data): And again here. (smtp_helo): Finally here. Also detect the ENHANCEDSTATUSCODES extension. (smtp_rcpt): Use the enhanced status codes if available. (smtp_rset): Here too. (smtp_quit): And finally here. * camel-transport.h: Removed gboolean supports_8bit since this is pretty local to only SMTP for now. 2002-01-24 Ettore Perazzoli <ettore@ximian.com> * Makefile.am: Remove some old cruft. 2002-01-24 Ettore Perazzoli <ettore@ximian.com> * tests/folder/Makefile.am: s/MAILER_LIBS/EVOLUTION_MAIL_LIBS/. * tests/message/Makefile.am: Likewise. * tests/mime-filter/Makefile.am: Likewise. * tests/misc/Makefile.am: Likewise. * tests/smime/Makefile.am: Likewise. * tests/stream/Makefile.am: Likewise. 2002-01-24 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (do_beep): Call the beep callback function. (camel_filter_driver_set_system_beep_func): New function to set the beep callback. 2002-01-22 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_remove_rule_by_name): New function to remove a filter rule by name. 2002-01-21 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (do_beep): As a temporary solution, just printf ("\a"); to make a beep :-) * providers/imap/camel-imap-command.c (imap_command_strdup_vprintf): Encode the mailbox to UTF-7 here. * providers/imap/camel-imap-utils.c (imap_parse_list_response): Decode the mailbox name as we parse the list response. (imap_mailbox_decode): It's only an illegal mailbox name if it didn't switch back to US-ASCII mode. 2002-01-18 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-utils.c (imap_mailbox_decode): New function to decode an IMAP mailbox name from modified UTF-7 encoding to UTF-8. (imap_mailbox_encode): New function to convert a mailbox name from UTF-8 to IMAP's modified UTF-7 encoding. 2002-01-17 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-basic.c (filter): Stop uudecoding once the CAMEL_UUDECODE_STATE_END state bit gets set. Set the CAMEL_UUDECODE_STATE_BEGIN state bit once we find the begin line. (reset): No longer have uu_begin or uulen state variables, these are now stuffed into a single state variable. * camel-mime-utils.c (uudecode_step): No longer needs a uulen argument and also keeps track of whether or not the end of the encoded data has been found in 'state'. (uuencode_step): Now stuffs uulen into state so that the uulen argument is no longer needed. (uuencode_close): Same. 2002-01-16 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-basic.c (filter): If we don't want to corrupt the uuencoded data by overwriting it with base64 decoded data afterward, we need to add a break statement! * camel-folder-summary.c (summary_build_content_info): Add code to add a uu filter. (camel_folder_summary_finalize): Unref the uuencode filter. 2002-01-16 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-basic.c (filter): Fixed slight logic error to find the uuencode begin line. Fixes bug #18754. 2002-01-15 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_send_to): Updated to match the new send_to API. (smtp_send): Get the from address and pass that along to smtp_send_to(). * providers/sendmail/camel-sendmail-transport.c (sendmail_send_to): Updated to match the new send_to API. * camel-transport.c (camel_transport_send_to): Now takes a from argument too. * providers/imap/camel-imap-folder.c (imap_update_summary): Sort the needheaders UID array and fixed to respect the UID_SET_LIMIT. This should now finish the fixification of bug #2529. There's still the possible issue that a command-line (The only command-line I can think of that can still be too long is a SEARCH command, but this can't possibly be fixed until we rewrite the imap code to use Zucchi's ImapEngine idea). 2002-01-14 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (imap_expunge_uids_online): Move the command-lock outside the loop. 2002-01-14 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (imap_expunge_uids_online): Updated to use the new imap_uid_array_to_set() interface. (imap_expunge_uids_resyncing): Same. (do_copy): Here too. (imap_update_summary): Added a FIXME comment to rewrite allowing for a uid-set limitation. (get_matching): Copy some of the logic over from imap_uid_adday_to_set() to limit the length of the uid-set string. (imap_sync_online): Added a comment to explain what is going on with get_matching() since the behavior has changed slightly. * providers/imap/camel-imap-utils.c (imap_uid_array_to_set): Modify the interface so that we can limit the size of the uid set string returned. 2002-01-14 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-search.c (imap_body_contains): Rewritten to use a cache for body searches when online. Will need some heavy testing but so far seems to be beneficial. * providers/imap/camel-imap-folder.c (imap_search_by_expression, search_by_uids): dont initialise search object here. (camel_imap_folder_new): Setup search object here with pointer to cache dir. 2001-12-01 Not Zed <NotZed@Ximian.com> * camel-store-summary.[ch]: New class to store a store's folder list in. Not yet completed. 2002-01-11 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (imap_update_summary): Kludge around a bug in Exchange 5.5 that reports 2 messages with the same UID. Fixes bug #17694. Replaces the fix from yesterday. 2002-01-10 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-local-folder.c: If PATH_MAX doesn't exist, use _POSIX_PATH_MAX. * providers/imap/camel-imap-folder.c (imap_update_summary): If mi is NULL, don't bother updating it. Should fix bug #17694. 2002-01-10 Jeffrey Stedfast <fejj@ximian.com> * camel.h: #include camel-mime-filter-tohtml.h * providers/imap/camel-imap-folder.c (imap_update_summary): Kludge around Microsoft Exchange 5.5 (bug #5348) by forgetting our currently selected folder and re-SELECTing it so that the Exchange server has a chance to realise it has new messages. 2002-01-09 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-local-folder.c (camel_local_folder_construct): If the mbox file is a symlink, follow the symlink and get the One True Path so that we can rewrite the mbox later without worrying about clobbering the symlink. 2002-01-08 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-search.c (TODO): There are a few sexp callbacks that could be modified to use fms->info rather than using a message object (like date and possibly mlist stuff) but *only* if the date exists on the CamelMessageInfo object (since it may be blank except for message flags). (camel_filter_search_get_message): New internal convenience function to make sure that the FilterMessageSearch has loaded the message (and to load the message if this isn't the case). (check_header): Call camel_filter_search_get_message(). (header_exists): Same. (header_regex): Here too. (header_full_regex): And here. (body_contains): Again here. (body_regex): Here too. (get_sent_date): Here also. (get_received_date): Same. (get_source): Here if we need to. (camel_filter_search_match): Now takes a callback function/data pair for on-demand message loading so that we don't necessarily have to load the message if the defined filter rules don't require it. * camel-filter-driver.c (camel_filter_driver_filter_folder): Don't bother fetching the message here, let camel_filter_driver_filter_message() worry about this. (get_message_cb): New utility callback to fetch a message. (camel_filter_driver_filter_message): Only fetch the message if we absolutely need it to get a CamelMessageInfo. Instead of passing a message object to camel_filter_search_match(), pass get_message_cb and some user_data so that the matching code can fetch the message on demand. 2002-01-07 Jeffrey Stedfast <fejj@ximian.com> * camel-folder.c (filter_filter): Flush the only-once actions. * camel-filter-driver.c (camel_filter_driver_filter_message): Don't increment a filtered_count here any longer. (camel_filter_driver_reset_filtered_count): Removed. (camel_filter_driver_get_filtered_count): Removed. (do_beep): New action. (play_sound): New action to play a sound (do_only_once): Another new action. (camel_filter_driver_finalise): Free the only_once hash if the driver has not been "flushed". (camel_filter_driver_flush): Flush all of the only-once actions. * camel-charset-map.c: Moved windows-1251 to the end of the list since it contains the euro and we'd prefer to use iso-8859-15 when the euro is requested than a windows charset if possible. * camel-charset-map-private.h: Regenerated. 2001-12-12 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.c (content_info_load): Don't try setting a content-type parameter if either the name or value is NULL. * camel-mime-utils.c (header_set_param): NULL-protection. 2002-01-02 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (ssl_bad_cert): Do more like what mutt does so hopefully this'll fix bug #16363 and #16300. 2001-12-21 Jeffrey Stedfast <fejj@ximian.com> * broken-date-parser.c (parse_broken_date): Completely rewritten. It is now a load faster and a heck of a lot more accurate, also now returns a time_t and sets the saveoffset variable rather than returning a new char* buffer for the normal camel date parser to re-parse. This saves a fair number of cpu cycles :-) * camel-mime-utils.c (header_decode_date): Cleanup the broken date parsing code. 2001-12-20 Jeffrey Stedfast <fejj@ximian.com> * camel-address.h: Change the prototype for camel_address_get_type to return a CamelType (since internally this is what it returns and also in case we decide to write a replacement for the current CamelObject it'd be easier to drop in). * camel-internet-address.h: Same but for camel_internet_address_get_type() * providers/smtp/camel-smtp-transport.c (smtp_send_to): Updated to use a CamelAddress of recipients. (smtp_send): Since smtp_send_to now takes a CamelAddress recipients argument, our lives have been simplified and we can now just concat To/Cc/Bcc into a recipients addr and send away. * providers/sendmail/camel-sendmail-transport.c (sendmail_send_to): Updated to use a CamelAddress of recipients. * camel-transport.c (camel_transport_send_to): Now takes a CamelAddress argument for the recipient list rather than a GList. 2001-12-19 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/Makefile.am: Remove the providerdir variable. * providers/sendmail/Makefile.am: Same. 2001-12-17 Jeffrey Stedfast <fejj@ximian.com> * camel-charset-map.c (camel_charset_iso_to_windows): New function to map ISO charsets to the Windows charsets. * camel-mime-part-utils.c (broken_windows_charset): Detect Windows charsets. (simple_data_wrapper_construct_from_parser): Simplify a tad and also check for iso-8859-* charsets that are really Windows charsets. Fixes bug #12631. 2001-12-17 Dan Winship <danw@ximian.com> * Makefile.am (INCLUDES): define CAMEL_PROVIDERDIR to be the configure.in-defined camel_providerdir. * providers/Makefile.am: Remove some outdated comments * providers/imap/Makefile.am (camel_provider_LTLIBRARIES, camel_provider_DATA): renamed from provider_LTLIBRARIES, provider_DATA. * providers/local/Makefile.am: Likewise * providers/nntp/Makefile.am: Likewise * providers/pop3/Makefile.am: Likewise * providers/sendmail/Makefile.am: Likewise * providers/smtp/Makefile.am: Likewise 2001-12-16 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (get_content): Reverted my previous changes here since it doesn't actually work afterall. * providers/imap/camel-imap-wrapper.c (imap_wrapper_hydrate): Update to do uudecoding when appropriate. 2001-12-16 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (get_content): Try to use the original boundary so luis will stop bugging me about "data corruption". Also preserve other params in the multipart content-type by dumping it to a string and setting it on the mime part. 2001-12-14 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_reset_filtered_count): Reset the filtered_count to zero. (camel_filter_driver_get_filtered_count): Return the private filtered_count value. (camel_filter_driver_filter_message): Increment the filtered_count. (camel_filter_driver_set_shell_exec_func): New function to set the shell-exec func. (shell_exec): New ESExp filter action callback. 2001-12-14 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-mime.c (camel_pgp_mime_part_verify): Removed x-inline-pgp-hack kludge because it doesn't work. (camel_pgp_mime_part_decrypt): Same. 2001-12-13 Chris Toshok <toshok@ximian.com> * camel-data-cache.c: include stdlib.h (for alloca on freebsd) and only include alloca.h if HAVE_ALLOCA_H is defined. 2001-12-11 Zbigniew Chyla <cyba@gnome.pl> Fixes #17085 * camel-charset-map.c: Added #include <errno.h>. * camel-charset-map-private.h: Recreated (following instructions from camel-charset-map.c) 2001-12-11 Jeffrey Stedfast <fejj@ximian.com> * camel-object.c (camel_type_register): Keep a name-to-type hash so that we can make sure that the type has not yet been registered (prevents a race condition such as the one in bug #16559). * camel-service.c (camel_service_connect): Make sure that the connect_op is non-NULL before unregistering/unreffing it. 2001-12-04 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_content_type_simple): Protect against either of the types being NULL. 2001-12-05 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-basic.c (filter): If complete() allocates len+2 bytes for the out buffer, so should this. See bug #16371 for an example case. 2001-12-05 Jeffrey Stedfast <fejj@ximian.com> * camel-sasl-digest-md5.c: iconv() returns a size_t, not an int. * camel-pgp-context.c: The return value of iconv() is a size_t, not an int. * camel-mime-part-utils.c (convert_buffer): Always use size_t args for iconv(). * camel-mime-filter-charset.c (complete): Always use size_t args for iconv(). (filter): Same. * camel-mime-utils.c (header_address_fold): Make headerlen a size_t instead of an int. (header_fold): Same. (base64_encode_close): We should be returning a size_t and inlen should also be a size_t. (base64_encode_step): Same here. (base64_decode_step): Here too. (base64_encode_simple): And here... (base64_decode_simple): Same. (uuencode_close): We should also use size_t's here... (uuencode_step): And here too. (uudecode_step): And also here. (quoted_encode_close): Same idea here. (quoted_encode_step): Again here. (quoted_decode_step): Here too. (quoted_encode): Input length should be a size_t. (rfc2047_decode_word): Same. (g_string_append_len): Here too. (append_8bit): " (rfc2047_encode_word): " (quote_word): " (hex_decode): " (rfc2184_decode): Use size_t's with iconv(). (header_decode_param): Same. 2001-12-09 Jon Trowbridge <trow@ximian.com> * camel-folder-summary.c: Add "NeedsReply" to the flag_names array for CAMEL_MESSAGE_NEEDS_REPLY. * camel-folder-summary.h: Added CAMEL_MESSAGE_NEEDS_REPLY flag. 2001-12-07 Dan Winship <danw@ximian.com> * camel-mime-message.c (camel_mime_message_set_date): Fix the tm_gmtoff case (its sign is the opposite of "timezone"). Fixes #14678 2001-11-29 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-search.c (search_body_contains): Don't use regex matching. Fixes bug #16227. * camel-mime-message.c (best_encoding): Check the content-object's mime type, not the mime part types. Should fix bug #15843. 2001-11-27 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (get_content): Return NULL if construct_from_stream fails. (get_message): Same. (get_message_simple): Here too. (add_message_from_data): And here. 2001-11-14 <NotZed@Ximian.com> * camel-vee-folder.c (folder_changed_remove_uid): Use the uid rather than vuid for unmatched. Also add the uid to unmatched if it wasn't in the unmatched_uids table at all. (folder_changed_change): If checking for added, if the new ones dont match, then try to add them to unmatched. Fixes #6893. * camel-folder-summary.c (camel_folder_summary_index): Change lock order, always summary_lock before ref_lock. (camel_folder_summary_array): " (camel_folder_summary_uid): " (camel_folder_summary_remove_uid): " Fixes a deadlock. 2001-11-30 Not Zed <NotZed@Ximian.com> * providers/nntp/camel-nntp-*.c: Completely new implementation of NNTP. Doesn't support subscriptions yet (lists all folders), but should be more reliable (faster?), and has an integrated cache. * camel-exception.c (camel_exception_new): Use e_memchunks for exception blocks. (camel_exception_free): Same. * camel-data-cache.[ch]: New object for managing on-disk caches of anything that can be stored in a camel-stream. * camel-file-utils.c (camel_file_util_mkdir): New function, just a nicer place to put this (than camel-store), should be removed from camel-store. (camel_file_util_safe_filename): New function to url-encode a filename. * camel-mime-parser.c (drop_states): New func to drop the parser state to initial state. (folder_scan_init_with_fd): (folder_scan_init_with_stream): Call above func to reset state if the stream is changed on us so we can change streams to reuse a parser object. 2001-11-25 Not Zed <NotZed@Ximian.com> * providers/nntp/camel-nntp-folder.c (nntp_folder_get_message): If the uid doesn't have a ',' in it, fail to crash. * providers/nntp/camel-nntp-newsrc.c (camel_nntp_newsrc_article_is_read): check group != NULL before scanning. (camel_nntp_newsrc_get_highest_article_read): " (camel_nntp_newsrc_get_num_articles_read): " (camel_nntp_newsrc_mark_range_read): " * providers/nntp/camel-nntp-store.c (camel_nntp_store_get_overview_fmt): IF we dont have nntp_list_follows, dont try and get a list response. (nntp_store_get_folder_info): Set path part of folderinfo. 2001-11-20 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (get_smtp_error_string): Check to see that errno is non-zero before returning g_strerror. If it's 0, then we have an unknown error. 2001-11-26 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-basic.c: For the uudecoding mode, garble up the "begin <mode> <filename>" line before decoding. * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): Add a uudecoder if the transfer encoding is x-uuencode. * camel-mime-part.c (write_to_stream): Handle x-uuencoded content too. 2001-11-19 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (stream_read): Added a check to see if the operation has been cancelled. (stream_write): Same. 2001-11-19 Jeffrey Stedfast <fejj@ximian.com> Updates for compliance with rfc2231 * camel-mime-utils.c (header_encode_param): camel_mime_special_table[c] & IS_ESAFE should have been !(camel_mime_special_table[c] & IS_ESAFE). Also added a few comments for how to improve the code at some future date. (header_decode_param): Now takes an argument rfc2184_part so our caller can get this information as well. (header_decode_param_list): Pass an rfc2184_part argument to header_decode_param and also added a few comments on where to improve on rfc2184/rfc2231 compliance. (rfc2047_decode_word): Updated to respect the updated ABNF syntax of rfc2047 encoded words, yay. 2001-11-16 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-utils.c (imap_parse_body): Abort if body == NULL. * camel-mime-filter-tohtml.c (camel_mime_filter_tohtml_new): New mime filter to convert plain text to html. 2001-11-16 Jeffrey Stedfast <fejj@ximian.com> Since some mail clients like Outlook are broken, we need to set a name parameter on pgp signed and encrypted parts. * camel-pgp-mime.c (camel_pgp_mime_part_encrypt): Set a name parameter. (camel_pgp_mime_part_sign): Same. 2001-11-14 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-command.c (camel_imap_command_response): If we get a BYE response, call camel_service_disconnect() and set an exception. Also do the check for "* BYE" first instead of passing it off imap_read_untagged() since we'll just waste time in there mallocing left and right only to arrive at the single response line "* BYE" again :-) 2001-11-13 <NotZed@Ximian.com> * camel-filter-search.c (get_source): If we have no source string, then use undefined, which should always evaluate to FALSE. Fix for #15267. 2001-11-12 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (camel_store_get_folder): Added a g_return_val_if_fail for folder_name != NULL. 2001-11-08 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (crypto_exec_with_passwd): If any of the pipe()'s fail, clean up any pipes that may have succeeded. Also close the password fds. 2001-11-09 <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (imap_auth_loop): If we get a user cancel and that causes us to fail to authenticate, abort rather than loop forever. Fix for #14951. 2001-11-06 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-command.c (camel_imap_command_continuation): Now takes a command-length argument so we can 1) avoid duping the command string yet again, yay. 2) we now don't have to worry about embedded nul-chars screwing us over (we still need to avoid allowing them into the string but at least now it won't mess us up). * providers/imap/camel-imap-folder.c (do_append): Instead of appending a nul char to the end of the byte array and then passing that off as if it were a string to camel_imap_command_continuation, instead pass the byte-array length since that function now takes a length argument. Yay. Also encode any 8bit parts to avoid the possibility of sending embedded nul chars to the imap server. * providers/imap/camel-imap-store.c (try_auth): Updated to pass a command-length argument to camel_imap_command_continuation(). 2001-11-07 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (save_ssl_cert): Oops, pass a mode argument to the open() call. * camel-tcp-stream-openssl.c (save_ssl_cert): Oops, pass a mode argument to the open() call. 2001-11-07 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-mime.c (camel_pgp_mime_part_decrypt): Add code to kludge around any x-inline-pgp-hacks by prepending some content headers to the decrypted data. 2001-11-02 Jeffrey Stedfast <fejj@ximian.com> * camel-data-wrapper.c (camel_data_wrapper_init): Initialize our private mutex. (camel_data_wrapper_finalize): Destroy our private mutex. (write_to_stream): Lock around camel_stream_write_to_stream() so that we don't get 2 threads trying to read from our stream at the same time. 2001-11-01 <NotZed@Ximian.com> * camel-vee-folder.c (folder_changed_change): Move all searching operations outside of any locks except subfolder lock, same as build_folder. Fix for #14294. 2001-10-31 <NotZed@Ximian.com> * providers/local/camel-spool-folder.c (spool_get_message): Same as below. * providers/local/camel-maildir-folder.c (maildir_get_message): Same as below. * providers/local/camel-mbox-folder.c (mbox_get_message): Set USER_CANCEL if failed due to EINTR. * camel-filter-driver.c (camel_filter_driver_filter_mbox): If construct from parser fails due to user cancel, set USER_CANCEL on exception. * camel-mime-part.c (construct_from_parser): Return error if the parser had an io error. * camel-mime-message.c (construct_from_parser): Check error on parser/return error. * camel-mime-parser.c (folder_scan_init): Init error number. (camel_mime_parser_errno): New function, return errno of any io failures. (folder_read): Set errno if a failure occured. (folder_seek): Same. (folder_scan_init_with_fd): Setup errno depeding on ok/failure. (folder_scan_init_with_stream): Same. 2001-10-30 <NotZed@Ximian.com> * camel-operation.c (camel_operation_new): Dont setup cancel_fd. (camel_operation_cancel_fd): If cancel_fd not created, set it up. (camel_operation_cancel): Set cancel flag before sending message. (cancel_thread): Same. * camel-filter-driver.c (camel_filter_driver_filter_mbox): Make sure we unref the mimeparse when we're done (successfully). This was leaking an fd every get-mail! :( * camel-lock-client.c (camel_lock_helper_init): Close all fd's above 2. * providers/local/camel-mbox-folder.c (mbox_get_message): Remove the X-Evolution header before anyone else gets to see this private thing. Shoudl fix 11676 since we were inheriting flags from X-Evolution headers we didn't want to. * camel-folder-search.c (match_message): If we can't retrieve the message, ignore it, it can't match. * providers/imap/camel-imap-search.c (camel_imap_search_class_init): Setup parent class pointer. (imap_body_contains): If offline, just use the parent body_contains method which will get the messages and search them manually. * providers/imap/camel-imap-folder.c (imap_search_by_expression, imap_search_by_uids): Remove offline check. 2001-10-30 Dan Winship <danw@ximian.com> * camel-mime-part.c (free_headers): Fix this to match get_headers. 2001-10-30 <NotZed@Ximian.com> * providers/local/camel-local-store.c (rename_folder): Dont try to move ibex if we have none. * camel-tcp-stream-openssl.c (stream_write): Fixed compilation errors. 2001-10-29 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (camel_mime_utils_init): Removed unused variable. 2001-10-30 <NotZed@Ximian.com> * camel-mime-utils.c (header_decode_mailbox): w(x) out some warnings. * providers/local/camel-local-summary.c (camel_local_summary_load): Remove the warning about not loading summary file - its a valid case for new folders. 2001-10-29 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-openssl.c (my_SSL_write): Removed. (stream_write): Keep looping (non-blocking case) if errno is EAGAIN, EINTR or EWOULDBLOCK. For NONBLOCKing I/O, sync up with CamelTcpStreamRaw. As with CamelTcpStreamRaw/SSL - make sure to write out everything before returning. (my_SSL_read): Removed. (stream_read): Just call ssl_error_to_errno() and check the errno values that we care about so we can keep the general look of all this stream code the same. Also when checking the return value of SSL_read, check for <0 instead of ==-1 since the man page for SSL_read doesn't say it will return -1 on fail, it just says <0. (stream_flush): Don't fsync() since syncing on a socket is a Bad Thing (tm). * camel-tcp-stream-ssl.c (stream_write): Make sure we write out everything just like in camel-tcp-stream-raw.c. * camel-stream-buffer.c (camel_stream_buffer_gets): If camel_stream_read() returns -1, don't necessarily return -1 to our caller since it's possible that we did actually "read" some data (ie, we copied some pre-buffered data into the out buffer). * camel-stream-buffer.h: Removed CAMEL_STREAM_BUFFER_NEWLINE since it never got used anywhere and it isn't supported anyway. 2001-10-30 <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (parse_list_response_as_folder_info): If we get an empty name, then mark that as NoSelect. Workaround fix for #13239. 2001-10-29 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (get_message_simple, imap_get_message): Set an X-Evolution-Source header on the returned message so replies come from the right identity. 2001-10-29 <NotZed@Ximian.com> * camel-vee-folder.c (folder_changed_change): If we're not going to search on a changed uid, make sure we change out copy still. Fixes #13916. 2001-10-29 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (subscribe_folder): Don't emit a folder_subscribed signal if we are in the process of renaming folders. (unsubscribe_folder): Same here but for the unsubscribe signal. (rename_folder): Unsubscribe from the folder being renamed and any subfolders it may have before actually renaming. Once the folder has been renamed, re-subscribe to it and it's subfolders with the new name. * camel-store.c (camel_folder_info_build): Sort the folder info's before constructing the tree. * camel-mime-utils.c (header_decode_param): Minor fix - should have been using a strNcmp when checking if it was an rfc2047 encoded word. * camel-pgp-context.c (pgp_sign): Notify the user that PGP 2.6.x is no longer supported and use camel_exception_set() instead of setv() where appropriate. (pgp_clearsign): Same. (pgp_verify): And again here. (pgp_encrypt): Here too. (pgp_decrypt): And finally here. 2001-10-29 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.c (unsubscribe_folder): Don't free the hash table key if the folder wasn't found in the hash table. I'm not sure where the bug is/was that made it possible to add a folder to the tree after creating it failed, but now if that happens, it won't crash if you try to delete it again. (#11492) * camel-disco-store.c (camel_disco_store_status): if the service is being connected and thinks it's online, but the session is offline, then mark the service offline as well. Fixes 13683. 2001-10-29 <NotZed@Ximian.com> * camel-mime-utils.c: Turn off w() -> warnings. * providers/imap/camel-imap-store.c (rename_folder): Also rename the message cache. (rename_folder): Fix subscribed folders table with rename(s). * providers/imap/camel-imap-folder.c (imap_rename): Implement, rename the cache dir pointer. * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_set_path): New method to set the path used by a message cache. * camel-vee-store.c (build_info): Fix the 'is this a matching subfolder or subfolder' logic to match camel-store get_subfolders'. * camel-store.c (get_subfolders): Fix the 'is this a matching folder or subfolder' logic. 2001-10-28 <NotZed@Ximian.com> * providers/local/camel-local-store.c (rename_folder): Changed to call ibex_move to rename it internally. * camel-store.c (camel_store_rename_folder): Use the right variable, not info. 2001-10-28 Jeffrey Stedfast <fejj@ximian.com> * camel-stream-buffer.c (camel_stream_buffer_read_line): Make sure we've actually read data before checking if p[-] is '\r' or not. * camel-tcp-stream-raw.c (stream_write): Same. * camel-stream-fs.c (stream_write): If errno isn't EAGAIN, just exit with -1 and instead of using continue (which doesn't go to the top of the do-while thus making our logic suck), just set w to 0. Still seems to be a bug in select() if it's telling us the socket is ready for data when it really isn't but oh well. 2001-10-28 <NotZed@Ximian.com> * providers/local/camel-local-folder.c (camel_local_folder_construct): Use path not protocol as the path part of the uri of the folder_created event. * providers/local/camel-maildir-store.c (scan_dir): If FAST is set, dont try and scan for unread counts. * providers/local/camel-local-store.c (create_folder): Implement, just return the folderinfo of the folder matched, not all of it. * camel-store.c (camel_store_rename_folder): Rename the camelfolders before emitting the folderchanged event (otherwise vstore fails to pick up change. 2001-10-27 <NotZed@Ximian.com> * camel-store.c (camel_store_init): Always use a default dir_sep of '/'. * camel-store.h (struct _CamelStore): Added a dir_sep to the base store so rename and whatnot can work, temporary fix for mixed dir separator in folder name api's. * providers/imap/camel-imap-store.c (rename_folder): Finished implementation. (imap_connect_online): Also set the store's dir_sep here. (imap_connect_offline): " * providers/local/camel-local-folder.c (local_rename): Implement local rename of folder objects. 2001-10-26 <NotZed@Ximian.com> * camel-vee-folder.c (camel_vee_folder_class_init): Hook into rename function. (vee_rename): Implement the veefolder rename function. (camel_vee_folder_class_init): (folder_changed_change): Kill a warning with a cast. * camel-vee-store.c (vee_rename_folder): Emit a folder_renamed event properly, also call parent to do some rename stuff. * camel-store.h: Added a CamelRenameInfo for the rename event. * camel-folder.c (camel_folder_rename): New function to rename a folder object. (camel_folder_class_init): Added a renamed event. (folder_rename): Default impl, set full_name, and set name assuming the dir separator is '/'. * camel-store.c (camel_store_class_init): Added folder_renamed event. (camel_store_rename_folder): Rename an active folder object if we have one, and update the folder table. (rename_folder): Make a default implementation that handles updating the folder tree. 2001-10-26 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (delete_folder): Don't bother checking ex since we can rely on the response being NULL on error. (rename_folder): Implemented. 2001-10-26 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (imap_get_message): Check to make sure that the dataset is non-NULL before using and/or freeing. * camel-tcp-stream-raw.c (stream_write): Check for EWOULDBLOCK too? (stream_read): Same. (stream_flush): Return 0 always, fsync() and tcp sockets don't play nice :-) * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_get): Lets try putting a camel_stream_reset() here. Logic being that the stream may have been read from since it was cached and thus our caller may try reading from it and get no data from it since it is already at the EOS. This may fix bug #12943. 2001-10-25 <NotZed@Ximian.com> * camel-search-private.c (utf8_get): Remove the stupid warning that should never have been there in the first place. * camel-sasl-digest-md5.c (digest_response): s/iconv/e_iconv/ * camel-pgp-context.c (pgp_verify): " * camel-mime-utils.c (rfc2047_decode_word, rfc2047_decode_word, append_8bit, rfc2047_encode_word, rfc2184_decode, header_decode_param): " * camel-mime-part-utils.c (convert_buffer, convert_buffer): " * camel-mime-filter-charset.c (reset, complete, filter): " 2001-10-24 <NotZed@Ximian.com> * camel-mime-filter-basic.c (complete): For qp decoding, if the data isn't really qp encoded, we could possible grow the buffer by upto 2 bytes above the input size, fix allocations/assertions appropraitely. * camel-vee-folder.c (folder_changed_change): If we're not autoupdate, only search for new matches against changed uid's that we dont already have. 2001-10-24 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.c (unsubscribe_folder): Don't let fi->name be NULL. 2001-10-24 Jeffrey Stedfast <fejj@ximian.com> * camel-stream-fs.c (stream_read): Same as in the TcpStreamRaw code. (stream_write): And again here... * camel-tcp-stream-raw.c (stream_read): Handle the EAGAIN error case as well. (stream_write): Same here, this might fix the SMTP truncation thing? I hope? 2001-10-23 Dan Winship <danw@ximian.com> * camel-mime-utils.c (mail_list_magic): Remove an extra * in one of the regexps that glibc apparently doesn't mind but bsd does. 2001-10-23 <NotZed@Ximian.com> * providers/local/camel-mbox-summary.c: made d(x) x recompile again. 2001-10-22 <NotZed@Ximian.com> * providers/local/camel-spool-summary.c (summary_rebuild): No, use Storing, as the other code does :p (spool_summary_check): Check for consistency. 2001-10-22 Jon Trowbridge <trow@ximian.com> * providers/local/camel-spool-summary.c (summary_rebuild): s/summarising/summarizing/. (spool_summary_check): s/summarise/summarize/. 2001-10-22 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-mime.c (camel_pgp_mime_part_verify): If we are kludging around a inline-pgp signed part, do some charset conversion to protect any 8bit text. 2001-10-22 <NotZed@Ximian.com> * camel-filter-search.c, camel-folder-search.c (check_header): Use search_type_mlist for mailing list searches. * camel.c (camel_init): call camel-mime-utils-init func. * camel-mime-utils.c: Changed mail mail_list_magic to include a domain part, also pre-compile all the patterns. They are all backward compatible except List-Id: which now uses the mail-address-like <list-name.host.name> part rather than the plain-text part (which might've been blank anyway). (camel_mime_utils_init): Initialisation function to setup any static data required by camel-mime-utils at run-time. We now setup the base64/charset class tables here, so it doesn't need to be done statically. (camel_mime_special_table, camel_mime_base64_rank): No longer statically initialise these. (main): Removed + the tests at the end of the file. (header_raw_check_mailing_list): Dont compile regex's here, already compiled in utils_init. Use the regex patterns to remove leading <'s from addresses. Also, if there is a domain part matched, add that after a '@' is added. * camel-search-private.c (camel_search_header_match): Match check against all addresses in a multiple address match. Should fix #13051. (camel_search_header_match): Oops, missing i=0. (camel_search_header_match): Added SEARCH_TYPE_MLIST for mlist search types. It ignores the domain part if either (but not both) of the values dont have it. 2001-10-22 Dan Winship <danw@ximian.com> * camel-pgp-mime.c (camel_pgp_mime_part_verify): Add a hack: if the multipart/signed part has a "x-inline-pgp-hack" parameter in its Content-Type, don't pass the MIME headers as part of the data to be verified. 2001-10-22 Rodrigo Moya <rodrigo@ximian.com> * Makefile.am: added $GNOME_INCLUDEDIR to INCLUDES, needed for latest libgnome, which install headers in $(prefix)/gnome-1.0 2001-10-21 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): Only do the rawtext checks if the part is a text part, otherwise don't bother wasting cpu cycles. * camel-store.c (camel_folder_info_build): Updated comments to make the code a bit more clear. 2001-10-21 <NotZed@Ximian.com> * camel-store.c (CS_CLASS): Dont do the macro stuff - basically so you can actually debug calls. (camel_store_uri_cmp): New function to compare store objects. * camel-vee-folder.c (vee_folder_add_info): oops, dont free the vuid, since its alloca'd now. (folder_changed_change): Kill a compile warning. 2001-10-20 <NotZed@Ximian.com> * camel-folder.c (move_messages_to): If no exception supplied, use a local one. (move_message_to): Set the seen flag also when we delete it. 2001-10-19 Jeffrey Stedfast <fejj@ximian.com> * camel-vtrash-folder.c (vtrash_move_messages_to): If the destination folder is the vfolder source folder, then set_message_flags on it. 2001-10-19 <NotZed@Ximian.com> * camel-session.c (register_provider): When registering provider, translate all strings. * camel-vee-folder.c (camel_vee_folder_remove_folder): Lock the right lock for unmatched subfolder list stuff. If the sub folder is deleted & part of unmatched, or the last ref to the sub folder is gone from unmatched, blow it away completely from unmatched, but only do it once. (vee_folder_remove_folder): Added arg to kill unmatched unconditionally. Also handle deleted folders, always remove stuff. (folder_changed_change): Lock subfolder lock for entirety of operation, and also check to see if the subfolder is still there before doing any fancy work (incase it got removed while we were waiting to execute). (folder_changed_change_uid): Use alloca for vuid, and not sprintf(".lengths") (vee_search_by_expression): Dont use sprintf for vuid. (vee_search_by_uids): " (vee_folder_add_info): " (folder_changed_remove_uid): " (folder_changed_change): " (folder_changed_change_uid): Also check folder_unmatched for changed uid's so we properly update/propagate unmatched changes. (folder_changed_change): Ok, so dont hold the subfolder lock for the whole duration of the changed event, although we probably should, requires a recursive mutex. (folder_changed_change): Change logic, we always add changed stuff if it now matches, but dont remove it unless its auto-remove, only propagate changes for it. (vee_set_message_flags): Call parent method after doing our work. (vee_set_message_user_flag): Same here. 2001-10-18 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): If a charset isn't specified or claims to be utf-8, check the validity of the text and if it's invalid, set the rawtext bit to TRUE. If the charset is x-unknown or some other x- charset, always set the rawtext bit to TRUE. 2001-10-18 <NotZed@Ximian.com> * camel-store.c (camel_store_unsubscribe_folder): If we are unsubscribing from a folder we have loaded, set the folder deleted, as with the delete_folder case. * providers/(imap|local|pop3|sendmail|smtp)/Makefile.am: Added CAMEL_CFLAGS so we get the right iconv stuff, patch from Yanko Kaneti <yaneti@declera.com> 2001-10-18 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (crypto_exec_with_passwd): Make sure the cancel_fd isn't -1. 2001-10-18 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-message-cache.c (insert_finish): Flush the stream here, not sure it really matters but it might (hopefully) fix bug #12943. * camel-pgp-context.c (crypto_exec_with_passwd): Make user-cancellable. * providers/local/camel-local-provider.c: String cleanup for bug #6640. * providers/imap/camel-imap-folder.c (imap_sync_online): Fix my Cyrus imapd workaround...instead of checking to see if info->flags == FLAGGED, just check to see if it contains any IMAP flags. 2001-10-17 <NotZed@Ximian.com> * providers/local/camel-maildir-folder.c (camel_maildir_folder_new): Use '.' as the inbox name for filter new messages test. * camel-store.c (camel_folder_info_clone): New function to clone a folderinfo tree. 2001-10-17 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-mh-folder.c (mh_append_message): Same as maildir and mbox append-message. * providers/local/camel-maildir-folder.c (maildir_append_message): Same as mbox append. * camel-exception.c (camel_exception_set): Make sure the new description isn't the same as the old description pointer before freeing the old one. (camel_exception_setv): Don't free the old description until we set the new one this way we can reuse the old description in the new description. * providers/local/camel-mbox-folder.c (mbox_append_message): If errno == EINTR, then we got a user-cancel so set the exception appropriately so that we don't make the user shit his pants. * camel-filter-driver.c (do_copy): Add an optimization for when destination folder is the same as the source folder. 2001-10-17 <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (get_folder_info_online): (parse_list_response_as_folder_info): Patch from Danw, Use unread count of -1 to mark unflagged messages. Make sure any folder we dont lookup explicitly is marked as -1. Should fix #9947 and friends. * providers/local/camel-mbox-summary.c (mbox_summary_sync): Only touch the summary if the timestamp or size changed. 2001-10-17 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_message): Set the DELETED flag on directly on the info if the source folder doesn't have summary capability. 2001-10-17 Not Zed <NotZed@Ximian.com> * camel-tcp-stream-raw.c (stream_write): As below. * camel-stream-fs.c (stream_write): If not checking cancellation, make sure we write out everything asked of us. 2001-10-16 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_verify): Rearange the order of the gpg options slightly otherwise it might try to use "--no-tty" as the filename to verify :-) * camel-object.[c,h]: If CAMEL_DEBUG is defined, print some useful ref/unref info. * providers/imap/camel-imap-store.c (delete_folder): Fixed an assignment warning. * camel-uid-cache.c (camel_uid_cache_new): Make sure that the parent directory exists before trying to open the filename, if it doesn't, create it. 2001-10-16 <NotZed@Ximian.com> * camel-mime-utils.c (header_address_decode): If no content, dont try and decode further and spit meaningless warnings. 2001-10-16 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (do_flag): Only perform the operation on the folder if the folder has summary capability, else just poke the info directly. Fixes the "Filters don't work" bug. (do_score): Same. (do_colour): And here too. 2001-10-15 Jeffrey Stedfast <fejj@ximian.com> * camel-stream-fs.c (stream_read): Save the errno before calling fcntl so fcntl doesn't overwrite errno in case of a previous error. (stream_write): Same. 2001-10-13 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_data): Replace the Bcc headers in a single location and don't flush the data unless the write was successful. 2001-10-12 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Remove the stripheader filter from the build. * camel-mime-filter-stripheader.[c,h]: Removed. * providers/smtp/camel-smtp-transport.c (smtp_data): Check the return value of camel_stream_flush to make sure it flushed the data successfully before continuing. Don't use the stripheader filter, it was completely broken - instead remove the header and set it again after we send. 2001-10-12 <NotZed@Ximian.com> * camel-vee-folder.c (camel_vee_folder_remove_folder): Unhook from all events before removing folder. (camel_vee_folder_add_folder): Hook onto delete event. (camel_vee_folder_finalise): Unhook from deleted event for suibfolder. (subfolder_deleted): When the folder is deleted, remove it from the vfolder automagically. 2001-10-11 <NotZed@Ximian.com> * camel-vee-folder.c (vee_search_by_uids): convert the uids list to a subset appropriate for each subfolder before trying to search on it. * camel-charset-map.[ch]: Removed charset/locale charset lookup and iconv_open/close functions, moved to gal. Fixed all callers. 2001-10-11 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (ssl_cert_is_saved): Same. * camel-tcp-stream-openssl.c (ssl_cert_is_saved): Instead of opening the file, stat it and make sure that it belongs to us. 2001-10-10 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-openssl.c (ssl_verify): Same hack as below. * camel-tcp-stream-ssl.c (ssl_bad_cert): Hack around the fact that adding a cert to nss's certdb seems to not work. 2001-10-10 <NotZed@Ximian.com> * camel-vee-folder.c (vee_sync): Only re-build the folder if we're expunging. Also only clear out the changed folders list if we're expungung and everything worked. 2001-10-10 Jeffrey Stedfast <fejj@ximian.com> * camel-charset-map.c (camel_charset_map_init): Added a hack for Solaris systems. 2001-10-10 <NotZed@Ximian.com> * providers/local/camel-maildir-store.c (scan_dir): oops, we want get_unread_message_count, not get_message_count! 2001-10-09 <NotZed@Ximian.com> * camel-service.c (camel_service_disconnect): Duplicate connect code that unregisters a cancel op if we created one. * camel-operation.c (camel_operation_progress): Another go at getting the logic right again. Make transients only update after 5 seconds (CAMEL_OPERATION_TRANSIENT_DELAY) (camel_operation_end): Likewise. * providers/local/camel-spool-folder.c (spool_search_by_uids): Implement. * providers/imap/camel-imap-search.c (imap_body_contains): If searching a sub-set of the total message count, then use a UID range to search only specific messages. * camel-vee-folder.c (vee_folder_change_match): Removed. (folder_changed_add_uid): Helper func for changed code. (folder_changed_remove_uid): " (folder_changed_change_uid): " (folder_changed): Rewritten. Supports proper auto-updating of changes, but not removals till a sync occurs. (vee_search_by_uids): Implement. (folder_changed): Changed to call an async threaded function to do the actual folder updating. * camel-folder-summary.c (camel_flag_list_copy): New func to copy a whole list of flags. (camel_tag_list_copy): New func to copy a whole list of flags. * providers/imap/camel-imap-folder.c (imap_search_by_uids): Implement. * providers/local/camel-local-folder.c (local_search_by_uids): Implement. * camel-folder.c (camel_folder_search_by_uids): New function, search a subset of uid's. (search_by_uids): Default impl, return error. 2001-10-08 Dan Winship <danw@ximian.com> * camel-folder.h (struct _CamelFolder): replace the ever-growing list of gbooleans with a single guint32 for flags. * camel-folder.c: Update folder flag setting/checking. * providers/pop3/camel-pop3-folder.c (camel_pop3_folder_init): * providers/nntp/camel-nntp-folder.c (camel_nntp_folder_new): * providers/local/camel-spool-folder.c (spool_init, camel_spool_folder_new): * providers/local/camel-maildir-folder.c (camel_maildir_folder_new): * providers/local/camel-local-folder.c (local_init): * providers/imap/camel-imap-folder.c (camel_imap_folder_init, camel_imap_folder_new): * camel-vtrash-folder.c (camel_vtrash_folder_init): * camel-vee-folder.c (camel_vee_folder_init): * camel-digest-folder.c (camel_digest_folder_init): update folder flag setting. 2001-10-07 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.c (get_folder_online): Unlock the store before returning error. (noted by NotZed) 2001-10-06 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_decode_text): Fix a slight error in logic. Just because the last word was an encoded word doesn't mean we ignore the lwsp padding, we only ignore it if the last word and *this* word are encoded. 2001-10-06 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-basic.c (filter): Implemented uuencoding and decoding. (complete): Implemented uuencoding and decoding. * camel-mime-utils.c (uuencode_close): New function to flush the uu encoder. (uuencode_step): New function to uuencode a block of data. 2001-10-05 <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (subscribe_folder): Use a folder_subscribed event rather than a folder_created one. (unsubscribe_folder): Similarly for unsubscribed/deleted. (create_folder): Emit a folder_created event on the folders we added. Not sure hsould probably check which ones are new and which ones are just existing ones? (delete_folder): Emit a folder_deletd event when we've deleted the folder. Part of fix for #11831. * camel-store.c (camel_store_class_init): Added folder_subscribed/unsubscribed events. 2001-10-05 Jeffrey Stedfast <fejj@ximian.com> * camel-service.c (camel_service_finalize): Turn off warnings. (get_name): Same. * camel-sasl.c (sasl_challenge): Turn off warnings. * camel-tcp-stream.c (tcp_connect): Turn off warnings. (tcp_getsockopt): Same. (tcp_setsockopt): Here too. (tcp_get_socket): And here. * camel-folder.c (folder_sync): Turn off warnings. (expunge): Same. (append_message): Here too. (get_message): And here. (search_by_expression): And again here. (folder_changed): Here too. * camel-store.c (get_folder): Set an exception and turn off debugging g_warnings. (create_folder): Same. (delete_folder): Here too. (rename_folder): And here. (get_trash): And here. (get_folder_info): Same. (free_folder_info): And again here. (camel_folder_info_build): Here too. (folder_subscribed): Same. (subscribe_folder): Here too. (unsubscribe_folder): And here. 2001-10-05 <NotZed@Ximian.com> * camel-mime-utils.c (header_decode_mailbox): Try harder to get broken names out of addresses. Unencoded ,'s in names will still break it, but well what can you do eh? (header_decode_mailbox): Always add .'s into address we've scanned past a '.', even if we can't decode the next part. Fix for some annoying bug #. 2001-10-04 <NotZed@Ximian.com> * camel-store.c (camel_store_delete_folder): Change the order of execution slightly. We delete the folder first, then call the classes delete folder, then cleanup the hashtable afterwards. This is so deletes of vfolders work, since vee-store uses the folder hashtable to store all of is vfolders :-/ 2001-10-04 Jeffrey Stedfast <fejj@ximian.com> * camel-service.c (camel_service_connect): Unregister the operation before unreffing it. 2001-10-03 <NotZed@Ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_changed): Remove now unused 'recents' array/freeing func. * camel-object.c (camel_object_trigger_event): Oops, always run the prep function, even if we have no hooks listening. Causes incoming imap recents to get ignored bythe camel-folder filter inbox code. * providers/local/camel-local-folder.c (local_sync): Dont explicitly save metadata, its saved in summary_sync. * providers/local/camel-mh-summary.c: Call superclass. (mh_summary_check): Dont save ibex. * providers/local/camel-maildir-summary.c: Call superclass method. (maildir_summary_check): Dont save ibex explictly. * providers/local/camel-mbox-summary.c: Call super-class sync when done, saves summary/ibex, etc. (mbox_summary_check): Dont save ibex. (mbox_summary_sync): Call summary_check rather than trying to update from our known position. * providers/local/camel-local-summary.c (camel_local_summary_check): Dont save the summary or index here. (local_summary_sync): Save the summary/index here. * camel-search-private.c (camel_search_build_match_regex): Added MATCH_NEWLINE flag, -> REG_NEWLINE. * camel-lock-helper.c: Include <string.h> to kill some warnings. * camel-filter-search.c (get_full_header): Put \n's between each line we generate. (header_full_regex): Dont drop the first argument to the command -> empty matches! (header_full_regex): Set match_newline for building regex. (body_regex): And here too. These all fix #9877. 2001-10-03 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (camel_store_delete_folder): Call camel_folder_delete() on the folder if it's "alive". * camel-folder.c (camel_folder_delete): Mark the folder as deleted, perform any folder specific cleanup and emit the deleted event. (camel_folder_sync): Only sync the folder if it hasn't been deleted. (camel_folder_expunge): Same. 2001-10-03 Dan Winship <danw@ximian.com> * providers/local/camel-local-provider.c (local_url_equal): Strange are the ways of alloca. Fixes 11543. 2001-10-02 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (ssl_bad_cert): Import the certificate if the user accepts it. 2001-10-02 <NotZed@Ximian.com> * providers/local/camel-local-provider.c (local_url_hash, local_url_equal): for hashing/comparing local url's, we ignore trailing /'s in paths (maybe shuld handle multiple /'s too). (camel_provider_module_init): Use local_url_hash/equal for all functions. * camel-sasl-digest-md5.c (digest_response): Change to camel_charset_iconv_open/close. * camel-pgp-context.c (pgp_verify): Change to camel_charset_iconv_open/close. * camel-mime-part-utils.c (convert_buffer): Change to camel_charset_iconv_open(). * camel-mime-filter-charset.c (camel_mime_filter_charset_new_convert, finalise): Change to camel_charset_iconv_open, etc. * camel-mime-utils.c: Use the camel_charset_iconv_open/close() functions to open/close it. * camel-charset-map.c (camel_charset_iconv_open): New function, wrap iconv_open, so we can cache ic's. (camel_charset_iconv_close): Likewise for close. (camel_charset_map_init,shutdown): Init/free iconv cache. 2001-10-02 Jeffrey Stedfast <fejj@ximian.com> * camel-charset-map.c (camel_charset_to_iconv): Revert my x-unknown special-case hack - this may mask other problems. * camel-mime-utils.c (rfc2047_decode_word): If the iconv conversion fails, for whatever reason, retry using the user's locale charset. 2001-10-02 <NotZed@Ximian.com> * providers/local/camel-maildir-store.c (get_folder_info): Go back to using the store url's path, not the toplevel_dir thing. * camel-operation.c (camel_operation_progress): Fix the progress logic, so we dont update too often. * camel-object.c (camel_object_get_hooks): Change the lock to a recursive e-mutex. (camel_object_hook_event): Maintain list length of hook list. (camel_object_unhook_event): " (camel_object_unhook_event): If we are in an event, just mark the pair as removed, without removing it. (camel_object_trigger_event): Before running events, copy the list, and also ignore 'removed' events. After running events, if we're all out of events, then free up any pending-removed events. (camel_object_free_hooks): Add some new assertions on the state of the hook structure. Removed the #error if threads not defined. It _should_ actually work without threads. (camel_object_free_hooks): Free mutex when done. 2001-10-01 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (delete_folder): Remove any cached messages that belonged to the deleted folder. (subscribe_folder): Don't ever let the info->name be NULL and don't use the stupid concat kludge to generate the URL, just use a CamelURL to do it. This way we don't risk breaking stuff by having a url like imap://fejj@imap//folder * camel-charset-map.c (camel_charset_to_iconv): If the charset is x-unknown, return the locale_charset. 2001-10-01 Dan Winship <danw@ximian.com> * camel-provider.h: Fix up the provider flags to specify things more completely so we don't have to hardcode provider names in the mailer. * providers/local/camel-local-provider.c: MH, mbox, and Maildir are LOCAL. MH and mbox are no longer STORAGE by the new definition. * camel-session.c (vee_provider): The vfolder store is a STORAGE. 2001-10-01 Dan Winship <danw@ximian.com> * providers/local/camel-local-store.c (construct): Don't munge the URL; CamelSession's caching relies on it not changing. Instead, add a toplevel_dir field to CamelLocalStore, and set that to the path, but always ending with /. (camel_local_store_finalize): Free toplevel_dir (camel_local_store_get_toplevel_dir): Return toplevel_dir rather than url->path. * providers/local/*: Lots of s/url->path/toplevel_dir/ * providers/local/camel-spool-store.c (construct): Likewise, don't try to strip a trailing / from url->path here, but I didn't make the corresponding toplevel_dir change, because there's no good reason someone should expect "/var/spool/mail/danw/" to work since that's not a directory. * providers/pop3/camel-pop3-folder.c (pop3_refresh_info): if STAT returns 0, don't bother sending UIDL. Speeds things up slightly and also works around a bug in a particular random POP server. (ximian bug 11369). 2001-09-29 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_sign): Add --no-secmem-warning and --no-greeting to the gpg command-line options. (pgp_clearsign): Same. (pgp_verify): Here too. (pgp_encrypt): And here. (pgp_decrypt): And finally here. 2001-09-28 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-search.c (get_source): Same here. * camel-mime-message.c (camel_mime_message_set_source): Don't use the e_url_shroud hack, use CamelURL functions instead. 2001-09-28 Jeffrey Stedfast <fejj@ximian.com> * providers/nntp/camel-nntp-provider.c (camel_provider_module_init): Init the nntp url hash and url_equal functions. * providers/sendmail/camel-sendmail-provider.c (camel_provider_module_init): Init the sendmail url hash and url_equal functions. * providers/smtp/camel-smtp-provider.c (camel_provider_module_init): Init the smtp url hash and url_equal functions. * providers/pop3/camel-pop3-provider.c (camel_provider_module_init): Init the pop3 url hash and url_equal functions. * providers/imap/camel-imap-provider.c (camel_provider_module_init): Init the imap url hash and url_equal functions. * providers/local/camel-local-provider.c (camel_provider_module_init): Init the local url hash and url_equal functions. * camel-session.c (camel_session_class_init): Init the vfolder url hash and url_equal functions. * camel-provider.h: Added url_hash and url_equal function pointers to the structure. * camel-vtrash-folder.c (vtrash_move_messages_to): Oops, a CamelFolder is not a CamelFolderClass. 2001-09-28 Dan Winship <danw@ximian.com> * camel-tcp-stream-openssl.c (ssl_error_to_errno): Make this take an SSL * instead of a CamelTcpStreamSSL *, since it can get called from open_ssl_connection, when the CamelTcpStreamSSL isn't set up right yet. Fixes a crash on connection failure. 2001-09-27 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_fetch_data): Grab the store's command_lock before grabbing the folder's cache_lock to prevent deadlock if another thread is processing an EXPUNGE response. * providers/imap/camel-imap-folder.c (imap_expunge_uids_resyncing): Fix a compiler warning that might point out a real bug... * providers/imap/camel-imap-folder.c (get_content): and one that doesn't 2001-09-27 <NotZed@Ximian.com> * camel-service.c (camel_service_connect): Dont re-register the connect_op if we got it by calling 'operation_registered', which returns an already-registered one. (camel_service_disconnect): Likewise here. This removes all the re-registered warnings. * camel-object.c (camel_object_unref): Fill the finalised data with nonsense, rather than 0's - more easily detect bad data. (camel_object_new): Clear instance data when we retrieve it off the list. * camel-object.h (_CamelObject): Added an event lock pointer for uh, locking event stuff. Also change the hooklist hashtable into a list. Changed all the code to handle it. Result: thread-safe events, event-hooks, and save memory too, and SHOULD FINALLY FIX THAT UNREAD COUNT PROBLEM. 2001-09-27 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-mime.c (camel_pgp_mime_part_decrypt): CRLF decode the stream before parsing it into a MIME part. This fixes bug #10521. * camel-store.c: Remove the old folder from the vTrash folder. 2001-09-27 Dan Winship <danw@ximian.com> * camel-service.c: Change "gboolean connected" to "CamelServiceConnectionStatus status", which can be disconnected, connecting, connected, or disconnecting. (camel_service_init, camel_service_finalize): create/destroy the connect_op_lock. Refer to service->status rather than service->connected. (camel_service_connect): When connecting, note the current operation (and create a new one if there's none registered) and mark the connection "connecting" until we succeed or fail. (camel_service_disconnect): Likewise in reverse. (camel_service_cancel_connect): New function to cancel a connection attempt. (cancel_connect): Default implementation: Call camel_operation_cancel on the connect_op. * camel-disco-store.c (disco_connect): Only call CamelRemoteStore's connect func if we're online. (disco_cancel_connect): Fall back to offline if a connection gets cancelled. (disco_get_folder_info): Kludge: call connect explicitly before deciding whether to do the online or offline version, so if the connect fails, we fall back correctly. * camel-session.c (camel_session_get_service_connected): s/svc->connected/svc->status/ * camel-remote-store.c (camel_remote_store_finalise): Change service->connected check to service->status check. (remote_connect): Don't set service->connected here: camel_service_connect() itself does that. * camel-operation.c (camel_operation_registered): Deal with the possibility that there's no registered op. 2001-09-26 <NotZed@Ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_message): If we get a search error, just abort, dont try and fall back and copy to inbox. (camel_filter_driver_filter_message): Make sure we ALWAYS use exceptions for important things - like moving messages to inbox! (camel_filter_driver_filter_message): If we have the source folder, use camel_folder_set_message_flags rather then poking the info directly, which skips changed events. This means filtering immediate doesn't lose changed events. (do_flag): Same here. (do_colour): Same thing but using set_tag. (do_score): Same again. (camel_filter_driver_filter_folder): Use ~0 as the 'set' arg to set_flags, just saves typing. (open_folder): We cache folders that we couldn't open as well as those ones we could. Use magic token FOLDER_INVALID = (void *)~0 as the marker. (close_folder): Handle FOLDER_INVALID case properly. 2001-09-26 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-message.c (camel_mime_message_get_part_by_content_id): New function to convenience Larry ;-) (check_content_id): Oops, return !found instead of found. This callback has to return whether or not to keep searching, not whether or not it found what it's looking for. Do'h! * camel-pgp-mime.c (camel_pgp_mime_is_rfc2015_signed): block out some code if ENABLE_PEDANTIC_PGPMIME is not defined. 2001-09-26 <NotZed@Ximian.com> * camel-vee-store.c: Emptied VeeStorePrivate, member wasn't used. * camel-vee-folder.c (camel_vee_folder_set_expression): Lock changed list separately. (camel_vee_folder_remove_folder): " (camel_vee_folder_set_folders): Use changed_lock for changed list. (vee_refresh_info): Use changed_lock, also just grab the list, reset it in the lock, and do the work unlocked. (vee_sync): Use changed lock for changed list. (folder_changed): ". All this kills a deadlock with sync/expunge. * camel-private.h (CamelVeeFolderPrivate): Added a new lock for the folders-changed list. (CamelVeeStorePrivate): Removed, defined in camel-vee-store.c, thanks to a pedantic. 2001-09-26 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Fix Ettore's fix. 2001-09-25 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (camel_pgp_context_new): No longer takes a remember argument. (pgp_sign): Only uncache the passphrase on failure. (pgp_clearsign): Same. (pgp_encrypt): Here too. (pgp_decrypt): And here. (pass_free): New function to zero the passphrase before freeing it. 2001-09-25 Ettore Perazzoli <ettore@ximian.com> [Patch for Automake 1.5 compatibility pointed out by Richard Boulton <richard@tartarus.org>, as per #9258.] * providers/nntp/Makefile.am: Remove some commented parts that make Automake 1.5 choke. * Makefile.am (camel-lock-helper): Add $(EXEEXT) for Automake 1.5 compatibility. 2001-09-25 <NotZed@Ximian.com> * camel-folder.c (get_unread_message_count): Only unref info if we got it. * camel-operation.c (camel_operation_unref): Fix for !threads enabled not ccompiling. (camel_operation_ref): Assert refcount > 0. (struct _CamelOperation): Removed the lock. On further investigation, I dont think this will always work, the registration operations assume that a lookup in the operation_active table will return a ref, that will remain valid until we ref it, which needn't be the case. So now i'm using a single global lock, since we'd need to do that for unref anyway, and every operation is fast & memory-bound. Changed all the code to handle this. (camel_operation_progress_count): Since the code is identical, just call progress() for now. (camel_operation_register): No longer refcount, use unref to check/clear the active table. (camel_operation_unregister): Same here. (camel_operation_unref): Check if operation is in active table, if so, warn, remove. 2001-09-25 Dan Winship <danw@ximian.com> * camel-tcp-stream-openssl.c (my_SSL_read, my_SSL_write): call SSL_read/write, looping on SSL_ERROR_WANT_READ/WRITE. I'm sure there's a perfectly good reason that the API works this way. No, really. (stream_read, stream_write): use my_SSL_read and my_SSL_write. Fixes at least ximian 8593, and probably 6024 and maybe 10366, at least for OpenSSL. There may be a parallel NSS bug? 2001-09-25 <NotZed@Ximian.com> * camel-stream-buffer.c (stream_write_all): A "safe write" function that always writes out all data asked of it, till finished. (stream_write): Rewritten. What was i on when i wrote that originally, how did it ever work with tcp streams ... 2001-09-24 <NotZed@Ximian.com> * camel-folder.c (append_message): Set the exception to FOLDER_INVALID rather than making out it worked. A fix for #7025. (search_by_expression): Here too. * providers/imap/camel-imap-folder.c (imap_update_summary): In the 'lose' case, free messages, not fetch_data twice! * camel-filter-search.c (check_header): If the type is encoded, get the fallback charset from the message content-type. * camel-mime-part-utils.c (check_html_charset): Dont check for charset==null before calling charset_to_iconv. (simple_data_wrapper_construct_from_parser): " * camel-mime-message.c (process_header): Try use content-type charset param as the fallback charset. * camel-charset-map.c (camel_charset_to_iconv): Handle name == NULL, return NULL. * camel-folder-summary.c (camel_folder_summary_format_address): (camel_folder_summary_format_string): Made private again, removed #warning about it. Renamed to s/camel_folder//. (summary_format_string): Take default charset param. (camel_message_info_new_from_header, message_info_new): Decode content-type field to get the charset parameter to use as the default charset for decoding strings. * camel-search-private.c (camel_search_header_match): Pass NULL as the charset, the locale charset is always tried. (camel_search_header_match): Supply a default_charset parameter to be used with TYPE_ENCODED params. * camel-mime-utils.c (header_param): get rid of the g_strcasecmp crap. (header_set_param): Same here. (header_decode_param_list): And here. (header_decode_text): Totally rewritten. 30% of its size. If the word is not rfc2047 encoded, always try default_charset if supplied, if that fails, try locale charset if it exists, if that fails then assume latin1/7 bit ascii. (append_8bit): Changed to return FALSE if we can't convert for whatever reason, and dont append anything. * camel-mime-part.h (struct _CamelMimePart): Move content_type and headers out of the 'private' section. * camel-mime-part.c (get_headers): Dont do any conversion on the header. (process_header): Get the content-type charset as the fallback charset for decode_string. (construct_from_parser): IF we have a content-type header, process it before doing anything else, so we have access to a fallback charset for invalid headers. 2001-09-23 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-openssl.c (ssl_verify): Don't even try to alert the user if the session isn't interactive. (errlib_error_to_errno): Make the default errno EINTR so that we act just like CamelTcpStreamSSL. * camel-pgp-context.c (pgp_sign): When the password is not provided, set the exception to USER_CANCEL. (pgp_clearsign): Same. (pgp_encrypt): And here. (pgp_decrypt): Here too. 2001-09-21 <NotZed@Ximian.com> * camel-store.c (create_folder): Set the exception if create_folder isn't implemented. 2001-09-21 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (create_folder): Go back to not using camel_folder_info_build() here because this seems to break folder creation (returning nodes from the root path when it shouldn't be?). 2001-09-21 <NotZed@Ximian.com> * camel-vee-store.c (vee_get_folder_info): Force a refresh of vfolders before storing their count. This essentially makes 'get_folderinfo' refresh all vfolders, so that get-mail works as expected. * camel-vee-folder.c (camel_vee_folder_finalise): Clear lists. (folder_changed): If we're not autoupdating the folder, then add this folder to the list of 'changed' folders for this vfolder. (camel_vee_folder_set_expression): Reset the folders_changed list. (camel_vee_folder_remove_folder): Remove the folder from the folders_changed list. (camel_vee_folder_set_folders): If we have a folder already, but its changed, use this opportunity to update it. (vee_sync): Remove any synced folders from the folders_changed list. (vee_refresh_info): Refresh folder, re-run queries on any changed folders. 2001-09-20 <NotZed@Ximian.com> * camel-folder-summary.c (camel_folder_summary_header_load): New function to just load the summary. Not much use yet, until unread counts are stored in the summary (which now looks like too much work!). * camel-mime-utils.c (header_decode_mailbox): Crappy fix for stupid mailers. If we get 'name b. name <foo>' then treat it as '"name b. name" <foo>'. See 8147. (header_decode_mailbox): Another fix for really broken mailers that have things that look like addresses followed by <address> bits. (rfc2047_encode_word): Escape ??= to avoid some bizarre warning about trigraphs. * providers/local/camel-local-folder.c (camel_local_folder_construct): Setup unread count properly, and also setup the url properly. * providers/local/camel-maildir-store.c (camel_folder_info_new): Build the path on the folderinfo. (scan_dir): Dont include empty host part in url for folder. (scan_dir): Lookup folder counts for any folders we currenty have open. (scan_dir): If we dont have the folder open, then scan the 'new' and 'cur' folders for new and unread messages to count instead. * providers/local/camel-spool-store.c (get_folder_info): If we have this folder open, then set unread from the folder itself. (get_folder_info): Change the INBOX path to not include the leading /. * providers/local/camel-spool-folder.c (camel_spool_folder_construct): Fix path. (camel_spool_folder_construct): Properly setup the url field when we say we've got a new folder created. * providers/imap/camel-imap-store.c (unsubscribe_folder): Build path on folderinfo. * camel-store.c (camel_folder_info_build_path): Prepend / to path. * camel-vee-folder.c (camel_vee_folder_construct): Use macro for unmatched folder name. (camel_vee_folder_new): " * camel-vee-store.c (vee_get_folder_info): Use macro for unmatched folder name. (vee_delete_folder): " (vee_rename_folder): " (vee_get_folder): Prepend / onto path. (vee_delete_folder): Setup path in folderinfo. (vee_get_folder_info): Fix path in folderinfo. * camel-vee-folder.h (CAMEL_UNMATCHED_NAME): New define for unmatched (internal) name. 2001-09-20 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-spool-folder.c (camel_spool_folder_construct): And finally here. * providers/local/camel-local-folder.c (camel_local_folder_construct): And here. * camel-vee-store.c (vee_get_folder): ANd here too. (vee_get_folder_info): And here. * providers/imap/camel-imap-store.c (create_folder): Use camel_folder_info_build() and do better error checking. (subscribe_folder): Set the path on the folder info. * camel-store.c (camel_folder_info_build): Set the folder-info path if it hasn't already been set by the caller. (camel_folder_info_free): Free the path. * camel-store.h: Added a path member to the FolderInfo type. 2001-09-20 Dan Winship <danw@ximian.com> * camel-charset-map.c (camel_charset_to_iconv): Nuke debug warning. 2001-09-19 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c: Updated exception strings to be more clear. * camel-pgp-mime.c (pgp_mime_part_sign_restore_part): Stupid mistake, fixes bug #9975. * camel-tcp-stream-ssl.c (ssl_bad_cert): Oops, don't fail if `data' isn't a CamelService...cuz it's not supposed to be! It's a CamelTcpStreamSSL object...doh! 2001-09-19 <NotZed@Ximian.com> * General cleanup of camel debug printfs. * camel-lock.c (camel_lock_fcntl): Changed to return 'success' if the error indicates file locking isn't supported on this filesystem. Still return a warning just incase (if its the first time). Might fix a lot of reported bugs. * providers/local/camel-spool-store.c (get_folder_info): Dont include the empty // host part in the uri. This 'breaks' the service lookup. 2001-09-18 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c: Match mailing-list header List-Owner. 2001-09-18 <NotZed@Ximian.com> * camel-vee-store.c (vee_get_folder): Fix the uri genereated for the folderinfo for the folder_created event. * camel-store.h: Added 'total' to CamelFolderInfo. 2000-09-18 <NotZed@Ximian.com> * providers/local/camel-maildir-folder.c (camel_maildir_folder_new): If filter inbox is set on the store, and we're opening inbox '', then enable filtering on new messages. * providers/local/camel-local-folder.c (camel_local_folder_construct): After loading the summary, check it, and only abort if that fails. Also maintain the changes count. * providers/local/camel-local-summary.c (camel_local_summary_load): Remove summary_check code from here. (camel_local_summary_check): Sync index/summary here, if we were successful. * providers/local/camel-spool-folder.c (camel_spool_folder_new): If we have filter-new-messages-on-inbox set and we just opened inbox, turn on filtering on this folder. (camel_spool_folder_construct): Keep track of changes for the folder, so that filter-new-messages works right (?) * providers/local/camel-spool-store.c (get_folder): Pass 'INBOX' as the folder name, not the path. * camel-folder-search.c (search_not): Modified patch from <peterw@ximian.com> since the summary is messageinfo's, not strings. ** Ok so the problem with the stuff below is that maildir/spool 'summary_load' throws away all events, including recents, joy eh? * providers/local/camel-maildir-summary.c (maildir_summary_check): Add new messages to the recent changeinfo. * providers/local/camel-spool-summary.c: Mark 'new' message as recent, for later processing if required (i.e. 'filter new messages'). * camel-store.c (construct): new function, cascade up construct method and check for 'filter' param, and set accordingly for any one that might want it. * providers/imap/camel-imap-store.c (construct): map the param_filter_inbox flag to the store->flags as CAMEL_STORE_FILTER_INBOX. * camel-store.h (CAMEL_STORE_FILTER_INBOX): new flag to tell it to filter inbox. * providers/imap/camel-imap-folder.h: Removed do_filtering flag from CamelImapFolder. * providers/imap/camel-imap-folder.c (imap_update_summary): Remove the 'recents' parameter, use the 'changes' parameter instead to convey this info. (camel_imap_folder_changed): Changed for update_summary api change. Now always just emit the changed event if we have any changes to report, filtering code removed, handled upstream. (filter_proc): (filter_free): Removed old filtering code. (camel_imap_folder_new): Set the filter_recent flag on the folder if we're the inbox and filtering of inbox is enabled. * camel-folder.c (folder_changed): If we have 'recent' messages, and are set to filter recents, then freeze the folder and launch a thread to do the processing (or similar if threading not enabled). (thaw): Make sure we emit the changed signal outside of owning the lock and if things have changed. Also, no longer bother downgrading folder_changed events to message_changed events. * camel-folder.h (struct _CamelFolder): Added filter_recent flag -> signifies all recent messages in folder should be filtered. * camel-session.c: (camel_session_thread_msg_new, camel_session_thread_msg_free, camel_session_thread_queue, camel_session_thread_wait): code to handle async operations as part of camel processing. (camel_session_finalise): free thread_lock, destroy thread, active hash, etc. (camel_session_init): init thread, active hash, etc. (camel_session_class_init): Init virtual functions. (session_thread_msg_new, session_thread_msg_free, session_thread_destroy, session_thread_received, session_thread_queue, session_thread_wait): default implementation of session threads stuff. 2001-09-17 <NotZed@Ximian.com> * camel-folder.c (camel_folder_change_info_recent_uid): New function to add a 'recent' uid to the change info. (camel_folder_change_info_clear): Clear recent list. (camel_folder_change_info_free): Free recent list. (camel_folder_change_info_new): Setup recent list. * camel-folder.h: Added a uid_recent item to the folder_changed event data. * providers/local/camel-maildir-store.c (scan_dir): Free new in the right block. * providers/local/camel-local-provider.c: Add local config entries to filter on new messages in spool and maildir provider. * camel-vee-folder.c (vee_folder_construct): Remove the assertion which stops ? in names from being allowed. 2001-09-18 Dan Winship <danw@ximian.com> * camel-data-wrapper.c (camel_data_wrapper_is_offline): Virtualize this. * camel-medium.c (is_offline): A medium is offline if its content object is offline. * camel-multipart.c (is_offline): A multipart is offline if any of its subparts are offline. 2001-09-18 Jeffrey Stedfast <fejj@ximian.com> * camel-charset-map.c: Added korean charset conversion. 2001-09-17 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (delete_folder): Don't process the ImapResponse data, we don't want to block forever on data we don't care about... (create_folder): Use get_folders_online() instead of get_folder_info_online(). * camel-folder.c: Reverted previous changes. 2001-09-17 JP Rosevear <jpr@ximian.com> * Makefile.am: Helpful install-exec-hook debugging output 2001-09-16 Jeffrey Stedfast <fejj@ximian.com> * camel-folder.c (camel_folder_get_uri): New function to get the URI of a folder. (get_uri): Default implementation. 2001-09-14 <NotZed@Ximian.com> * camel-vee-folder.c (camel_vee_folder_set_folders): New function, set the complete list of folders on a vfolder, all at once. (camel_vee_folder_set_expression): If we set the query to be the same thing, dont do anything. * camel-vee-store.c (camel_vee_store_init): Turn off vtrash for this store. * camel-store.c (camel_store_init): Enable vtrash by default via flags. (camel_store_get_trash): REturn NULL if the store doesn't support vtrash. (init_trash): Dont init if store doesn't support vtrash. * camel-store.h: Add a new flags CAMEL_STORE_VTRASH -> store supports vtrash. 2001-09-13 <NotZed@Ximian.com> * camel-vee-store.c (vee_get_folder_info): Implement. (build_info): Used to build a folder record from the folders hashtable. (vee_delete_folder): Implemented, remove folder from hashtable. (vee_rename_folder): Implemented, remove old folder from hashtable, add new one and rename its guts too. * camel-store.c (camel_store_rename_folder): Do nothing if we're not asked to actually change names. Also dont do the renamed cache check till after we've called the subclass. (camel_store_delete_folder): Call the subclass delete firs,t then make sure the cache is right. * camel-vee-folder.c (vee_folder_construct): Remove support for name?query syntax to setup vfolder. Abort if such syntax is used. (camel_vee_folder_new): Removed code that handles ? syntax, etc. (vee_folder_build_folder): Handle unset expression, treat it as an empty search. (camel_vee_folder_set_expression): Oops, actually set the expression. * camel-vtrash-folder.c (camel_vtrash_folder_new): Dont use name?query syntax to setup vfolder, but set the expression directly. Also fixes a small memleak. 2001-09-12 <NotZed@Ximian.com> * camel-store.c (camel_store_delete_folder): Fixed warnings with a cast. (camel_store_rename_folder): " 2001-09-14 Jeffrey Stedfast <fejj@ximian.com> * camel-url.c (camel_url_set_param): Use g_datalist_set_data_full so that the data will be free'd when we clear the list. 2001-09-14 Dan Winship <danw@ximian.com> * camel-file-utils.c (camel_file_util_encode_string, camel_file_util_decode_string): Don't claim failure when reading/writing the empty string. 2001-09-14 JP Rosevear <jpr@ximian.com> * Makefile.am: use install hook instead of install rule to guarantee we run after installation 2001-09-13 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_folder): Don't reuse the exception if it has already been set. (camel_filter_driver_filter_message): Same here. Also use the new return value from camel_filter_search_match(). * camel-filter-search.c (camel_filter_search_match): Return an integer (matched, no-match, or error). * providers/local/camel-mbox-folder.c (mbox_append_message): Do error-checking based on function return values rather than exceptions as it's possible for them to be NULL. (mbox_get_message): Same. * providers/imap/camel-imap-folder.c (imap_append_offline): Pass an exception to the cache. (imap_append_online): Same. (imap_append_resyncing): Here too. (imap_copy_offline): And here. (handle_copyuid): Pass NULL as the exception here... (parse_fetch_response): And finally here. * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_insert): (insert_setup): (camel_imap_message_cache_insert_stream): (camel_imap_message_cache_insert_wrapper): (camel_imap_message_cache_copy): Take an exception argument and set it on error. * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): Convert the charset to the iconv-friendly name. * camel-charset-map.c (camel_charset_to_iconv): Add code to convert windows-[cp]#### charsets to their iconv-friendly format of cp####. * camel-pgp-context.c (crypto_exec_with_passwd): Return -1 on failure. 2001-09-12 Larry Ewing <lewing@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_mbox): free the source url in the nonfailure case. 2001-09-12 <NotZed@Ximian.com> * camel-folder-search.h: Change the 'not' virtual method from an immediate e-sexp function to a non-immediate one, which is what it should be. * camel-folder-search.c (search_not): Implement a 'not' function ourselves. If 'not' on a vector value, then not over all items in the summary. (builtings[]): Change the 'not' builtin to be a non-immediate function. 2001-09-12 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_mbox): Unref the message info. 2001-09-10 <NotZed@Ximian.com> * providers/local/camel-maildir-store.c (get_folder_info): Pass in a hashtalbe used for finding visited nodes. (inode_hash): (inode_equal): (inode_free): Helper functions for visited node checks, hashes on inode + device number. (scan_dir): Dont re-visited directories we've already visited, use inodes + device number. Also, if the folder doesn't contain cur/tmp/new, mark it as noselect, but continue to recurse folders, if asked. * camel-folder.c (move_messages_to): Progress reporting while moving messages. * camel-store.c (camel_store_delete_folder): Make sure we remove the folder from our vtrash, if we have it in our opened set, and also remove it from our hashtable. (camel_store_rename_folder): Remove the old folder name from the folder cahce, if it is there. 2001-09-10 Jeffrey Stedfast <fejj@ximian.com> * camel-charset-map.c (camel_charset_to_iconv): Get rid of the g_warning, we no longer really need that. 2001-09-10 <NotZed@Ximian.com> * providers/imap/camel-imap-store.c (camel_imap_msg_new): Forgot to return the newly allocated msg. 2001-09-08 Not Zed <NotZed@Ximian.com> * providers/local/camel-spool-summary.c (spool_summary_sync_full): Likewise. * providers/local/camel-mbox-summary.c (mbox_summary_sync_full): Fix the wording. * camel-tcp-stream-raw.c (stream_read): Save errno around fcntl call in cancellable read case. Also, loop while we get EINTR. This might fix some weird reconnect behaviour with pop. (stream_write): " 2001-09-07 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (ssl_bad_cert): #if 0 out the code that was hopefully going to fix bug #5325 because the functions I used seem to have been deprecated. *sigh*. * camel-remote-store.c (camel_remote_store_finalise): Disconnect from the service here otherwise the locks will be destroyed and by the time CamelService finalizes we'll crash or hang. * providers/smtp/camel-smtp-transport.c (smtp_data): Make sure to unref the mime filters when we finish with them. 2001-09-07 Ettore Perazzoli <ettore@ximian.com> * providers/local/camel-mbox-summary.c (mbox_summary_sync_full): s/Synchronising/Synchronizing/. (mbox_summary_sync_quick): Likewise. (summary_rebuild): s/Summarising/Summarizing/. 2001-09-07 Jeffrey Stedfast <fejj@ximian.com> * camel-multipart.c (camel_multipart_set_boundary): Take a const char * instead of a char *. * camel-mime-part-utils.c (camel_mime_part_construct_content_from_parser): Reuse the original boundary instead of generating our own. 2001-09-06 <NotZed@Ximian.com> * providers/imap/camel-imap-folder.c (filter_proc): Thread function to perform filtering. (filter_free): Free the filter object. (camel_imap_folder_changed): If we have threads enabled, then queue up a request to perform filtering. For bug #4422. * providers/imap/camel-imap-store.c (camel_imap_msg_new): Create a new 'imap msg'. (camel_imap_msg_queue): Queue an 'imap msg'. (async_received): Handle receiving of imap msg in async thread. (async_destroy): Handle destroying of imap msg. (camel_imap_store_finalize): Destroy thread when done. (camel_imap_store_init): Setup thread. * providers/imap/camel-imap-store.h: Added an ethread to the store, for async processing. 2001-09-06 Jeffrey Stedfast <fejj@ximian.com> * camel-remote-store.c (remote_recv_line): Use camel_stream_buffer_read_line() instead of duplicationing the functionality. Also, the previous way was broken anyway. What if a line was the same length as our buffer? Then we'd go and read a second line and a third and so on until they weren't the same length, leaving \r's in the middle of the buffer. 2001-09-06 Dan Winship <danw@ximian.com> * providers/pop3/camel-pop3-store.c (pop3_get_response): Fix this to always set an exception if it returns POP3_FAIL, as documented. * providers/pop3/camel-pop3-folder.c (pop3_get_message_stream): Revert. 2001-09-06 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-folder.c (pop3_get_message_stream): Set an exception on POP3_FAIL as well. 2001-09-06 Dan Winship <danw@ximian.com> * providers/pop3/camel-pop3-provider.c: #ifdef out the "delete after N days" option in pop3_conf_entries, since it won't be done for 1.0. 2001-09-04 Jeffrey Stedfast <fejj@ximian.com> * camel-remote-store.c (remote_disconnect): Too late to sync the folders here... if this is gonna happen at all it should be done in the provider. * camel-service.c (camel_service_finalize): Do a clean disconnect. 2001-08-30 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-spool-store.c (get_inbox): Return the default folder (ie. INBOX) rather than saying we don't support an inbox. * camel-tcp-stream-ssl.c: #include pk11func.h. (ssl_bad_cert): Use CERT_GetDefaultCertDB. 2001-08-29 Jeffrey Stedfast <fejj@ximian.com> * camel-smime-context.c (smime_sign_prepare): Do the same here as what we did in the pgp code a moment ago. (smime_sign_restore): And again here. (smime_sign): And update thise code to pass in the address of the encodings list when restoring parts. * camel-pgp-mime.c (pgp_mime_part_sign_prepare_part): Special-case message/rfc822 mime parts - don't set an encoding on these, instead traverse into their subparts and set the encodings for those. (pgp_mime_part_sign_restore_part): Reverse any operations done to message/rfc822 parts in the above prepare_part() function and also take a pointer to a GSList of encodings instead of just a GSList so we can properly keep track of which encoding we are on. (camel_pgp_mime_part_sign): Pass in the address to the encodings list when restoring parts. 2001-08-29 Not Zed <NotZed@Ximian.com> * providers/pop3/camel-pop3-folder.c (pop3_sync): Add progress reporting to deleting messages. Dont change the calculation to a bloody float. Bug #6189. 2001-08-28 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (crypto_exec_with_passwd): Initialize buf and diag_buf to NULL, hopefully this fixes the crash if the select() times out. (pgp_encrypt): Argh. What was I thinking? Don't use "-r recipient" as an argument, instead use "-r" and "recipient" as 2 separate arguments to pgp. This might fix the "gpg only encrypts to my private key" bug? 2001-08-28 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-store.c (create_folder): Whoops. Double-free and FMR = bad. * camel-remote-store.c (remote_send_string): Don't reveal the user's IMAP password if their server supports LITERALPLUS. Also add some more dashes to make it line up nicely :-) 2001-08-28 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (ssl_bad_cert): If the user accepts the certificate, add it to the database as a trusted CA. 2001-08-28 Peter Williams <peterw@ximian.com> * camel-object.c (camel_object_unref): Make the locking more fine-grained on destruction; specifically, don't hold the type lock while calling the finalize functions. 2001-08-27 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-local-provider.c: Changed the description/names of some of the local providers to try and clarify the meanings. * providers/pop3/camel-pop3-provider.c: Simplify the POP description - we want short but sweet. * providers/imap/camel-imap-provider.c: s/IMAPv4/IMAP 2001-08-27 Not Zed <NotZed@Ximian.com> * camel-mime-part-utils.c (convert_buffer): re-read the iconv man page, and treat the return value properly. It returns the number of non-reversible conversions performed, not the number of output characters, sigh. (check_html_charset): Changed to just take a buffer of data, and not the mime parser. (simple_data_wrapper_construct_from_parser): Since we dont need the charset till we have all the data, search for the charset after we've read the data, if we have html data with no charset in the header. (simple_data_wrapper_construct_from_parser): Remove the seekable_source stuff. * Re-apply patches from before. 2001-08-25 Not Zed <NotZed@Ximian.com> ["Summarising" and "Synchronising" are spelt with a "s" in ENGLISH] * providers/local/camel-mbox-summary.c (mbox_summary_sync_quick): s/Synchronizing/Synchronising/. * providers/local/camel-mbox-summary.c (summary_rebuild): s/Summarizing/Summarising/ 2001-08-24 Jeffrey Stedfast <fejj@ximian.com> * camel-disco-diary.c (camel_disco_diary_replay): Use fseek instead of fseeko since we want to be portable and use ftell rather than ftello for the same reason. (camel_disco_diary_empty): Same here. 2001-08-24 Peter Williams <peterw@ximian.com> * Revert Not Zed's previous commits as they apparently can cause serious data loss for POP3 users. 2001-08-24 Ettore Perazzoli <ettore@ximian.com> ["Summarizing" and "Synchronizing" are spelt with a "z" in the default locale, that happens to be "en_US".] * providers/local/camel-mbox-summary.c (mbox_summary_sync_quick): s/Synchronising/Synchronizing/. * providers/local/camel-mbox-summary.c (summary_rebuild): s/Summarising/Summarizing/ 2001-08-24 Not Zed <NotZed@Ximian.com> * camel-mime-part-utils.c (convert_buffer): Limit the maximum output buffer size. iconv bugs? returns E2BIG forever on some (valid) data? * providers/local/camel-spool-summary.c (summary_rebuild): Likewise. * providers/local/camel-mbox-summary.c (summary_rebuild): Summarising is spelt with an s. 2001-08-23 Not Zed <NotZed@Ximian.com> * camel-mime-part.c (write_to_stream): If we have rawtext, then dont do any charset conversion, only encoding/crlf conversion. * camel-data-wrapper.h (struct _CamelDataWrapper): Added 'rawtext' member, says the character encoding is raw, not utf8. * providers/local/camel-spool-summary.c (spool_summary_sync_quick): Synchronising is spelt with an s, not a z. (spool_summary_sync_full): " * providers/local/camel-mbox-summary.c (mbox_summary_sync_full): No, synchronising is spelt with an s. (mbox_summary_sync_quick): " * camel-mime-part-utils.c (camel_mime_part_construct_content_from_parser): Remove the warnings which aren't going anywhere fast. (convert_buffer): Function to convert a bytearray of textual content from one charset to another. If the charset doesn't exist or fails, then do no conversion. (simple_data_wrapper_construct_from_parser): First, read in data, then try and convert it using the specified charset if supplied. If that fails, then dont do any conversion, and leave as raw. Also, if we have any x-* charsets, then dont do any processing. 2001-08-23 Peter Williams <peterw@ximian.com> * providers/local/camel-mbox-summary.c (summary_rebuild): Add a missing \n to the end of a printf. * providers/local/camel-spool-summary.c (summary_rebuild): Same. 2001-08-22 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-folder.c (pop3_generate_uids): Oops. Fixed a mis-use of GPtrArray. If we set_size, then using g_ptr_array_add will assume `size' elements have already been initialized and will start appending on to the end. 2001-08-21 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_verify): Turn on --batch and reenable gpg keyserver lookups. (pgp_decrypt): Disable keyserver lookups if we are disconnected. 2001-08-21 Dan Winship <danw@ximian.com> * camel-folder-thread.c (camel_folder_thread_messages_new_summary): Fix the duplicate message check so it doesn't mess up threading when a reply appears in the folder before the message it's a reply to. Fixes bug #3357. (camel_folder_thread_messages_new): Likewise, although I don't think this function is used any more. 2001-08-20 JP Rosevear <jpr@ximian.com> * Makefile.am: build a static camel for the addressbook conduit 2001-08-20 Ettore Perazzoli <ettore@ximian.com> * providers/local/camel-mbox-summary.c (mbox_summary_sync_full): s/Synchronising/Synchronizing/. (mbox_summary_sync_quick): Likewise. * providers/local/camel-spool-summary.c (spool_summary_sync_full): Likewise. (spool_summary_sync_quick): Likewise. 2001-08-20 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (create_folder): Get the entire directory structure for the folder we just created, meaning if we created a folder named "test.mailbox" where test didn't previously exist, get the listing for "test" and "test.mailbox". * providers/imap/camel-imap-utils.c (imap_parse_folder_name): New function. 2001-08-19 Chris Toshok <toshok@ximian.com> * camel-store.h: #include <sys/types.h> for mode_t on freebsd. 2001-08-19 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-folder.c (pop3_generate_uids): Make sure the generated UID is non-NULL before appending it to the array of UIDs. How would this even happen? I have no clue... 2001-08-17 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (imap_update_summary): Start scanning new messages at maxuid + 1 rather than uidval + 1? Maybe I'm missing the logic, but anyways...this might fix bug #5348. 2001-08-16 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (imap_keepalive): If we are online and message flags have changed in the current folder, sync the flags else do a NOOP. This fixes bug #4666. Also lock around the NOOP command. * providers/imap/camel-imap-folder.c (imap_sync_online): Don't ever send "STORE FLAGS.SILENT ()" since Cyrus imapd doesn't like this (and maybe others too?) so what we'll do instead is to send "STORE -FLAGS.SILENT (\Seen \Answered \Deleted \Marked)" to clear all the flags (since this is in essence what we want to do anyway). 2001-08-16 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-store.c (imap_connect_online): Fix this again. Sigh. Problem of namespace = "" may not yet be completely solved. 2001-08-15 Peter Williams <peterw@ximian.com> * camel-tcp-stream-openssl.c (ssl_error_to_errno): New function. Try to approximate the SSL error into errno. (errlib_error_to_errno): New function, try to approximate OpenSSl's error library's error (ERR_*) into errno. (stream_read): Try to set errno using ssl_error_to_errno. (stream_write): Same. (open_ssl_connection): Same. * providers/imap/camel-imap-store.c (imap_connect_online): Oh crap, huge killer typo. 2001-08-14 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-store.c (imap_connect_online): Canonicalize store->namespace to end in store->dir_sep, once both values are known. The %F-related code makes this assumption. Probably fixes a bug reported on the mailing list. 2001-08-13 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (camel_store_get_folder): After talking to NotZed, it turns out I was wrong after all. (store_sync): Create a copy of the folder-cache that owns a ref on each of the folders so that if one of the folders get's finalized inside store_sync(), we don't run into any locking issues. This is mostly meant to solve a problem in IMAP (#6089). 2001-08-13 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-store.c (delete_folder): Fix a leak. * providers/imap/camel-imap-utils.c (imap_namespace_concat): Bleah, handle when namespace = NULL (can happen upon initial open of mailbox.) * providers/imap/camel-imap-command.c (imap_command_strdup_vprintf): Don't crash when %F'ing with an empty folder name and NULL namespace. 2001-08-13 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (camel_store_get_folder): We need to be ref'ing the folder if/when we add it to the hash. This may fix bug #6089? (camel_store_get_folder): Nevermind, reverted the fix. (camel_store_get_folder): No, I wasn't on crack before after all...if we don't ref here, then we're open to a race condition later. 2001-08-11 Ettore Perazzoli <ettore@ximian.com> * providers/local/camel-spool-summary.c (summary_rebuild): s/summarising/summarizing/. (summary_rebuild): Likewise. * providers/local/camel-mbox-summary.c (mbox_summary_sync_full): s/summarise/summarize/. (mbox_summary_sync_quick): Likewise. (summary_rebuild): Likewise. (mbox_summary_check): Likewise. 2001-08-10 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-utils.c (imap_quote_string): Added an assert - at this point a backtrace would be more useful than a transaction log for debugging. * providers/imap/camel-imap-command.c (imap_command_start): Return FALSE here, not NULL. * providers/imap/camel-imap-folder.c (imap_rescan): Don't g_strdup() the uid into the trigger_event call. * providers/pop3/camel-pop3-folder.c (pop3_get_message_stream): Set the total bytes expected to 0 if the response is empty too. (pop3_refresh_info): Return of there was an error generating pop3 uids. * camel-mime-utils.c (header_raw_check_mailing_list): regfree the pattern before continue'ing. This was showing up quite a bit under purify. 2001-08-09 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-utils.c (imap_parse_string_generic): A quoted string cannot contain \n's so check for those as well. (imap_atom_specials): Update this to fix bug #6553. 2001-08-08 Jon Trowbridge <trow@ximian.com> * camel-filter-search.c (get_source): Remove trailing stuff from URLs before returning. (Done in an evil way, not in the good way suggested by fejj. :)) (Bug #4876) 2001-08-08 jacob berkman <jacob@ximian.com> * camel-search-private.c (camel_search_header_match): check for NULL name and addr (fixes a bug i was about to report) 2001-08-08 Not Zed <NotZed@Ximian.com> * camel-filter-search.c (check_header): Properly determine match type to pass to header_match. (address_matches_exactly): Removed, effectively added to camel_search_header_match. * camel-folder-search.c (check_header): Properly determine the match type to pass to header_match. * camel-search-private.c (camel_search_header_match): Add a new parameter 'type' which is the type of header we're matching against. ASIS means utf8 format, ADDRESS means an internet address ('formatted'), ADDRESS_ENCODED means a raw address header, ENCODED means rfc 2047 encoded text. (header_match): Move original logic here, have search_header_match call it as appropriate for the 'type' of match. 2001-08-07 Not Zed <NotZed@Ximian.com> * camel-session.c (camel_session_class_init): Only init the vee provider struct once (if we're subclassed this will get called multiple times). * camel-object.c (obj_finalize): Removed a bit of a debug that crept in with jacob's poolv patch (?). 2001-08-07 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-search.c (address_matches_exactly): New function to do danw's address "is" comparison (as specified in bug #5886. (check_header): Special-case address-type headers when we are doing an exact match to use the address_matches_exactly function. 2001-08-06 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-folder.c (pop3_get_message_stream): Modify to take a headers_only argument which will then only fetch message headers. (pop3_generate_uids): Pass in TRUE for headers_only. (pop3_get_message): Pass in FALSE for headers_only and move the camel_operation() stuff out of get_message_stream and in to here instead. (pop3_generate_uids): We also need to be using i+ 1 as the id since we are starting at 0. 2001-08-06 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-folder.c (pop3_get_message_stream): New internal function to get a message stream by message sequence id (so that we can share code between pop3_get_message and pop3_generate_uids). (pop3_get_message): Use pop3_get_message_stream. (pop3_generate_uids): New function to generate UIDs based on the message - this fixes bug #1256. (pop3_refresh_info): If the server doesn't support UIDL, call pop3_generate_uids to get our list of UIDs. * providers/imap/camel-imap-store.c (get_folder_info_online): Don't strstr for noselect=yes, that's just plain broken. 2001-08-06 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (imap_rescan): Fix off-by-one error in cleaning up deleted messages at the end of the folder. Fixes bug #5585 and maybe others. * providers/imap/camel-imap-store.c (get_folder_info_online): Do a refresh_info on the selected folder before looking at its unread count. Should fix bug #4944. 2001-08-06 Peter Williams <peterw@ximian.com> Completely hide the namespace from everything external to the IMAP code, which Dan W says is the way it should be. * providers/imap/camel-imap-command.c (imap_command_strdup_vprintf): Add a new %F argument, which is like %S but will add the namespace (for folder names). (camel_imap_command): Use %F here. * providers/imap/camel-imap-utils.c (imap_parse_list_response): Changed to strip out the namespec when returning *folder. In order to do this we need to be passed the CamelImapStore. (imap_concat): Move to here from camel-imap-store.c, un-static (imap_namespace_concat): New function, adds the namespace to the folder name, unless it's INBOX. * providers/imap/camel-imap-utils.h: Prototypes. * providers/imap/camel-imap-store.c (imap_connect_online): Extra arg to imap_parse_list_response. (imap_connect_offline): Here too. (get_folder_status): Use %F. (get_folder_online): Here too. (delete_folder): Here too. (create_folder): Here too, and arg to imap_parse_list_response. (parse_list_response_as_folder_info): Arg to i_p_l_r. (get_subscribed_folders_by_hand): Use %F. (get_folders_online): Here too. (get_folder_info_online): Instead of checking for NULL @name, check for name = NULL or "", and set to "" instead of namespace. Pass "" instead of namespace to camel_folder_info_build. (subscribe_folder): Use %F. (unsubscribe_folder): Here too. * providers/imap/camel-imap-folder.c (imap_get_full_name): This now just returns folder->full_name. (do_append): Use %F (do_copy): Here too. 2001-08-06 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_verify): Always do --no-auto-key-retrieve. 2001-08-03 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_verify): Minor changes to the iconv code so that it does the check in the return value of iconv() just like the code in camel-mime-utils.c in case this fixes Bo Rosen's gpg bugs (but I doubt this will change anything?). 2001-08-03 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_connect): First check to make sure that service->url->authmech is non-NULL before passing it into strcmp to avoid a segfault. * camel-pgp-context.c (pgp_decrypt): Sigh, if gpg returns '1' it means that the key is bad but it did succeede in decrypting the block so to make users happy (even though I find this a questionable action) we are going to ignore the error and show it to the user anyway. This fixes bug #6136. 2001-08-03 Not Zed <NotZed@Ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_connect): special case popb4smtp auth before we try and connect, and do the magic here first. 2001-08-02 Not Zed <NotZed@Ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_connect): Check for POPB4SMTP separate to the esmtp auth list. (smtp_auth): If creating the sasl object means it is already authenticated, then exit early. Sort of 'clean hack' to help popb4smtp work. (smtp_auth): Unref the sasl object, clean up a memleak i think. * providers/smtp/camel-smtp-provider.c (camel_provider_module_init): Added POPB4SMTP auth type. * camel-sasl.c (camel_sasl_authtype): Added POPB4SMTP type. * camel-sasl-popb4smtp.c: New file for pop before smtp 'authentication'. * Makefile.am (libcamel_la_SOURCES, HEADERS): Add camel-sasl-popb4smtp.[ch]. 2001-08-01 Not Zed <NotZed@Ximian.com> * providers/local/camel-mbox-folder.c (mbox_lock): If we fail to get the lock, make sure we close our locking fd, and reset it. Fixes the crash part of 5095. * providers/imap/camel-imap-folder.c (camel_imap_folder_changed): Slightly re-arranged filter/change notification logic. If we;re filtering, freeze the folder till we're finished to smooth the ui for new mail. 2001-07-31 Not Zed <NotZed@Ximian.com> * camel-folder.c (camel_folder_init): Chagned mutexes to e_mutexes, folder lock recursive. This fixes the imap filtering bug, or at least some cases of it. (camel_folder_finalize): Free mutexes. * camel-private.h (CamelFolderPrivate): Changed mutexes to e_mutexes. 2001-07-25 Not Zed <NotZed@Ximian.com> * camel-folder.c (camel_folder_change_info_*): Change the return if fail's into asserts. (camel_folder_change_info_free): Don't let it accept a NULL argument. This isn't g_lib so we dont have to. 2001-07-27 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_decode_param): Fix the fix that fixes broken mailer behavior. We want to make sure that the user's locale charset is non-NULL before we pass it off to iconv_open(). If it *is* NULL, we want to use iso-8859-1. 2001-07-26 Jeffrey Stedfast <fejj@ximian.com> * camel-url.c (camel_url_free): Allow the url to be NULL just like the libc free convention. 2001-07-26 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-command.c (camel_imap_command_start): Send an IMAP command, but don't wait for responses. (camel_imap_command_response): Read a single line of response from the server. (camel_imap_command, etc): Reimplement in terms of the new code. * providers/imap/camel-imap-folder.c (imap_rescan): Use camel_imap_command_start and camel_imap_command_response, and call camel_operation_progress after each line read from the server. (imap_update_summary): Likewise, although with more fudging on the precentages... Also, fix this so that if none of the new messages are cached, it only does a single FETCH, and if some of them are cached, it does two FETCHes (one to get the UIDs, FLAGS, and SIZEs, and another to get the bodies of all of the messages that aren't cached now that it knows the relevant UIDs). This should speed up startup a bunch (especially if you have high bandwidth but also high latency to the IMAP server). 2001-07-25 Dan Winship <danw@ximian.com> * camel-mime-utils.c (mail_mlist_magic): Add another Sender pattern. 2001-07-24 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_changed): Change the break into a continue, we should process as many as we can find. * camel-folder.c (camel_folder_move_messages_to): If we have no uid's to copy, exit here before going any further. Some code internally [imap] assumes there are uids to copy otherwise it can segfault. (camel_folder_copy_messages_to): Same. 2001-07-24 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_decode_param): Handle broken mailers that send unencoded 8bit header params. And there was much rejoicing. Rah. * camel-url.h (CAMEL_URL_HIDE_ALL): New #define, and there was much rejoicing. Rah. 2001-07-24 Not Zed <NotZed@Ximian.com> * camel-operation.c (camel_operation_register, unregister): Added some warnings for bad cases. 2001-07-23 Not Zed <NotZed@Ximian.com> * camel-operation.c (camel_operation_register): Only insert a hash entry if we haven't already. (camel_operation_unregister): Only remove the has entry if the id is a real thread. 2001-07-23 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (do_move): Slight fix for when source == destination (we don't want to do this action because on-demand filtering would then delete the messages it supposedly "moved"). 2001-07-23 Not Zed <NotZed@Ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_auth, smtp_helo, send_to): Added some operation progress reporting. Actual data transfer is 'tricky' because of the layers used. (smtp_auth): Instead of checking exception_is_set, use challenge==NULL to test if sasl_challenge_base64 failed. * providers/local/camel-mbox-summary.c (mbox_summary_sync): Remove peters changes below and put in a better fix. These functions already return -1 on error, just use that, and not worry about building our own exception redundantly. * camel-service.c (camel_get_host_byname): Turn the progress into a transient event. 2001-07-20 Peter Williams <peterw@ximian.com> * providers/local/camel-mbox-summary.c (mbox_summary_sync): More usage fixes for CamelException. Check our own exception for summary_update and xfer it out if an error occurred. 2001-07-20 JP Rosevear <jpr@ximian.com> * Pull up test fixes to get them building again 2001-07-20 Not Zed <NotZed@Ximian.com> * camel-charset-map.c (camel_charset_get_iconv_friendly_name): renamed to camel_charset_to_iconv() to make it just a little more consistent and not so long & fixed callers. 2001-07-19 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-filter-charset.c (camel_mime_filter_charset_new_convert): Convert to the iconv-friendly charset names. * providers/imap/camel-imap-store.c (create_folder): Fixed a compiler warning about returning without a value in a non-void function. Blah. * camel-mime-part.c (process_header): Pass the locale charset as the default_charset to header_decode_string(). * camel-folder-summary.c (camel_folder_summary_format_string): Pass the locale charset as the default_charset to header_decode_string(). (content_info_new): Same. * camel-mime-message.c (process_header): Pass the locale charset as the default_charset to header_decode_string(). * camel-mime-utils.c (append_8bit): New helper function who's purpose is similar to append_latin1() but for 8bit text that we are assuming is not latin1. (header_decode_text): Now takes a default_charset parameter and calls append_8bit when appropriate. (header_decode_string): Also takes a default_charset parameter now. (header_decode_mailbox): Pass NULL as the default_charset to header_decode_string(). 2001-07-19 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_verify): Modified to treat the return value from camel_charset_locale_name() as a const char*. * camel-sasl-digest-md5.c (digest_response): Modified to treat the return value from camel_charset_locale_name() as a const char*. * camel-charset-map.c (camel_charset_locale_name): Modify to return const char* by returning the static locale_charset which is created inside of camel_charset_map_init(). (camel_charset_map_init): Find the locale charset here and set the static variable. 2001-07-19 Peter Williams <peterw@ximian.com> Policy change: NULL url's are no longer allowed in CamelFolderInfos. They used to signify that the folder was, in IMAP jargon, NoSelect; now the same effect is achieved by adding a "noselect=yes" parameter to the end of the URL. As far as I know, IMAP is the only affected provider. * providers/imap/camel-imap-store.c (delete_folder): New function. Implement folder deletion. (camel_imap_store_class_init): Set the delete_folder class function here. (get_folder_status): New function. Utility wrapper around the STATUS command. (create_folder): If the parent folder is NoSelect but is empty, delete it and recreate it as a a subfolder-containing folder. If it is NoSelect but contains messages, set an exception. (parse_list_response_as_folder_info): Always set the FolderInfo's URL, but add a NoSelect parameter if it isn't selectable. (get_folder_info_online): Change logic of removing the namespace to reflect URL change. Same for logic of checking unread counts. (get_folder_info_online): Use get_folder_status to simplify this. * camel-store.c (camel_folder_info_build): When creating dummy parents, copy the child's URL and set the NoSelect parameter. 2001-07-19 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): No need for `acharset' anymore. (check_html_charset): Return a const char* and also use camel_charset_get_iconv_friendly_name() internally. * camel-mime-utils.c (rfc2047_decode_word): Remove my iso8859-1 -> iso-8859-1 hack and use camel_charset_get_iconv_friendly_name() instead. (rfc2184_decode): Use camel_charset_get_iconv_friendly_name() * camel.c (camel_init): Call camel_charset_map_init(). * camel-charset-map.c (camel_charset_map_init): New function to initialize the charset-equivalent lookup table. To be called by camel_init(). (camel_charset_get_iconv_friendly_name): New function to try and convert a charset into something that iconv is more likely to accept. 2001-07-19 Peter Williams <peterw@ximian.com> * Revert 7/11/2001 patch for IMAP INBOX filtering at NotZed's request. 2001-07-17 Peter Williams <peterw@ximian.com> Clean up some exception misusage. * providers/imap/camel-imap-command.c (camel_imap_command): Use our own internal exception for sending the string and transfer it to @ex if anything goes wrong. (imap_read_response): Use our own internal exception for reading the untagged responses and blah blah blah. * camel-session.c (get_service): Use our own internal exception when constructing the service and transfer it to @ex if anything goes wrong. * camel-remote-store.c (remote_recv_line): Instead of having gboolean exception, use our own internal exception and copy it to @ex if anything goes wrong. * camel-store.c (store_sync): Create an internal exception because sync_folder() checks it for validity. Transfer it to @ex when done. * camel-exception.c (camel_exception_get_description): If @ex is NULL, complain - passing NULL exceptions to Camel is okay, but there should be no circumstances under which they're then examined. (camel_exception_get_id): Same here, (camel_exception_xfer): NULL-protect and warn if transferring from a NULL exception. 2001-07-17 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (init_trash): Use CAMEL_VTRASH_NAME. * camel-vtrash-folder.h: #define CAMEL_VTRASH_NAME 2001-07-16 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-command.c (camel_imap_response_free): Given: * 4 EXISTS * 1 EXPUNGE We have to pass 3, not 4 to camel_imap_folder_changed for the exists count. Fixes ximian bug #2112 (finally!). 2001-07-14 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_changed): If the message info for an expunged message is NULL, then just break out - maybe a message was expunged that we were never notified even existed? 2001-07-13 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (rfc2047_decode_word): Added a hack to convert charsets in the format iso8859-1 to iso-8859-1 because it seems to be more iconv friendly. It has been reported that on some systems, iconv doesn't know about iso8859-1 while it *does* know about iso-8859-1. See bug #4530. 2001-07-13 Peter Williams <peterw@ximian.com> * Makefile.am (install-exec-local): Let people install as non-root, but give them a bigass warning so they're not allowed to complain when it doesn't work right. * camel-remote-store.c (sync_remote_folder): New function: hash table callback. (remote_disconnect): If cleanly disconnecting, sync our folders. Fixes deadlocks on exit (folders syncing after store disconnects) and also makes sense. 2001-07-13 Jeffrey Stedfast <fejj@ximian.com> * camel-uid-cache.c (camel_uid_cache_new): We now use a structure as the bucket data rather than just a cache level so set the save state to TRUE. (maybe_write_uid): We only save the uid if the cache levels are the same *and* if the save state is TRUE. (free_uid): Free the state value. (camel_uid_cache_get_new_uids): New uids that get added to the cache start with a save state of FALSE. (camel_uid_cache_save_uid): Set the save state of the uid to TRUE. What should we do if the uid isn't already in the cache? Currently I make it add the uid, but maybe it shouldn't? * providers/imap/camel-imap-folder.c (imap_filter_timeout): Update to reflect CamelFilterDriver API changes. * camel-filter-driver.c (camel_filter_driver_filter_folder): Take a cache argument so we can tell the cache whether or not the uid should be saved (meaning we have successfully filtered it). 2001-07-12 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_message): Okay, I've decided that that user is just on crack. We don't want to allow filtering of deleted messages or we could get some unexpected behavior. (do_move): Oops. My last fix needs to be after the if-statement, not inside it. (do_copy): And here too. 2001-07-12 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (do_move): Only set p->copied and p->deleted if the messages are copied without an exception. (do_copy): Same - but obviously we don't set the p->deleted here at all. (camel_filter_driver_filter_message): Since users complained that they couldn't filter deleted messages for "safe keeping" (WTF? safe keeping? why were they deleted in the first place then??), I have taken out the code that optimized filtering by not allowing deleted messages to be filtered. This fixes bug #4473. Note: Users have a warped sense of logic. 2001-07-12 Not Zed <NotZed@Ximian.com> * camel-mime-part-utils.c (extract_metatag_charset): Removed. (check_html_charset): Replaced with this. (simple_data_wrapper_construct_from_parser): Call check_html_charset if we dont have a charset on the content-type, and we have text/html data. (check_html_charset): We also need to do qp/base64 decoding ourselves, sigh. * camel-mime-utils.c (html_meta_param_list_decode): Removed. This was very wrong, the rules for mail headers vastly different from rules for decoding html elements. (rfc2184_decode): Move the malloc inside the iconv_open worked, otherwise we have a memleak. * camel-mime-filter-html.c (camel_mime_filter_html_finalize, init, run, reset): Changed to use camelhtmlparser, and fixed a tiny memleak. * camel-html-parser.c: Made the html indexer tokeniser re-usable. ONLY TO BE USED INTERNAL TO CAMEL HOWEVER. (tokenise_step): Slight fix to non-quoted values. * camel-folder-summary.c (camel_folder_summary_info_new_from_message): Removed some code i wasn't supposed to commit. * providers/local/camel-local-summary.c (local_summary_add): Only set info->size if it is not zero. (local_summary_add): If we dont get a size from the info passed in, calculate it using a null stream write. Should do #4392. 2001-07-11 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part-utils.c (extract_metatag_charset): New function to extract a meta-tag charset value if it exists. (simple_data_wrapper_construct_from_parser): Along the same lines as the code I previously ripped out, but this time use the mime-parser's seek ability to help us along. Currently I read up to a 2k buffer size - this is probably overkill, 1k is probably plenty. * camel-mime-utils.c (html_meta_param_list_decode): When we get to an `=', we must skip past it before trying to grab the param value. duh. 2001-07-11 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): Ripped out my code since it was never being used since the mime parser is not using a seekable stream (ever?) which pretty much means my code needs to be done someplace else. 2001-07-11 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_changed): Run the filter in its own thread to prevent locking issues. (imap_filter_timeout): New function that does the filtering called as a CamelSession timeout. 2001-07-11 Not Zed <NotZed@Ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_mbox): Create the messageinfo itself, so we can properly set the size. * camel-movemail.c (camel_movemail_solaris): Write out the from line between each message. 2001-07-10 Not Zed <NotZed@Ximian.com> * providers/local/camel-local-summary.c (local_summary_add): Copy the size across from the source message info if supplied. * camel-stream-null.c: Added a 'written' member which keeps track of how much has been written to the stream. * camel-movemail.c (camel_movemail): If we have BROKEN_SPOOL defined, then use the solaris movemail to quote from lines that sendmail didn't. (camel_movemail_solaris): Compile this in if BROKEN_SPOOL defined. 2001-07-10 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (html_meta_param_list_decode): New function to parse an HTML meta-tag. * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): If the Content-Type did not contain a charset parameter and it's also a text/html part, we have 1 last place to look - in the META html tags. *sigh* * camel-mime-message.c (camel_mime_message_get_source): s/gint/unsigned since that's what it should be. 2001-07-09 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_sign): Forget the passphrase if the user has set that option. (pgp_clearsign): Same. (pgp_encrypt): And here... (pgp_decrypt): And finally here. (camel_pgp_context_new): Take a `remember' argument. 2001-07-09 Not Zed <NotZed@Ximian.com> * providers/local/camel-maildir-summary.c (maildir_summary_sync): Oops, e_poolv_set free's it for us, so dont double-free here. 2001-07-06 Not Zed <NotZed@Ximian.com> * camel-vee-folder.c (camel_vee_folder_finalise): Changed so we dont have the list changing under us, just going to the next node before we call a function that might change the list is potentially dangerous (slight mod of peters fix). Hmm, i think it would've double-unref'd it too(?) 2001-07-08 Chris Toshok <toshok@ximian.com> * camel-tcp-stream-openssl.c (camel_tcp_stream_openssl_finalize): openssl_table is gone. we now store/get the stream from the SSL_CTX's app_data. (stream_read): rework the non-blocking case to account for SSL possibly buffering data (in which case select will block even though data is ready to be read), and to account for FreeBSD's strange behavior of returning -1/EAGAIN even though select said the fd was ready to be read. (ssl_verify): openssl_table is gone. (open_ssl_connection): set the SSL_CTX's app_data to be the stream, remove the openssl_table code. 2001-07-06 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_encode_param): Don't allow in to be NULL, so instead of doing if (in == NULL) return;, make it a g_return_val_if_fail and later we can make it an assert or something. * providers/local/camel-maildir-store.c (get_inbox): Fixes bug #1138. * providers/local/camel-mbox-summary.c (camel_mbox_summary_build_from): Make the `day-of-month' digit take up 2 chars by using "%2d". Fixes bug #3989 for lame mailers like Pine. * camel-pgp-context.c (crypto_exec_with_passwd): Set the pipe fd's to nonblocking. 2001-07-06 Peter Williams <peterw@ximian.com> * camel-object.c (camel_object_unref): Make sure that after the finalization event has happened and the finalization functions have been called that the object still has a zero refcount. 2001-07-06 Joe Shaw <joe@ximian.com> * providers/nntp/camel-nntp-grouplist.c: Add locking to camel_nntp_get_grouplist_from_server(). * providers/nntp/camel-nntp-resp-codes.h: Added NNTP_EXTENSIONS_SUPPORTED (202). * providers/nntp/camel-nntp-store.c (camel_nntp_store_get_extensions): Check for both NNTP_LIST_FOLLOWS and NNTP_EXTENSIONS_SUPPORTED from a LIST EXTENSIONS request. (Dunno if NNTP_LIST_FOLLOWS ever comes out of this, but that's what was already there...) Also, put some locking around it. (finalize): e_mutex_destroy() the command lock (camel_nntp_store_init): e_mutex_new() the command lock. * providers/nntp/camel-nntp-store.h: Add locking macros. 2001-07-05 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.c (camel_message_info_string): Added an assert to make sure that `mi' isn't NULL. (camel_message_info_set_string): Same. * providers/imap/camel-imap-command.c (camel_imap_response_free): Create and use a temporary CamelException for use with camel_imap_folder_changed. 2001-07-05 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_verify): Send the --no-auto-key-retrieve argument to gpg if we are in offline mode so that we don't have to worry about gpg hanging if it can't connect to the key servers, because now it shouldn't even attempt to. 2001-07-05 Chris Toshok <toshok@ximian.com> * camel-tcp-stream-openssl.c (ssl_verify): use X509_STORE_CTX_get_ex_data to get at the SSL* - don't use the X509_STORE_CTX to look up our stream, since it's not what we used to insert our stream into the hashtable. (open_ssl_connection): insert the stream into the hashtable before calling SSL_connect, as this can cause ssl_verify to be called, and we need to look up the stream there. remove the stream from the hashtable if there's an error connecting. (stream_connect): pass the CamelTcpStreamOpenSSL* to open_ssl_connection since it handles the hashtable stuff. remove hashtable stuff from here. 2001-07-05 Jeffrey Stedfast <fejj@ximian.com> * camel-folder.c (camel_folder_copy_messages_to): if source == dest, just no-op. 2001-07-05 Peter Williams <peterw@ximian.com> * camel-vee-folder.c (camel_vee_folder_finalise): Move this before the camel_vee_folder_remove_folder because that function modifies p->folders messing up our iteration. (camel_vee_folder_finalise): Don't unref our summary; camel-folder now does this. * camel-object.h (CamelObject): Add a 'destroying' flag to CamelObject. * camel-object.c (obj_init): Clear 'destroying'. (camel_object_unref): If 'destroying' then do not send the finalize event and do not call finalize functions. Otherwise, set destroying so just in case we get refed (eg event code) we don't get doubly finalized. 2001-07-05 Not Zed <NotZed@Ximian.com> * camel-mime-filter-basic.c (filter): Fix the assertion slightly, if we have a little bit of input the output size could be larger, since we store upto 3 chars in the state. * camel-folder-summary.c (camel_folder_summary_info_new_from_message): When indexing a new record, create a pseudo word 'ibexindexed' so we can always tell that a message has already been idnexed, even if it contains no words of its own. Things like maildir use this check to see if its already been processed, and it matters if it is incorrect in this case (not just wasted cycles). (camel_message_info_dup_to): Assign the to->strings from the e_poolv_cpy() call, since it may allocaote a new poolv if the lengths do not match. (camel_folder_summary_info_new_from_parser): And same here. * providers/local/camel-maildir-summary.c (maildir_summary_sync): Changed the logicfor epoolv code to be different, we dont need to update hash references or any tricky stuff. (maildir_summary_check): Samehere. * camel-folder-summary.h: Removed include of e-memory.h. 2001-07-03 Joe Shaw <joe@ximian.com> * providers/nntp/camel-nntp-resp-codes.h: Added NNTP_NO_PERMISSION (502) to the list of response codes. * providers/nntp/camel-nntp-store.c (nntp_store_connect): If we receive a NNTP_NO_PERMISSION, don't get into an infinite loop trying to reconnect, get extensions, etc. Just give up and return FALSE. 2001-07-03 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_clear): Doh. Don't remove things from the hash table while foreach'ing it. (And can't use foreach_remove either because we have to remove them in a weird order). Fixes #3618. * providers/imap/camel-imap-folder.c (imap_get_message): If the server returns OK from the FETCH BODY, but there's no parseable BODY response, it's probably because there's an UN-parseable BODY response, implying the message is badly formatted, MIMEwise. In that case, fall back to fetching the message as a single part. 2001-07-02 Sam Creasey <sammy@oh.verio.com> * providers/nntp/camel-nntp-folder.c: Implemented nntp_folder_search_by_expression and nntp_folder_search_free. Basic search functionality e.g. unread marking now works for NNTP folders. * camel_filter_search.c (get_size): Added get-size sexp directive to get the size of a message for filters. * providers/nntp/camel-nntp-folder.c (camel_nntp_folder_new): Always check with the NNTP server after summary load -- this function now always expires old summary entries and syncs with the news server. * providers/nntp/camel-nntp-utils.c (camel_nntp_get_headers): Only fetch headers for articles not already logged in the summary file. * providers/nntp/camel-nntp-grouplist.c (camel_nntp_get_grouplist_from_*): change from g_list_append() to g_list_prepend() + g_list_reverse. Traversing 40,000 element linked lists sucks. * providers/nntp/camel-nntp-store.c (camel_nntp_command): Should the NNTP connection die with CAMEL_EXCEPTION_SERVICE_NOT_CONNECTED, make a single retry attempt. Timing out the NNTP link is less painful this way. 2001-07-02 Peter Williams <peterw@ximian.com> * README (Introduction): Add comment noting that Camel actually exists now :-) 2001-07-02 Dan Winship <danw@ximian.com> * camel-session.c (camel_session_get_filter_driver): new method to get an application-initialized filter driver. * camel-filter-driver.c (camel_filter_driver_new): Remove the get_folder function and data args from here... (camel_filter_driver_set_folder_func): ...and add this function to set/change them. * providers/imap/camel-imap-folder.c (camel_imap_folder_new): If this folder is INBOX and we're filtering INBOX, set a flag on the folder for later. (imap_update_summary): Add another argument (GPtrArray *recents), and if it's non-NULL, add the uids of any \Recent new messages to it. (camel_imap_folder_changed): If doing filtering in this folder, create a recents array and pass it to imap_update_summary. Then get a filter driver and use it to filter the recent messages. * providers/imap/camel-imap-summary.h: * providers/imap/camel-imap-utils.c (imap_parse_flag_list): Add support for the \Recent flag. * providers/imap/camel-imap-provider.c (imap_conf_entries): enable the "filter" option. * camel-types.h: add CamelFilterDriver typedef here 2001-07-02 Not Zed <NotZed@Ximian.com> * camel-lock-client.c (camel_lock_helper_init): properly return error if we can't create pipes. * camel-folder-summary.c (summary_build_content_info): re-enable html indexing. * camel-mime-filter-html.c: Completely re-implemented using a custom parser. 2001-07-02 Dan Winship <danw@ximian.com> * tests/misc/url.c (main): Change a camel_exception_clear to camel_exception_init to fix a sometimes-crash 2001-06-29 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c: Added Originator: header as a header to look for when looking for the mailing list. * tests/mime-filter/Makefile.am: Build the test-crlf test program. * tests/mime-filter/test-crlf.c: New test suite for the crlf filter. * camel-mime-filter-crlf.c (filter): Fixed to correctly encode and decode dots. 2001-06-28 Peter Williams <peterw@ximian.com> * camel-mime-filter-stripheader.c: New file. Filter that strips a header from mime output. * camel-mime-filter-stripheader.h: New file. Header for the above. * providers/smtp/camel-smtp-transport.c (smtp_data): Use the stripheader filter to remove the "Bcc" header. * Makefile.am: Add the stripheader files. * tests/lib/Makefile.am (INCLUDES): Get this to compile again. * tests/mime-filter/test-stripheader.c: New file. Test suite for the CamelMimeFilterStripHeader. * tests/mime-filter/Makefile.am: New test section: mime filters. 2001-06-28 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_decode_param_list): Fix a memory leak. Also if the decoded value is NULL, that means it wasn't encoded so we don't want to toss the old value. (header_param_list_format_append): Correctly wrap long parameter values according to rfc2184. 2001-06-28 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (imap_refresh_info): If the folder is frozen, don't refresh, just record that we need to do it later. (imap_append_online): If the APPEND doesn't trigger an immediate EXISTS response (because the folder isn't the selected folder, or because the server doesn't do that until the next command), call imap_refresh_info on the folder. (imap_copy_online): Likewise. (Replacing the unconditional NOOP that was there before, which absolutely killed filter performance by forcing the IMAP provider to switch back and forth between folders after every copy or move.) (imap_thaw): If the folder needs a refresh, do it. * camel-folder.c (camel_folder_is_frozen): New method 2001-06-27 Jeffrey Stedfast <fejj@ximian.com> * camel-transport.c (camel_transport_init): Initialize the private send_lock. (camel_transport_finalize): Free the private send_lock. (camel_transport_get_type): Set the init and finalize functions. (camel_transport_send): Lock the transport. (camel_transport_send_to): Same. * camel-private.h: Add CAMEL_TRANSPORT_(UN)LOCK macros. 2001-06-27 Jeffrey Stedfast <fejj@ximian.com> * tests/folder/test9.c (main): Updated to match the current API. * tests/folder/test3.c: #include <gtk/gtk.h> since we use gtk_init(). * camel-internet-address.c (camel_internet_address_encode_address): Use the new header_address_fold. * camel-mime-utils.c: Removed some old #if 0'd code of mine. (rfc2047_encode_word): If enclen is 0, don't write an encoded word token (=?iso-8859-7?Q??= would be an invalid token). (header_address_fold): New function to wrap address headers - header_fold() was force-wrapping rfc2047 encoded words which was making the test suite fail. The *real* solution, however, is to not create rfc2047 encoded words longer than 72 chars. 2001-06-26 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (open_folder): Since we want an error opening a folder to result in the message being copied to Inbox (assuming no other filters get a chance to `move` it), don't record any exceptions that may occur in this function. 2001-06-26 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_selected, imap_append_offline, imap_append_online, camel_imap_folder_changed): lock the cache around accesses * providers/imap/camel-imap-store.c (get_folder_online): ref the newly-created folder (as current_folder) before calling camel_imap_folder_selected, in case that needs to do something that causes another folder to become current... * camel-service.c (camel_service_get_url): D'oh. s/FALSE/CAMEL_URL_HIDE_PASSWORD/ 2001-06-25 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-search.c (get_source): g_strdup() the message source here or we will have problems later when we go to free the result ;-) 2001-06-25 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.c (try_auth): Fix this: IMAP doesn't allow an extra argument to AUTHENTICATE to avoid a useless round trip like some other SASL bindings do. * providers/imap/camel-imap-command.c (imap_read_response): Deal with IMAP servers that (incorrectly) return "+\r\n" rather than "+ \r\n" for an empty continuation response. (camel_imap_response_extract_continuation): Likewise. 2001-06-21 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (word_types_compatable): New function that decides if 2 word types are mergeable. An atom and a qstring are mergeable; 2 qstrings are mergeable; and 2 encoded words are mergeable. (header_encode_phrase_merge_words): If 2 words are merged, the new word type is the MAX of the combined types. This means that if we merge a qstring and an atom, the resulting word type is a qstring. * camel-internet-address.c (internet_format): s/g_string_sprintfa/g_string_append since this makes more sense in this particular case. (internet_encode): Same here. 2001-06-21 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_send): Use camel_address_length() rather than casting and accessing data members. 2001-06-20 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_raw_check_mailing_list): If the first char of the mailing-list name is '<', chop it off to make Ettore happy. Fixes bug #2153. 2001-06-19 Not Zed <NotZed@Ximian.com> * camel-lock-helper.c (unlock_id): Fixed a bug when not running setuid it wouldn't unlock. * camel-movemail.c (camel_movemail): use new locking daemon, also return an error code. 2001-06-18 Jeffrey Stedfast <fejj@ximian.com> Note: Except for the info_free(), the NULL checks are g_return's. I felt that since g_free() handles NULL that our _free() functions should also. * camel-folder.c (camel_folder_change_info_free): Check to make sure that the info pointer isn't NULL. (camel_folder_change_info_change_uid): Same. (camel_folder_change_info_changed): Same. (camel_folder_change_info_remove_uid): Same. (camel_folder_change_info_add_uid): Same. (camel_folder_change_info_build_diff): Same. (camel_folder_change_info_cat): Same. (camel_folder_change_info_add_source): Same. (camel_folder_change_info_add_source_list): Same. (camel_folder_change_info_add_update): Same. (camel_folder_change_info_add_update_list): Same. 2001-06-18 Dan Winship <danw@ximian.com> * tests/stream/Makefile.am (LDADD): * tests/smime/Makefile.am (LDADD): * tests/misc/Makefile.am (LDADD): * tests/message/Makefile.am (LDADD): * tests/folder/Makefile.am (LDADD): Remove DB3_LDADD * Makefile.am (install-exec-local): Set camel-lock-help setgid or setuid as needed 2001-06-18 Not Zed <NotZed@Ximian.com> * camel-folder-summary.c (camel_folder_summary_info_new_from_parser): Only create a missing uid if we have indexing turned on. * camel-lock-helper.c (setup_process): Function to setup process/sanity/security checks. Change to the real uid as soon as we can. (lock_path): First try to lock as the real uid, if that fails, try the root uid. (unlock_id): Unlock as the uid we created the lock as. * Makefile.am (INCLUDES): Added -DCAMEL_SBINDIR for lock helper location. * providers/local/camel-spool-folder.c (spool_lock): Implemented, using lock helper locking. Need to work out if the locking requires a root created lock? (spool_unlock): Likewise. 2001-06-15 Not Zed <NotZed@Ximian.com> * camel-lock-helper.c: Setuid Lock helper process. Creates and manages .locks, keeping them active, removing them, etc. What real perms it needs is a little system dependent. 2001-06-14 Not Zed <NotZed@Ximian.com> * providers/local/camel-maildir-store.c (get_folder_info): Implement. (scan_dir): Does the work of scanning for maildir directories. 2001-06-13 Not Zed <NotZed@Ximian.com> * providers/local/camel-spool-store.c (get_folder_info): Implemented, just returns a hardcoded INBOX folder. (free_folder_info): implemented, free's the 1 possible level of folder info. * providers/local/camel-spool-folder.c (camel_spool_folder_construct): Set the real unread message count on the folder_created thing. 2001-06-15 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_max_uid): Return the largest real UID in the cache (as an integer). Eventually to be used for an optimization in the new-message fetch code. (cache_put): Keep track of max_uid. 2001-06-14 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c: Added Mailing-List header regex so that we can do mlist magic on that header. 2001-05-28 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_decode_param): Properly handle correct and incorrect (for the common case, ie Outlook and Netscape/Mozilla) rfc2184 values. (header_decode_rfc2184_param): Get the param name and see if we should expect a rfc2184 parameter value. (rfc2184_decode): Decode a rfc2184 value. (hex_decode): hex decode a string. (header_decode_param_list): Handle rfc2184 encoded parameters (ie parameters that have been split and perhaps encoded). (header_param_list_format_append): Encode the value before seeing if it will fit on the line. If the value does get encoded, be sure to put a '*' before the equal-sign. (header_encode_param): New function to rfc2184 encode a parameter value (maybe it should be renamed?) 2001-06-14 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part.c (init_header_name_table): Don't try to format the Reply-To header - we don't want to wrap the reply-to address. 2001-06-12 Not Zed <NotZed@Ximian.com> * providers/local/camel-spool-*.[ch]: A new provider, for spool mailboxes. Mostly a cut and paste of the mbox code, but not a subclass CamelLocal*. Not tested a lot, doesn't lock yet, use with extreme caution. * tests/lib/folders.c (test_folder_message_ops): Added spool arg, spool folders can't be deleted, renamed, etc. (test_folder_basic): Same. * tests/folder/test2.c (main): Added checks for spool type. * tests/[message|stream|folder|misc|smime]/Makefile.am (LDADD): Added db3 flags, so make check compiles, doesn't run though. 2001-05-24 Not Zed <NotZed@Ximian.com> * providers/local/camel-local-provider.c (camel_provider_module_init): Added spool provider. 2001-06-07 Jon Trowbridge <trow@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_folder): Add a separator between the service url and the folder name, if needed. * camel-filter-driver.c (camel_filter_driver_filter_mbox): Added an argument, so that the original source URI of the mbox can be passed in. This is needed because this function is called post-movemail, so we are never reading from the original mbox anymore. Without the original mbox URI, the X-Evolution-Source tag gets set incorrectly and filter-on-source will fail to work. (camel_filter_driver_filter_message): Also take an extra arg for the original source URI. It is the original URI, not the source URI, that is used for filtering and for setting the X-Evolution-Source tag. 2001-06-05 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (imap_rescan): Don't fetch the UID and FLAGS of messages we don't know about yet, since they'll just get discarded. 2001-06-01 Jeffrey Stedfast <fejj@ximian.com> * tests/smime/pgp.c: Fix to not get into a recursive loop in get_type(). * tests/smime/pgp-mime.c: Same. 2001-05-31 Jeffrey Stedfast <fejj@ximian.com> * camel-smime-context.c: Numerous fixes to get it to compile. * camel.h: Added camel-cipher-context.h, camel-cms-context.h, camel-smime-context.h, and camel-smime-utils.h * camel-cms-context.[c,h]: New virtual class for manipulating cryptographic message syntax messages (like S/MIME). * camel-smime-context.[c,h]: Modified to inherit from the CamelCMSContext class rather than the CamelCipherContext class. * camel-smime.[c,h]: Removed - just use camel-smime-context directly. * camel-smime-utils.[c,h]: New source files. Moved the 2 useful functions from camel-smime.[c,h] into here. 2001-05-30 Dan Winship <danw@ximian.com> * camel-charset-map.c: Redo the BUILD_MAP code to not depend on libunicode. Now it only generates a map of "popular" 8bit encodings. (It's not worthwhile to support obscure encodings, because any mailer that supports them will support UTF8 too. And Chinese and Japanese use mostly the same UTF8 characters so you need to decide between those encodings based on the locale or the charset of the message you're replying to or the input method you used. So this is sufficient for camel_charset_best's use.) * camel-charset-map-private.h: Regenerated. * camel.c (camel_shutdown): Move #ifdefs around to prevent a warning. 2001-05-29 Dan Winship <danw@ximian.com> * camel-disco-diary.c: Code for logging and replaying offline operations. * camel-disco-store.c (disco_construct): Set disco->status here (where we can base it on the session's offline status) rather than at init time. (disco_connect): If we connect online and have a non-empty diary, switch to RESYNCING mode and replay the diary to the server. (disco_get_folder, disco_get_folder_info): Add _resyncing variants. * camel-disco-folder.c (disco_sync, disco_expunge_uids, disco_append_message, disco_copy_messages_to, disco_move_messages_to): Add _resyncing variants to switches. (disco_expunge_uids, disco_append_message, disco_copy_messages_to, disco_move_messages_to): Remove #ifdef'ed out diary code: let the provider do it. (disco_append_message): Redo the append methods to no longer return the UID, since we're no longer doing the logging from here. * providers/imap/camel-imap-store.c (imap_connect_online, imap_connect_offline): Create a CamelDiscoDiary. (imap_disconnect_offline): And free it. * providers/imap/camel-imap-folder.c (camel_imap_folder_selected): If RESYNCING, don't do any sort of checking that the remote folder matches the summary, beyond making sure that the UIDVALIDITY is correct. (imap_rescan): Add a missing camel_folder_summary_info_free when removing a UID from the summary. (imap_expunge_uids_offline): Implement. Fairly simple. (imap_expunge_uids_resyncing): Implement. If the store supports UIDPLUS, we can just use imap_expunge_uids_online. If not, we need to temporarily undelete any messages marked deleted on the server that aren't supposed to get expunged. (imap_append_offline): Implement, using cache and summary operations, and triggering the folder_changed event by hand. (imap_append_resyncing): Implement. Redo imap_append_online a bit in the process to make them able to share more code. (imap_copy_offline): Implement. (imap_copy_online): Move parts of this out into a helper. (imap_copy_resyncing): Implement. In most cases this is just like imap_copy_online, but if you are copying a message that was itself copied or appended into the folder, and the server doesn't do UIDPLUS, it will be necessary to replace at least part of the copy operation with one or more appends. * providers/imap/camel-imap-command.c (imap_read_response): Don't record the current folder in the response when in RESYNCING mode. (This means that EXISTS and EXPUNGE responses won't be processed, which is needed because the summary may not match the folder at this point.) (imap_read_response): On error, call camel_imap_response_free_without_processing, not camel_imap_response_free. * providers/imap/camel-imap-utils.c (imap_uid_array_to_set): Make this work better when operating on UIDs that aren't in the summary. * providers/imap/camel-imap-summary.c (camel_imap_summary_add_offline): New routine used by imap_append_offline and imap_copy_offline to create new summary entries. 2001-05-28 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_set_param): Use g_strcasecmp() because some systems don't have strcasecmp(). Also, when removing a param, make sure to free the param->name too. 2001-05-27 Dan Winship <danw@ximian.com> * tests/lib/messages.c (test_message_read_file): Fix an fd leak * tests/lib/session.c, tests/lib/session.h: a CamelSession subclass for the test programs. * tests/lib/Makefile.am: include session.[ch] * tests/folder/test*.c: Use a CamelTestSession from libcameltest instead of cut+pasting everywhere. * tests/misc/url.c (main): Update for a camel_url_new change at some point. * tests/*/.cvsignore: Add stuff. * camel-mime-utils.c (rfc2047_encode_word): Fix a silly ==/!= mixup. 2001-05-26 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c: Add a "need_rescan" flag saying if we want to rescan the entire folder for flag changes next time it's selected. (camel_imap_folder_init): Set need_rescan TRUE. (camel_imap_folder_selected): If need_rescan is TRUE, call imap_rescan. (imap_refresh_info): Only do a full rescan if need_rescan is TRUE. Otherwise just do a NOOP, making this a MUCH more lightweight operation. Also, don't call imap_rescan directly if the folder isn't selected, since that could end up causing the folder to be scanned *twice* (imap_rescan -> camel_imap_command -> camel_imap_folder_selected -> imap_rescan). (imap_rescan): Set need_rescan FALSE. (imap_sync_online): Don't NOOP if no changes were pushed: the caller will call refresh_info if it wants to poll for changes. Fixes evolution-mail doing lots of unnecessary extra work at quit time. * providers/imap/camel-imap-store.c (refresh_folder_info): Set need_rescan on each folder before calling refresh_info. 2001-05-25 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_auth): Oops. Use a strNcmp to make sure the response begins with a 334 code, not a strcmp. Duh. * providers/local/camel-maildir-folder.c (maildir_append_message): remove the summary info so we are not out-of-sync with the maildir folder. * providers/local/camel-mh-folder.c (mh_append_message): remove the summary info so we are not out-of-sync with the mh folder. * providers/local/camel-mbox-folder.c (mbox_append_message): remove the summary info so we are not out-of-sync with the mbox. 2001-05-24 Jeffrey Stedfast <fejj@ximian.com> * camel-smime-context.c (camel_smime_context_new): Oops, I spelled the get-default-db function name wrong. 2001-05-23 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-local-folder.c (camel_local_folder_construct): s/folder_deleted/folder_created - result of a bad copy/paste. * providers/imap/camel-imap-search.c (imap_body_contains): Optimize the match "" case. 2001-05-21 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (get_folder_info_online): Store the CamelFolderInfo tree that was returned from camel_folder_info_build() in a new variable, 'tree', rather than 'fi' since we later use 'fi' when syncing folders. Not only does this fix a memory leak, but it also fixes the bug where the user would only see the last folder in the folder list and/or it's subfolders. 2001-05-16 Jeffrey Stedfast <fejj@ximian.com> * camel-session.c (camel_session_forget_password): The service can be NULL here too, thanks to Wayne Davis for pointing this out. * camel-smime-context.c (camel_smime_context_new): We don't need to be passed the certdb path anymore. 2001-05-16 Dan Winship <danw@ximian.com> * camel-medium.c (camel_medium_get_headers): New function to return an array of all headers. (camel_medium_free_headers): And free them. * camel-mime-part.c (get_headers, free_headers): Implement this for CamelMimePart. (Works for CamelMimeMessage too.) 2001-05-15 Jeffrey Stedfast <fejj@ximian.com> * camel-smime-context.c (camel_smime_context_finalise): Don't close or free (it wasn't allocated) the certdb. (camel_smime_context_new): If we get a NULL certdb handle, then don't bother trying to create a new certdb handle since NSS_Init* should have done that. 2001-05-15 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Don't use EXTRA_GNOME_*, use the new CAMEL_* variables so we don't link in tons of extra cruft that we don't need. * tests/smime/pkcs7.c: Added to the test suite - tests the S/MIME pkcs7 functions. 2001-05-14 Jeffrey Stedfast <fejj@ximian.com> * camel-smime-context.c: Lots of fixes to get this to compile. * camel-smime.c (camel_smime_part_verify): Updated to pass in a hash argument to camel_smime_verify(). * camel-pgp-mime.c (camel_pgp_mime_part_verify): Update according to the changes in the context API. * camel-pgp-context.c (pgp_verify): Updated to take a CamelCipherHash argument. * camel-cipher-context.c (camel_cipher_verify): Now takes a hash argument since the S/MIME code needs this. * Makefile.am: Add camel-smime-context.[c,h] and camel-smime.[c,h] to the build. * camel-smime.[c,h]: New source files for manipulating S/MIME message parts. 2001-05-14 Jon Trowbridge <trow@ximian.com> * camel-search-private.c (utf8_get): Replicate the semantics of the libunicode utf8 functions by returning NULL in the arg on invalid utf8. * camel-pgp-context.c (pgp_verify): Check for valid utf8, terminate loop if something looks wrong. * camel-mime-utils.c (header_encode_phrase_get_words): Properly check for invalid utf8. (header_encode_string): Properly check for invalid utf8. * camel-charset-map.c (camel_charset_step): Properly check for invalid utf8. 2001-05-14 Jeffrey Stedfast <fejj@ximian.com> * tests/folder/test9.c: * tests/folder/test8.c: * tests/folder/test7.c: * tests/folder/test6.c: * tests/folder/test5.c: * tests/folder/test4.c: * tests/folder/test3.c: * tests/folder/test2.c: * tests/folder/test1.c: * tests/smime/pgp-mime.c: * tests/smime/pgp.c: Subclass CamelSession since we can no longer specify the passwd callback any other way and update code accordingly. 2001-05-11 Dan Winship <danw@ximian.com> * camel-mime-utils.c (header_location_decode): Decode Content-Location, either correctly or Netscape-generated-brokenly. * camel-mime-part.c (camel_mime_part_set_content_location, camel_mime_part_get_content_location, etc): Deal with Content-Location header. 2001-05-11 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_auth): Don't check the initial auth response until we get into the while-loop otherwise we have problems if the SASL mechanism supported a client initiated challenge (like PLAIN and LOGIN do). 2001-05-11 Dan Winship <danw@ximian.com> * camel-stream-null.c (camel_stream_null_new): Make this return CamelStream * (like the other stream new functions) instead of CamelStreamNull * 2001-05-10 Chris Toshok <toshok@ximian.com> * camel-tcp-stream-ssl.c (stream_connect): call SSL_Enable after the SSL_ImportFD and before PR_Connect. Otherwise, NSS aborts during the connect. 2001-05-10 Jeffrey Stedfast <fejj@ximian.com> * camel-session.c (camel_session_get_password): Don't abort if the Service is NULL, this is perfectly valid for cipher contexts. 2001-05-09 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (imap_move_messages_to): Pass appropriate parameters to CF_CLASS and add comment explaining why my initial attempt at a solution didn't work. 2001-05-09 Dan Winship <danw@ximian.com> * camel-disco-store.c (camel_disco_set_status): Set the disconnected status. (camel_disco_store_can_work_offline): Return whether or not a given CamelDiscoStore can work offline or not. * camel-disco-folder.c (camel_disco_folder_cache_message): Explicitly tell a folder to cache a message. (Better than using get_message, because for IMAP that doesn't guarantee you'll get all the message parts.) (camel_disco_folder_prepare_for_offline): Prepare a folder for offline use by caching all messages meeting given search criteria (and doing anything else the particular folder implementation needs). * camel-session.c (camel_session_set_online, camel_session_is_online): A session-wide online/offline toggle. (camel_session_init): Set online to TRUE. * providers/imap/camel-imap-store.c (can_work_offline): Implementation of CamelDiscoStore::can_work_offline. (Checks that the store has been used online at least once.) (imap_get_folder_online, imap_get_folder_offline): Deal with request for "inbox" properly. ("Don't you mean... 'INBOX'?"). * providers/imap/camel-imap-folder.c (imap_cache_message): Implementation of CamelDiscoFolder::cache_message. * camel.h: Add camel-disco-store.h and camel-disco-folder.h 2001-05-09 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-part.c (camel_mime_part_set_content_id): Wrap the content-id with <>'s. 2001-05-04 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (connect_to_server): Add support for using OpenSSL. * camel-remote-store.c (remote_connect): Add support for using the OpenSSL implementation. * camel-tcp-stream-ssl.c (ssl_bad_cert): Hmmmm, don't pass in a NULL as the last argument to alert_user - prototype doesn't take that argument anymore? * camel-tcp-stream-openssl.c (camel_tcp_stream_openssl_finalize): (ssl_verify): Use a global hash table to try and lookup the CamelTcpStreamOpenSSL object given the ssl context since OpenSSL doesn't think one needs to pass data around, we should all be living in a world of global variables, duh! 2001-05-06 Dan Winship <danw@ximian.com> * Makefile.am (libcamelinclude_HEADERS): Fix another build problem. (Um, anyone else could have done this...) 2001-05-04 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.c (get_folder_online): Fix up refcounting on current_folder. * camel-disco-folder.c, camel-disco-store.h: Remove CamelDiscoDiary refs that weren't supposed to escape yet. 2001-05-03 Dan Winship <danw@ximian.com> * camel-disco-store.c: new abstract class for disconnectable remote stores * camel-disco-folder.c: new abstract class for folders on disconnectable stores. * Makefile.am: Add camel-disco-folder.[ch] and camel-disco-store.[ch]. * providers/imap/camel-imap-store.c: Make this a subclass of CamelDiscoStore, and fix up the offline interfaces for the changes since they were first written (particularly the fact that some IMAP stores don't just use subscribed folders). * providers/imap/camel-imap-folder.c: Make this a subclass of CamelDiscoFolder, although most ops still fail in disconnected mode. * camel-store.c (camel_store_get_folder_info): Change gboolean, gboolean, gboolean to guint32 in the prototype for this function. * providers/local/camel-local-store.c (get_folder_info): Update for prototype change. 2001-05-02 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.h: Clean this up a bit. Add a "tag_prefix" member. Move "useful_lsub" into capabilities. * providers/imap/camel-imap-store.c (camel_imap_store_init): Initialize the tag_prefix, based on a static variable. * providers/imap/camel-imap-command.c (camel_imap_command): Use the store's tag_prefix character rather than "A" at the start of the tag. Makes the verbose debug output easier to parse when connected to multiple IMAP servers. (Well, unless you're connected to more than 26 servers...) * providers/imap/camel-imap-utils.c (imap_uid_array_to_set): Fix this up... it was losing count in some cases and giving a more verbose answer than it needed to. * providers/pop3/camel-pop3-store.c: Rescue the KPOP code from bit rot. 2001-05-01 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-command.c (camel_imap_command): Redo command locking. Since command_lock is recursive, we can just get a lock here, and release it either on error, or when the caller frees the response data. (This simplifies a lot of stuff, and fixes some problems with camel_imap_folder_changed being called without the command_lock locked because of the 2001-03-22 change.) (camel_imap_response_free): (camel_imap_response_free_without_processing): (camel_imap_response_extract): (camel_imap_response_extract_continuation): These all take a CamelImapStore now as well, to deal with locking. * providers/imap/camel-imap-private.h: Add CAMEL_IMAP_STORE_ASSERT_LOCKED, which defaults to a noop, but can be made to call e_mutex_assert_locked. * providers/imap/camel-imap-folder.c, camel-imap-search.c, camel-imap-store.c: Simplify using new locking stuff. Add a few CAMEL_IMAP_STORE_ASSERT_LOCKED checks. 2001-05-01 Dan Winship <danw@ximian.com> Support the IMAP UIDPLUS extension (RFC 2359), which lets you resync after disconnected operation more efficiently, but also makes it possible to do appends and moves/copies more efficiently now. * providers/imap/camel-imap-folder.c (imap_append_message): If the server supports UIDPLUS, grab the APPENDUID response and cache the message into the folder's message cache. (imap_copy_messages_to): Likewise, for COPYUID, copy any message parts we have cached between the source and destination folder caches. (imap_get_message): If the entire message is already in the cache, just return that rather than building it from parts. (imap_update_summary): Fetch just the "UID FLAGS RFC822.SIZE" of the new messages first, then only fetch the headers for messages where we don't already have the headers cached. * providers/imap/camel-imap-message-cache.c: Add gtk-doc comments. (cache_put): Fix refcounting stuff here. (camel_imap_message_cache_insert_stream, camel_imap_message_cache_insert_wrapper): New. (camel_imap_message_cache_get): Fix a bug here so the memory caching actually works. (camel_imap_message_cache_copy): New routine, used by imap_copy_messages_to. * providers/imap/camel-imap-utils.c (imap_uid_set_to_array): Inverse operation of imap_uid_array_to_set. Used to parse COPYUID response. 2001-04-30 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-utils.c (imap_uid_array_to_set): New-and-improved version of get_uid_set() from camel-imap-folder.c. Looks at the summary as it goes so that if you ask for messages 5, 6, 8, and 9, and there is no message 7, then you get "5:9" instead of "5:6,8:9" * providers/imap/camel-imap-folder.c (imap_copy_messages_to): Use imap_uid_array_to_set() rather than get_uid_set(). (get_uid_set): Gone 2001-04-27 Dan Winship <danw@ximian.com> * camel-session.c: Redo this a lot so that instead of having a class full of callbacks, we have a subclassable class. Also, replace the increasingly horrifying camel_session_query_authenticator with three new routines, camel_session_get_password, camel_session_forget_password, and camel_session_alert_user. * camel-pgp-context.c: * camel-pkcs7-context.c: * camel-smime-context.c: * providers/imap/camel-imap-store.c: * providers/pop3/camel-pop3-store.c: * providers/smtp/camel-smtp-transport.c: Use camel_session_get_password / camel_session_forget_password. * camel-tcp-stream-ssl.c (ssl_bad_cert): Use camel_session_alert_user. 2001-04-26 Jeffrey Stedfast <fejj@ximian.com> * camel.c (camel_init): Automagically call camel_shutdown() atexit() ;-) (camel_init): Also initialize NSS databases as read/write for S/MIME and if NSS fails to init, try initializing with volatile databases. * camel-file-utils.h: #include <sys/types.h> for off_t. 2001-04-26 Dan Winship <danw@ximian.com> * camel-file-utils.c: New file, with the int, string, time_t, and off_t encode/decode routines from camel-folder-summary.c moved here and renamed, for the enjoyment of non-CamelFolderSummary subclasses. * Makefile.am (libcamel_la_SOURCES): Add camel-file-utils.c (libcamelinclude_HEADERS): and camel-file-utils.h * camel-folder-summary.c: Remove functions that were moved to camel-file-utils.c, update uses of them for the new names. (camel_folder_summary_{en,de}code_token are still here.) * providers/local/camel-mbox-summary.c: Use camel_file_util_* names * providers/imap/camel-imap-summary.c: Use camel_file_util_* names * providers/imap/camel-imap-store.c (imap_store_setup_online, imap_store_setup_offline): Use camel_file_util_* names, which makes much more sense since this isn't folder summary stuff. 2001-04-26 Dan Winship <danw@ximian.com> * Makefile.am (INCLUDES): Remove UNICODE_CFLAGS (and some other stuff that's redundant with EXTRA_GNOME_CFLAGS) (libcamel_la_LIBADD): Replace UNICODE_LIBS with GAL_LIBS. * camel-search-private.c: * camel-pgp-context.c: * camel-mime-utils.c: Use gunicode interfaces rather than libunicode. * camel-charset-map.c: Use gunicode rather than libunicode. (The charmap-regen code still depends on libunicode though.) * camel-mime-filter-charset.h: * tests/message/test2.c (convert): Use iconv rather than unicode_iconv. * providers/smtp/Makefile.am (libcamelsmtp_la_LIBADD): * providers/pop3/Makefile.am (libcamelpop3_la_LIBADD): * providers/local/Makefile.am (libcamellocal_la_LIBADD): Remove UNICODE_LIBS. * camel.c (camel_init): Remove call to unicode_init. * camel-mime-parser.c: Remove unused unicode.h include. 2001-04-26 Dan Winship <danw@ximian.com> * camel-service.c (get_host): Use e_gethostbyname_r. (camel_service_gethost): Remove unused var. 2001-04-26 Not Zed <NotZed@Ximian.com> * Applied jacob's patches for e-poolv stuff. 2001-04-23 Jeffrey Stedfast <fejj@ximian.com> * camel.h: Add camel-vee-folder.h and camel-digest-folder.h * camel-digest-folder.[c,h]: New class that can be used to browse a multipart/digest message as if it were a CamelFolder. 2001-04-22 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (pgp_verify): Oops, don't use the return value of iconv() as a string length, it only tells us the number of non-reversable character conversions. This fixes it so we actually see the gpg output in the message viewer. 2001-04-23 Mikael Hallendal <micke@codefactory.se> * providers/nntp/Makefile.am (INCLUDES): Changed GTK_INCLUDEDIR to EXTRA_GNOME_CFLAGS. 2001-04-20 Jeffrey Stedfast <fejj@ximian.com> * camel-smime-context.c (smime_sign): Implemented using the secmime.h convenience functions. (smime_clearsign): We just error our here, there is no clearsign for smime. (smime_verify): Copied code over from the pkcs7 verify - same state, ugly and unknown :-) (smime_encrypt): Implemented using the secmime.h convenience functions. (smime_decrypt): Same as the code in pkcs7-context. 2001-04-19 Jeffrey Stedfast <fejj@ximian.com> * camel-pkcs7-context.c (pkcs7_sign): Implemented. (pkcs7_clearsign): Implemented. (get_zero_len_passwd): Convenience function to try a 0-length key on the certificate db - if this works then there is no need to prompt the user. (get_password): Callback to get the password. (camel_cipher_hash_to_nss): Convenience function to convert a CamelCipherHash to an NSS HASH_HashType. (nss_hash_to_sec_oid): Converts a HASH_HashType to a SecOidTag. (pkcs7_digest): Digest function for use with pkcs7_sign(). (sign_encode_cb): Callback to write the signed data to the output stream. (pkcs7_verify): Hacked on a bit more but am still a bit on the confused side. This might work but I wouldn't hold my breath. The sign/clearsign should be close if not correct, but this one...god only knows. (pkcs7_decrypt): Implemented. (pkcs7_encrypt): Implemented (mostly, still need to handle when 'sign' is TRUE). 2001-04-18 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.c (camel_imap_store_finalize): Free base_url and storage_path. * providers/pop3/camel-pop3-store.c (finalize): Free the implementation string. (camel_pop3_command): Clarify documentation to mention that @ex isn't set (and *@ret is) on CAMEL_POP3_ERR. (connect_to_server): Set @ex properly on CAMEL_POP3_ERR. * providers/pop3/camel-pop3-folder.c (pop3_refresh_info, pop3_get_message): Set @ex properly on CAMEL_POP3_ERR. 2001-04-17 Jeffrey Stedfast <fejj@ximian.com> * camel-pkcs7-context.[c,h]: New source files to handle Pkcs7 encryption, decryption, signing, and verifying. Not yet complete. I'm sensing this is going to take a while seeing as how NSS is just so well documented. 2001-04-17 Dan Winship <danw@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_message): Fix a memory leak in the error case. 2001-04-16 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-mbox-summary.c (summary_rebuild): Don't we want to tell where we are by using the start of the message? I think this should be using camel_mime_parser_tell_start_from() instead. Also divide by the size *before* multiplying by 100 so that we don't overflow the int which was giving us negative values for our progress status (oops). 2001-04-16 Jeffrey Stedfast <fejj@ximian.com> * camel-operation.c: Added a prototype for stamp(). * camel-charset-map.c (camel_charset_locale_name): #include string.h so we don't get a warning about strchr being undefined and also init charset to NULL. * camel-pgp-context.c (pgp_verify): Go back to doing the utf8 conversion by hand so that we don't depend on gal. * Makefile.am: Remove the EXTRA_GNOME_CFLAGS include. * camel-store.c (camel_mkdir_hier): Convenience function that it seems a number of camel-store implementations used gal for. * providers/nntp/camel-nntp-store.c (ensure_news_dir_exists): Lets not depend on gal for just e_mkdir_hier() - use camel_mkdir_hier() instead. * providers/nntp/camel-imap-store.c: Lets not depend on gal for just e_mkdir_hier() - use camel_mkdir_hier() instead. * camel-session.c (camel_session_get_storage_path): Don't depend on e_mkdir_heir() anymore, use the CamelStore version. * camel-folder-search.h: Removed gal dependency, why was this even there in the first place? * providers/imap/camel-imap-folder.c: Don't need gal/util/e-util.h here, so remove it. * string-utils.c (strstrcase): New function, well more like old function brought back to life so we don't have to depend on gal. * providers/imap/camel-imap-store.c (imap_store_setup_online): Use strstrcase rather than e_strstrcase so we don't depend on gal. (get_unread_online): Same here. * providers/smtp/camel-smtp-transport.c (smtp_helo): Use strstrcase. 2001-04-15 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_raw_check_mailing_list): regex doesn't set errno and regcomp returns 0 on success and any other value for an error (so don't *just* check for -1). 2001-04-14 Jeffrey Stedfast <fejj@ximian.com> * camel-cipher-context.c: Check to make sure the context is a valid context in all the user functions. 2001-04-13 Jon Trowbridge <trow@ximian.com> * providers/imap/camel-imap-store.c (imap_disconnect): s/imap_store/store/, fixing a typo that was causing the build to fail. * camel-filter-search.c (get_source): Make the (previously unused) get-source command actually do the right thing, properly shrouding any passed-in source and falling back to use the source attached to the mime message. 2001-04-13 Dan Winship <danw@ximian.com> * camel-object.c (camel_object_hook_event): * camel-uid-cache.c (camel_uid_cache_get_new_uids): * camel-url.c (camel_url_new_with_base): Plug leaks. * camel-remote-store.c (camel_remote_store_init): Don't re-initialize the folders hash table. * providers/imap/camel-imap-store.c (imap_disconnect): Free authtypes since they'll be re-read on re-connect. (get_folder_info): Free folderinfos that we're discarding from the list. 2001-04-12 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (camel_pgp_context_get_type): Oops, lets not get into a recursive call here ;-) * tests/smime/pgp.c: Updated to reflect changes to the PGP code. * tests/smime/pgp-mime.c: Same. 2001-04-12 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-mime.c: Updated to reflect the few changes made to the CamelPgpContext class. * camel-pgp-context.c: Updated to subclass of CamelCipherContext. * camel-cipher-context.[c,h]: New base class for CamelPgpContext. Also contains the replacement for CamelPgpValidity, CamelCipherValidity. 2001-04-12 Dan Winship <danw@ximian.com> * camel-mime-filter-crlf.c (filter): Fix a pair of array bounds reads noted by purify 2001-04-12 Jeffrey Stedfast <fejj@ximian.com> * camel.h: #include the pgp stuff. 2001-04-12 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_remove): * camel-url.c (camel_url_to_string): Plug memory leaks. 2001-04-11 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-utils.c (imap_parse_string_generic): Simplify peterw's imap_is_atom_char fix. 2001-04-11 Jeffrey Stedfast <fejj@ximian.com> * tests/smime/Makefile.am: Added pgp-mime.c to the tests. * tests/smime/pgp-mime.c: Test suite for camel-pgp-mime.c functions. * Makefile.am: Add camel-pgp-mime.[c,h] to the build. * camel-pgp-mime.c: Made a number of fixes to get it to compile and also fixed a few logic errors (mostly forgetting to reset streams) so that it worked (thanks to the pgp-mime test program). 2001-04-11 JP Rosevear <jpr@ximian.com> * providers/imap/Makefile.am: user GNOME_INCLUDEDIR since gnome files are included in the top level camel headers and the gtk include dir is now versioned and such * providers/local/Makefile.am: ditto * providers/pop3/Makefile.am: ditto * providers/smtp/Makefile.am: ditto * providers/sendmail/Makefile.am: ditto * camel-service.c: use five arg version of gethostbyname_r if appropriate (camel_get_host_byname): check if msg->herr is non-zero instead of checking if msg->hp is null since we may not always have msg->hp 2001-04-11 Jeffrey Stedfast <fejj@ximian.com> * camel-sasl-digest-md5.c (digest_response): Use camel_charset_locale_name() to get the locale charset rather than checking the CHARSET environment variable. This is a much less ugly hack. Also: If we fail to be able to convert to UTF8, then disavow all knowledge of the charset parameter. * camel-charset-map.c (camel_charset_locale_name): New function to return the locale charset (or NULL if US-ASCII). 2001-04-11 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (camel_pgp_verify): Use e_utf8_from_locale_string() rather than trying to do it manually since Lauris's version works much better. 2001-04-06 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Added camel-pgp-context.[c,h] to the build. Also added tests/smime/pgp.c and Makefile.am * tests/smime/pgp.c: New test suite for PGP functions. * camel-pgp-context.c: Various fixes to get it to build. 2001-04-07 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-utils.c (imap_parse_string_generic): It would be a good idea to recognize '\0'. Before, this code was running off the ends of strings and intermittently coredumping (if it didn't hit an imap_atom_char first) -- whoops! 2001-04-05 Not Zed <NotZed@Ximian.com> * merge from evolution-0-10-branch to evolution-0-10-merge-0 2001-04-03 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Added tests/ back in * camel-mime-utils.c (header_decode_text): Ignore whitespace between encoded words (there were a few cases where it didn't before). (header_encode_string): Preserve whitespace between words that are to be encoded by encoding them too. (header_encode_phrase): Same here but with phrases. 2001-04-03 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (imap_sync): If we don't change any flags or expunge, send a NOOP to give the server a chance to send EXISTS, etc, messages. (imap_get_message): If the server_level < IMAP4rev1, always fetch messages all-at-once, since they don't support the [#.MIME] syntax. (imap_update_summary): When reading message flags, set the server_flags field as well, and don't overwrite any flags set by camel_folder_summary_info_new_from_message. Might help with some of the flag problems... 2001-04-02 Dan Winship <danw@ximian.com> * camel-service.h: Remove the "quick_login" member, which is unnecessary. * providers/smtp/camel-smtp-transport.c (smtp_auth): Remove the references to quick_login and fix this to use the CamelSasl interfaces correctly to do the same thing. (connect_to_server): Split this out of smtp_connect (smtp_connect): Use connect_to_server. When re-EHLO'ing after auth, ignore errors. (query_auth_types): Use connect_to_server rather than smtp_connect, so it doesn't try to authenticate. Add LOGIN authtype to the list of authtypes to check for. * providers/smtp/camel-smtp-provider.c (camel_provider_module_init): Add LOGIN authtype to the authtypes list explicitly. * camel-sasl.c (camel_sasl_authtype_list): Don't list LOGIN here: it's not a real SASL authtype and is only used for SMTP. * camel-sasl-plain.c: * camel-sasl-login.c: * camel-sasl-kerberos4.c: * camel-sasl-cram-md5.c: * camel-sasl-anonymous.c: * providers/pop3/camel-pop3-provider.c: Remove "quick_login" argument from authtypes. 2001-04-02 Gediminas Paulauskas <menesis@delfi.lt> * camel-filter-driver.c: marked missing report status' for translation. 2001-04-02 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.c (camel_pgp_sign): mutex lock & unlock the context. (camel_pgp_clearsign): Same. (camel_pgp_verify): Same. (camel_pgp_encrypt): Same. (camel_pgp_decrypt): And finally here... * camel-pgp-context.h: Update the function prototypes to match those found in camel-pgp-context.c. * camel-pgp-context.c: Updated. * camel-pgp-mime.c (camel_pgp_mime_part_sign): Implemented. (camel_pgp_mime_part_verify): Implemented. (camel_pgp_mime_part_encrypt): Implemented. (camel_pgp_mime_part_decrypt): Implemented. 2001-04-01 Jeffrey Stedfast <fejj@ximian.com> * camel-sasl-login.[c,h]: New files to handle the LOGIN SASL mechanism. * camel-sasl-plain.c: Removed the definition of camel_sasl_login_authtype. * camel-sasl.c (camel_sasl_new): Oops. I thought LOGIN was an alias to PLAIN. I was wrong. These two SASL objects have to be separate. * providers/smtp/camel-smtp-transport.c (smtp_auth): Updated to check for and use authmech->quick_login when available. 2001-04-01 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-provider.c: Updated the authtypes here too. * camel-sasl-plain.c: * camel-sasl-anonymous.c: * camel-sasl-digest-md5.c: * camel-sasl-cram-md5.c: * camel-sasl-kerberos4.c: Updated the authtype values. * camel-service.h: Added another field to CamelServiceAuthType that specifies whether or not the mechanism supports "quick auth" which means that the client can send the initial challenge in the AUTH request. * camel-sasl.c (camel_sasl_new): Add support for LOGIN. (camel_sasl_authtype_list): Here too. (camel_sasl_authtype): And finally here. * camel-sasl-plain.c: Define camel_sasl_login_authtype. * providers/smtp/camel-smtp-transport.c (smtp_auth): Only unref the SASL object if it exists. 2001-03-30 Dan Winship <danw@ximian.com> * providers/pop3/camel-pop3-store.c (connect_to_server, query_auth_types, pop3_connect): Move things around here to make this all work right again (so you don't get prompted for a password when checking the supported authtypes.) 2001-03-30 Jon Trowbridge <trow@ximian.com> * camel-mime-message.c (camel_mime_message_set_source): Shrould our source URL before putting it into X-Evolution-Source. 2001-03-29 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (get_folder_info): Don't blindly append "/INBOX", construct a CamelURL and set the path that way. (parse_list_response_as_folder_info): Don't blindly append "/<foldername>", construct a CamelURL and set the path that way instead. (get_folders_offline): And again here. 2001-03-30 Not Zed <NotZed@Ximian.com> * camel-service.c (camel_get_host_byname): Dont overwrite an exception if we got one already. Added camel_operation status to it, and comment out some debug. 2001-03-29 Not Zed <NotZed@Ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_connect): Free host name info when done ... blah blah. * camel-sasl-kerberos4.c (krb4_challenge): Free host name info after we're done with it. * camel-sasl-digest-md5.c (digest_md5_challenge): Free host name info after we're done with it. * camel-remote-store.c (remote_connect): Free the host name info from get_host after we're finished with it. * camel-service.c (camel_get_host_byname): New function to lookup a name, and still be cancellable. (camel_free_host): And a function to free the structure returned. (camel_service_gethost): Call get_host_byname for this. 2001-03-29 Jon Trowbridge <trow@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_message): Save the source URL using camel_mime_message_set_source. * camel-mime-message.c (camel_mime_message_set_source): Renamed camel_mime_message_set_identity to this. Sets the X-Evolution-Source header. (camel_mime_message_get_source): Returns the X-Evolution-Source header. 2001-03-29 Kjartan Maraas <kmaraas@gnome.org> * broken-date-parser.c: #include <stdio.h>, <stdlib.h> and <string.h> * broken-date-parser.h: Remove all includes but <glib.h> * camel-address.h: Added #include <glib.h> * camel-data-wrapper.c: Move system includes to the top. * camel-data-wrapper.h: #include <glib.h> * camel-exception.c: #ifdef HAVE_CONFIG_H * camel-exception.h: Remove #include <glib.h> * camel-filter-driver.c: #ifdef HAVE_CONFIG_H and move system headers to top. Mark strings in camel_exception_set*() for i18n. * camel-filter-driver.h: #include <glib.h> * camel-filter-search.c: #ifdef HAVE_CONFIG_H * camel-folder-search.c: #include <config.h> * camel-folder-search.h: Header shuffling * camel-folder-summary.h: Same here * camel-folder-thread.c: #ifdef HAVE_CONFIG_H * camel-folder.c: Here too. * camel-folder.h: #include <glib.h> * camel-internet-address.c: Move system headers up * camel-lock.c: #ifdef HAVE_CONFIG_H * camel-medium.c: Same here * camel-mime-filter-bestenc.c: Move system headers up * camel-mime-filter-from.c: Same here * camel-mime-filter-linewrap.c: And here * camel-mime-filter.c: #include <string.h> to fix warnings from new gcc * camel-mime-filter.h: System headers moved up * camel-mime-message.c: Here too. #ifdef HAVE_CONFIG_H added * camel-mime-part-utils.c: #ifdef HAVE_CONFIG_H added * camel-mime-part.c: Here too, and shuffle headers * camel-mime-utils.c: #ifdef HAVE_CONFIG_H. Remove const from two variables to remove warning. * camel-mime-utils.h: Shuffle headers * camel-movemail.c: #ifdef HAVE_CONFIG_H. Mark string in camel_exception_set * camel-multipart.c: #ifdef HAVE_CONFIG_H. Added includes to remove warnings from new gcc * camel-object.c: #ifdef HAVE_CONFIG_H * camel-object.h: Shuffle headers * camel-operation.c: #ifdef HAVE_CONFIG_H * camel-provider.c: #ifdef HAVE_CONFIG_H. Shuffle headers. Mark a string * camel-remote-store.c: #ifdef HAVE_CONFIG_H * camel-sasl-anonymous.c: #ifdef HAVE_CONFIG_H. #include <string.h> * camel-sasl-cram-md5.c: #ifdef HAVE_CONFIG_H. #include <string.h> and <stdio.h> to remove warnings. * camel-sasl-digest-md5.c: #ifdef HAVE_CONFIG_H. shuffle headers * camel-sasl-digest-md5.h: Shuffle headers. * camel-sasl-kerberos4.c: #ifdef HAVE_CONFIG_H. Shuffle headers. * camel-sasl-kerberos4.h: Shuffle headers. * camel-sasl-plain.c: #ifdef HAVE_CONFIG_H. Shuffle * camel-sasl.c: #ifdef HAVE_CONFIG_H. #include <string.h> * camel-search-private.c: #include <config.h> * camel-seekable-stream.c: #ifdef HAVE_CONFIG_H * camel-seekable-stream.h: Shuffle * camel-seekable-substream.c: #ifdef HAVE_CONFIG_H * camel-service.c: Same * camel-service.h: Shuffle * camel-session.c: #ifdef HAVE_CONFIG_H * camel-store.c: Same * camel-stream-buffer.c: Here too * camel-stream-buffer.h: Shuffle * camel-stream-filter.c: #include <string.h> * camel-stream-fs.c: #ifdef HAVE_CONFIG_H * camel-stream-fs.h: Shuffle headers * camel-stream-mem.c: Same as above * camel-stream-mem.h: Here too * camel-stream-null.c: #ifdef HAVE_CONFIG_H * camel-stream.c: Here too * camel-stream.h: Shuffle * camel-tcp-stream-openssl.c: #ifdef HAVE_CONFIG_H * camel-tcp-stream-raw.c: Same here. and move local includes down * camel-tcp-stream-ssl.c: Same as above * camel-tcp-stream.c: #ifdef HAVE_CONFIG_H * camel-tcp-stream.h: move local include down * camel-transport.c: #ifdef HAVE_CONFIG_H * camel-transport.h: #include <glib.h> * camel-uid-cache.c: #ifdef HAVE_CONFIG_H. Shuffle headers * camel-uid-cache.h: Shuffle * camel-url.c: #ifdef HAVE_CONFIG_H * camel-vee-folder.c: #ifdef HAVE_CONFIG_H. Shuffle headers * camel-vee-folder.h: #include <glib.h> * camel-vee-store.h: Same here * camel.c: #ifdef HAVE_CONFIG_H * gstring-util.c: Here too * hash-table-utils.c: Remove #include <glib.h> * hash-table-utils.h: #include it here instead. * string-utils.c: #ifdef HAVE_CONFIG_H * providers/cache/camel-cache-map.c: #ifdef HAVE_CONFIG_H, shuffle headers and mark strings in camel_exception_set*() * providers/cache/camel-cache-provider.c: #ifdef HAVE_CONFIG_H * providers/cache/camel-cache-store.c: Same here, and mark a string. * providers/imap/camel-imap-command.c: #ifdef HAVE_CONFIG_H * providers/imap/camel-imap-command.h: Shuffle header * providers/imap/camel-imap-folder.c: #ifdef HAVE_CONFIG_H * providers/imap/camel-imap-message-cache.c: #ifdef HAVE_CONFIG_H * providers/imap/camel-imap-provider.c: #ifdef HAVE_CONFIG_H and #include <string.h> * providers/imap/camel-imap-search.c: #ifdef HAVE_CONFIG_H * providers/imap/camel-imap-store.c: #ifdef HAVE_CONFIG_H and mark a string. * providers/imap/camel-imap-summary.c: Shuffle headers * providers/imap/camel-imap-wrapper.c: #ifdef HAVE_CONFIG_H. Shuffle * providers/local/camel-local-folder.c: #ifdef HAVE_CONFIG_H * providers/local/camel-local-provider.c: #ifdef HAVE_CONFIG_H * providers/local/camel-local-store.c: Same here * providers/local/camel-local-summary.c: Here too, shuffle, mark a string * providers/local/camel-maildir-folder.c: #ifdef HAVE_CONFIG_H * providers/local/camel-maildir-store.c: Same * providers/local/camel-maildir-summary.c: Here too * providers/local/camel-mbox-summary.c: And here * providers/local/camel-mbox-store.c: And here * providers/local/camel-mbox-summary.c: #include <config.h> * providers/local/camel-mh-folder.c: #ifdef HAVE_CONFIG_H * providers/local/camel-mh-store.c: same * providers/local/camel-mh-summary.c: #ifdef HAVE_CONFIG_H, shuffle headers and mark a string for translation. * providers/nntp/camel-nntp-auth.c: #include <config.h> * providers/nntp/camel-nntp-store.c: Remove superfluous arg to CAMEL_SERVICE_CLASS to make it compile * providers/pop3/camel-pop3-provider.c: #ifdef HAVE_CONFIG_H * providers/pop3/camel-pop3-store.c: Same here * providers/sendmail/camel-sendmail-provider.c: Here too * providers/sendmail/camel-sendmail-transport.c: Here too * providers/smtp/camel-smtp-provider.c: Same * providers/smtp/camel-smtp-transport.c: #ifdef HAVE_CONFIG_H, mark strings for translations. 2001-03-29 Dan Winship <danw@ximian.com> * camel-url.c (camel_url_to_string): change "gboolean show_passwd" to "guint32 flags". * tests/misc/url.c (main): Update calls to camel_url_to_string * providers/imap/camel-imap-store.c (construct): Call camel_url_to_string with flags to hide password, authtype, and params to create the base_url. (etc): Update for the fact that the base_url no longer has the "/" at the end. * camel-service.c (construct): Update calls to camel_url_to_string (pass CAMEL_URL_HIDE_PASSWORD) 2001-04-03 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_decode_text): Ignore whitespace between encoded words (there were a few cases where it didn't before). (header_encode_string): Preserve whitespace between words that are to be encoded by encoding them too. (header_encode_phrase): Same here but with phrases. 2001-04-04 Kjartan Maraas <kmaraas@gnome.org> * camel-filter-driver.c: Added #include <config.h>. Marked a bunch of strings for translation. * camel-folder-search.c: Added #include <config.h>. * camel-folder-summary.c: Same here. * camel-movemail.c: Marked strings for translation. * camel-provider.c: Same here. * camel-search-private.c: #include <config.h> * camel-vee-folder.c: Mark string for translation. * providers/local/camel-mbox-summary.c: #include <config.h> 2001-04-03 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (imap_sync): If we don't change any flags or expunge, send a NOOP to give the server a chance to send EXISTS, etc, messages. (imap_get_message): If the server_level < IMAP4rev1, always fetch messages all-at-once, since they don't support the [#.MIME] syntax. 2001-04-01 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (get_folder_info): Don't blindly append "/INBOX", construct a CamelURL and set the path that way. (parse_list_response_as_folder_info): Don't blindly append "/<foldername>", construct a CamelURL and set the path that way instead. (get_folders_offline): And again here. 2001-03-31 Not Zed <NotZed@Ximian.com> * camel-operation.c (struct _status_stack): change stamp to unsigned to avoid overflow. (camel_operation_end): (camel_operation_progress_count): (camel_operation_progress): Double the delay before transient events start showing their own progress. 2001-03-30 Not Zed <NotZed@Ximian.com> * providers/pop3/camel-pop3-folder.c (pop3_get_message): Change the pop download into a transient event. * camel-operation.c (stamp): Get a timestamp that changes 4x a second (or rather, arbitrarily more often than 1/second). (camel_operation_progress): Use stamp() for time updates. (camel_operation_progress_count): " (camel_operation_start): Store the status as a struct status_stack item. (camel_operation_start_transient): New function to start a transient event, which is only reported if it runs for too long/and is reset afterwards. (camel_operation_progress): (camel_operation_progress_count): Fix for status_stack changes, and account for transient updates. * camel-filter-driver.c (camel_filter_driver_filter_folder): When we remove things (set deleted), always set seen too, to mirror interface behaviour when you delete. (camel_filter_driver_filter_message): And same here too. 2001-03-30 Dan Winship <danw@ximian.com> * providers/pop3/camel-pop3-store.c (connect_to_server, query_auth_types, pop3_connect): Move things around here to make this all work right again (so you don't get prompted for a password when checking the supported authtypes.) 2001-03-29 Dan Winship <danw@ximian.com> * camel-url.c (camel_url_to_string): change "gboolean show_passwd" to "guint32 flags". * tests/misc/url.c (main): Update calls to camel_url_to_string * providers/imap/camel-imap-store.c (construct): Call camel_url_to_string with flags to hide password, authtype, and params to create the base_url. (etc): Update for the fact that the base_url no longer has the "/" at the end. * camel-service.c (construct): Update calls to camel_url_to_string (pass CAMEL_URL_HIDE_PASSWORD) 2001-03-28 Jeffrey Stedfast <fejj@ximian.com> * camel-pgp-context.[c,h]: New class for PGP encrypting, decrypting, signing and verifying. Next I'll be moving the contents of mail/mail-crypto into Camel. These files will basically replace mail/openpgp-utils.[c,h]. 2001-03-29 Not Zed <NotZed@Ximian.com> * camel-folder-search.c (search_body_contains): Optimise the match "" case, just match everything in the folder (if we have it). * camel-vtrash-folder.c (vtrash_move_messages_to): Access the folder directly from the message info. (vtrash_move_messages_to): I think we also need to call the real move_messages_to, and not try and bypass the lock, since we dont have the lock for that folder. * camel-vee-folder.h: Move CamelVeeMessageInfo to public, so subclasses can get to it. * camel-vee-folder.c (camel_vee_folder_add_folder): Dont let you add a vfolder to itself. (folder_changed): (vee_folder_remove_folder): (vee_folder_build_folder): Copy changes before triggering changed events, so we can do them outside of locks. (camel_vee_folder_get_message_folder): Removed function. (camel_vee_folder_hash_folder): Made public (renamed from hash_folder), simply for subclasses (probably not needed). (vee_expunge): Oops, call the class sync, not superclass sync. (vee_sync): Remove some accidental debug. 2001-03-27 Jeffrey Stedfast <fejj@ximian.com> * camel-service.c (get_path): Add the port (if user defined) to the path to fix bug #1996. Danw: I'm iffy on the "fix" so I'd like you to comment. 2001-03-27 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-folder.c (pop3_sync): If we aren't going to even check the resp/free buffer that gets returned from camel_pop3_command, don't even bother using it. Send in NULL instead. (pop3_get_message): Check to make sure that the `result' string is non-NULL. It's possible that some POP servers may return with "+OK\r\n<message>..." rather than the expected "+OK ### octets\r\n<message>..." 2001-03-27 Not Zed <NotZed@Ximian.com> * camel-folder-summary.c (camel_folder_summary_finalize): Unref the html filter when done. (summary_build_content_info): Disable using the html parser to help index html mails until it gets sorted out. * camel-vee-folder.c (folder_changed): Fix a cut n paste bug, we want uid_removed not uid_changed deref, fixes bug #1999. 2001-03-27 Dan Winship <danw@ximian.com> * camel-sasl.c (camel_sasl_authtype_list): add an argument to say whether or not you want "PLAIN" in the list (so you don't end up with "Password" twice in the config dialog). * providers/imap/camel-imap-provider.c (camel_provider_module_init): * providers/imap/camel-imap-store.c (query_auth_types): We don't want PLAIN. * providers/smtp/camel-smtp-provider.c (camel_provider_module_init): * providers/smtp/camel-smtp-transport.c (query_auth_types): But we do. 2001-03-27 Dan Winship <danw@ximian.com> * camel-provider.h: Add an "extra_conf" field to CamelProvider with structures and defines and stuff, for providers to specify additional configuration options that they want. Also add a "supports ssl" flag to the provider flags. * camel-url.c (camel_url_new_with_base): Fix parsing of params. (output_param): And unparsing... * camel-remote-store.c: add a "default_ssl_port" field. (remote_construct): If the URL has the "use_ssl" parameter, set the store's use_ssl flag. (remote_connect): If using SSL and no port specified, use the default_ssl_port rather than the default_port. * providers/smtp/camel-smtp-provider.c (smtp_provider): add CAMEL_PROVIDER_SUPPORTS_SSL and CAMEL_URL_ALLOW_USER. (The config gui code knows that the ALLOW_USER and ALLOW_AUTH go together.) (ssmtp_provider): gone * providers/smtp/camel-smtp-transport.c (smtp_construct): Set a flag if "use_ssl" param is set. (smtp_connect): Check the CamelSmtpTransport use_ssl flag rather than checking if this is smtp or ssmtp. * providers/imap/camel-imap-provider.c (imap_conf_entries): Add a bunch of IMAP-specific configuration options, like "check for new mail in all folders" (default TRUE), "show only subscribed folders" (default TRUE), "ignore server-supplied namespace", and "apply filters to INBOX" (not yet implemented). (imap_provider): We support SSL and we no longer allow a path in the URL. (namespace is handled via extra_conf) (simap_provider): Gone * providers/imap/camel-imap-store.c (camel_imap_store_init): Set default_ssl_port, don't set CAMEL_STORE_SUBSCRIPTIONS here (construct): remove simap stuff, deal with "use_lsub", "namespace", "check_all", and "filter" parameters. Set base_url to not include params. (imap_store_setup_online): Don't ask for the namespace if it was set explicitly. Don't get subscribed folders if !use_lsub. (imap_concat): Fix a bug. (get_folder_info): Support for not checking all folders. * providers/imap/camel-imap-folder.c (imap_get_full_name): Rewrite to use imap_store->namespace, not service->url->path * providers/pop3/camel-pop3-provider.c (pop3_conf_entries): "keep on server" (currently still implemented by the mailer code, not here), "delete after N days" (not yet implemented). (pop3_provider): we support SSL (spop_provider): gone * providers/pop3/camel-pop3-store.c (camel_pop3_store_init): Set default_ssl_port (pop3_connect): Remove spop code 2001-03-26 Jeffrey Stedfast <fejj@ximian.com> * camel-vee-folder.c (vee_folder_construct): Move all of the camel_vee_folder_construct() internals here. (camel_vee_folder_construct): This function basically becomes a wrapper for vee_folder_construct and also checks to make sure that the UNMATCHED vfodler is constructed otherwise constructs it. (camel_vee_folder_new): Call vee_folder_construct() instead so that we don't get into a dead-lock. 2001-03-26 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (init_trash): Use camel_vtrash_folder_new() to create the vtrash folder now. * camel-vtrash-folder.[c,h]: New subclass of CamelVeeFolder for our vTrash folders. * camel-folder.c (camel_folder_copy_messages_to): Don't watch for vtrash folders anymore. (camel_folder_move_messages_to): Same. * camel-vee-folder.c (camel_vee_folder_class_init): Update. (vee_move_messages_to): Rewrite to use the new move API. * camel-filter-driver.c (do_copy): Updated to reflect copy_message_to changes. Create a temporary uid array and use that. (do_move): Same. (camel_filter_driver_filter_message): And again, here... * providers/imap/camel-imap-folder.c (imap_copy_messages_to): Update to the new API. (imap_move_messages_to): Same. (get_uid_set): New function to create a `set' string based on an array of UIDs for use with imap_copy_messages_to. * camel-folder.c (camel_folder_copy_messages_to): Replaces camel_folder_copy_message_to (camel_folder_move_message_to): Replaces camel_folder_move_message_to. 2001-03-27 Not Zed <NotZed@Ximian.com> * camel-vee-store.c (vee_get_folder): Added folder_created event here. * camel-vee-folder.c (unmatched_finalise): Removed, moved into main finalise code. (message_changed): Just create a new change entry and promote it to a folder_changed thing. (vee_sync): Always rebuild folder on sync, even when not expunge. (folder_changed): If not autoupdating, make sure we remove any removed entries. (vee_folder_build_folder): (vee_folder_remove_folder): NOP if we're called on folder_unmatched. (vee_search_by_expression): Only search each folder once. Should we also order the result in summary order? 2001-03-20 Not Zed <NotZed@Ximian.com> * camel-store.c (init_trash): Fix calling for vee_folder_new(). * camel-folder-summary.c (camel_folder_summary_remove_index): Optimise slightly, use ptr_array_remove_index rather than have to search for the index by using summary_remove. * camel-vee-folder.h: Removed local member from VeeFolder, since it was never used. * camel-vee-store.c (camel_vee_store_finalise): Setup finalise function. * camel-vee-folder.c (camel_vee_folder_set_expression): New function to set the query of an existing vfolder. (camel_vee_folder_construct): New function for convenience/subclasses. (camel_vee_folder_new): Removed exception parameter, no longer needed (since we dont search yet). (camel_vee_folder_new): Changed to use folder_construct, and no longer raise the folder created signal. (vee_folder_build_folder): Make it return error on exception, so we can work without exceptions. (vee_folder_remove_folder): Use remove index rather than remove_uid, this is ok since we have the summary locked for our own access. (camel_vee_folder_add_folder): Only add the folder to the unmatched private if it is not a private folder. (camel_vee_folder_remove_folder): Only remove the folder from unmatched if it is not private. (vee_expunge): Just call sync with expunge set. (vee_sync): If expunging, also refresh the search. (vee_folder_build_folder): We do our own locking now, removed from callers, also trigger changed events here too (within locks), fixed callers appropriately. (vee_folder_remove_folder): Do our own locking, and trigger changed events. (vee_folder_add_info): Renamed of vee_folder_add, we now take a hash of the folder name, rather than use the folders address. (hash_folder): Convert a folder name into an 8 character hash. (vee_get_message, vee_search_by_expression, vee_set_message_flags, vee_set_message_user_flag, vee_move_message_to): Changed the uid to be an 8 byte hash + original uid, with no ':' anymore. 2001-03-26 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_decode_text): When writing the last token (if not rfc2047 encoded), don't use g_string_append_len, use append_latin1. * camel-mime-message.c (write_to_stream): Don't spew a warning if the message-id wasn't set by the application, not a big deal since we create one right here. 2001-03-26 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-command.c (camel_imap_command): keep a reference on the store's current_folder. * providers/imap/camel-imap-store.c (camel_imap_store_finalize): (imap_disconnect): unref the current_folder before clearing it. * camel-url.c (camel_url_get_param): Add missing accessor. * camel-session.c (camel_session_get_provider): New. Return the provider for a URL. 2001-03-25 Dan Winship <danw@ximian.com> * camel-url.c (camel_url_new_with_base): New URL parser with full RFC1808 relative URL support. Yum. (camel_url_new): Wrapper around camel_url_new_with_base now. * tests/Makefile.am: add misc subdir * tests/misc/url.c: relative URL test cases from RFC 1808 2001-03-25 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (do_copy): Don't use copy_to if the source folder doesn't support summary info's. (do_move): Same. (camel_filter_driver_filter_message): And again here. I think this will fix the problem of "colour" tags not staying with the message when it's copied to a folder during the filter process. 2001-03-24 Jeffrey Stedfast <fejj@ximian.com> * camel-sasl-digest-md5.c (digest_response): Do charset conversion on the username param here if needed. (parse_server_challenge): Protect against possibly empty-string charset values. 2001-03-23 Jon Trowbridge <trow@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_message): Don't call camel_mime_message_set_identity. (The call is commented out, left over from some earlier experimentation that I want to be able to remember later...) * camel-mime-message.c (camel_mime_message_set_identity): Added. A function to set the X-Evolution-Identity header. 2001-03-22 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-command.c (imap_read_response): Don't look at untagged responses other than "* BYE"... (camel_imap_response_free): ...do it here instead... (camel_imap_response_free_without_processing): ...but not here. * providers/imap/camel-imap-folder.c (camel_imap_folder_selected): Remove the "EXISTS" line from the response so camel_imap_response_free won't try to process it. (camel_imap_folder_selected): If we get an EXISTS response mixed in with the FETCH responses, record the new EXISTS value. Use camel_imap_response_free_without_processing so that an EXISTS response won't cause a loop. (imap_rescan): Use camel_imap_response_free_without_processing. (imap_update_summary): Use camel_imap_response_free_without_processing. If an EXISTS response shows up, call imap_update_summary again to get the new messages. (imap_protocol_get_summary_specifier): Use BODY.PEEK[0], not RFC822.HEADER for prev IMAP4rev1, since the FETCH parser won't deal with RFC822.HEADER responses. * providers/imap/camel-imap-store.c (get_folder_info): Fix an uninitialized variable 2001-03-21 Dan Winship <danw@ximian.com> * camel-provider.h: (CamelProvider) add a "GList *authtypes", so you can get the list of authtypes used by a provider without needing to have an actual CamelService object handy. (Will be needed by the new config druid.) (CAMEL_PROVIDER_ALLOWS, CAMEL_PROVIDER_NEEDS): New macros to test the URL part stuff, since the way it works is too complicated and everyone always does it wrong. * camel-service.c (camel_service_query_auth_types): Remove the @connected arg again: if you don't want to connect, you can just get the list of authtypes off the provider. (camel_service_free_auth_types): Remove this. All existing implementations do authtypes the same way, so just say the caller should "g_list_free" the list. (Oh, look, removing this function doesn't actually cause the mailer to not build. How 'bout that.) (construct, get_path): Use the new URL part macros. * camel-remote-store.c (remote_query_auth_types): Update (remote_free_auth_types): Nuke (camel_remote_store_authtype_list): New function for use by subclasses. * providers/imap/camel-imap-provider.c: * providers/pop3/camel-pop3-provider.c: * providers/smtp/camel-smtp-provider.c: Update CamelProvider structures. (camel_provider_module_init): Put all the SSL parts together so there's only 1 #ifdef. Set up the provider authtypes field using the SASL, CamelRemoteStore, and standard authtypes, as appropriate. Copy that from the normal provider to the SSL provider. * providers/local/camel-local-provider.c: * providers/sendmail/camel-sendmail-provider.c: * camel-session.c: Update CamelProvider structures. * providers/imap/camel-imap-store.c (query_auth_types): * providers/pop3/camel-pop3-store.c (query_auth_types): Update * providers/smtp/camel-smtp-store.c (query_auth_types): Update. Remove the no_authtype, because that's what "ALLOW_AUTH" rather than "NEED_AUTH" means. (free_auth_types): Nuke. 2001-03-21 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_message): Return -1 on fail. (camel_filter_driver_filter_folder): Same. (camel_filter_driver_filter_mbox): Same. (camel_filter_driver_filter_folder): Return -1 if an exception was set as well. 2001-03-19 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-openssl.c (camel_tcp_stream_openssl_finalize): Free the expected host. (camel_tcp_stream_openssl_new): Now takes a Service and an expected_host. Set them. 2001-03-20 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.c (camel_imap_store_check_online): Function to check if the store is online and set an exception if not. Currently controlled by an environment variable, but eventually there will be both a global (session-level) setting and a per-store setting. (construct): Set up storage_path and base_url here rather than at connect-time. (imap_auth_loop): Split out from imap_connect. (imap_setup_online): Split out from imap_connect. Do the post-authentication connection setup, and cache the results to disk. (imap_setup_offline): Set up a CamelImapStore with information saved from a previous imap_setup_online. (imap_connect): If online, do connect_to_server(), imap_auth_loop(), and imap_setup_online(). Otherwise, do imap_setup_offline(). (get_folder, get_folder_info): Add offline support. (create_folder, subscribe_folder, unsubscribe_folder): Disable these when offline (for now). * providers/imap/camel-imap-folder.c (camel_imap_folder_new): Remove the sync'ing-with-server stuff... it's done by camel_imap_folder_selected now, which only gets called if the store is online. (camel_imap_folder_selected): add the code removed from camel_imap_folder_new. Besides simplifying the folder_new and summary_new code, this also means now that we'll DTRT if a folder's UIDVALIDITY changes while we're connected. Also, when that happens, clear the message cache explicitly. (imap_refresh_info, imap_sync): These are no-ops when offline. (imap_expunge, imap_append_message, imap_copy_message_to, imap_search_by_expression): These don't yet work offline. (imap_get_message, camel_imap_folder_fetch_data): Return an error when trying to fetch a non-cached body part when we're offline. * providers/imap/camel-imap-summary.c (camel_imap_summary_new): Rewrite to not check the validity here. (We'll do it from camel_imap_folder_selected instead.) * providers/imap/camel-imap-command.c (camel_imap_command): Call camel_imap_folder_selected even when the selection is all we're doing, to match the changes in camel-imap-folder.c. * providers/imap/camel-imap-message-cache.c (camel_imap_message_cache_clear): New function to clear out a message cache. 2001-03-19 Christopher James Lahey <clahey@ximian.com> * Makefile.am (INCLUDES): Added $(EXTRA_GNOME_CFLAGS) 2001-03-19 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-store.c (pop3_connect): Set the errbuf to NULL after freeing it? I don't think this should fix bug #1801 but I guess it can't hurt. 2001-03-18 Jeffrey Stedfast <fejj@ximian.com> * camel-remote-store.c (remote_recv_line): Protect against a possibly NULL exception. * camel-filter-driver.c (camel_filter_driver_filter_message): Document and modify to return a boolean value denoting whether or not errors occured during processing. (camel_filter_driver_filter_mbox): Same as above and also modified to check return codes of filter_message rather than relying only on whether or not the exception was set (as it's possible to pass in NULL for the exception). (camel_filter_driver_filter_folder): Same as above. * camel-mime-utils.c (header_param_list_format_append): Quote the param value if it contains whitespace as well. 2001-03-17 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Don't include system include dir. * camel.c: Use quotes around the nss.h include so it doesn't try to use the systen nss.h file. * camel-tcp-stream-ssl.c: Same here. 2001-03-17 Not Zed <NotZed@Ximian.com> * providers/local/camel-maildir-summary.c (maildir_summary_sync): Modified patch from Dan Berger <dberger@ix.netcom.com> to re-check the maildir directory for new/updated messages at sync time. * camel-folder.c (camel_folder_ref_message_info): Implemented. (ref_message_info): And default implementation. * camel-folder.h: Added ref_message_info virtual method. 2001-03-16 Not Zed <NotZed@Ximian.com> * camel-folder-thread.c (camel_folder_thread_messages_new_summary): New function to create a thread tree from a supplied summary array. (camel_folder_thread_messages_destroy): Handle thread trees generated by the above function properly. 2001-03-16 Jeffrey Stedfast <fejj@ximian.com> * camel-private.h: Same. * camel-remote-store.h: Fixed #include to use <>'s instead of ""'s 2001-03-16 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (ssl_bad_cert): Print info about the issuer of the certificate. * providers/smtp/camel-smtp-transport.c (smtp_connect): Use camel_tcp_stream_get_socket(). * camel-tcp-stream-openssl.c (stream_get_socket): Implemented. * camel-tcp-stream-ssl.c (stream_get_socket): Implemented. * camel-tcp-stream-raw.c (stream_get_socket): Implemented. * camel-tcp-stream.c (camel_tcp_stream_get_socket): New function. 2001-03-16 Kjartan Maraas <kmaraas@gnome.org> * providers/pop3/camel-pop3-folder.c: Added #include <config.h> 2001-03-16 Dan Winship <danw@ximian.com> * camel-mime-utils.c (header_msgid_generate): Don't use getdomainname: it's NIS crap, not DNS. 2001-03-15 Jeffrey Stedfast <fejj@ximian.com> * camel.c (camel_init): Don't use the hardcoded cert db directory, use the one passed in. * camel-tcp-stream-ssl.c (ssl_get_client_auth): Wrote the default implementation. Not that we'll use this though, since this is the default implementation provided by NSS anyway. This more or less serves as a reference in case we want to change anything. (ssl_auth_cert): Same. (ssl_bad_cert): Changed the prompt string and free it when we're done. 2001-03-15 Dan Winship <danw@ximian.com> First batch of disconnected IMAP-related stuff. This adds local caching of message parts, but NOT any actual disconnected support. (But it should speed up IMAP use.) * providers/imap/camel-imap-message-cache.c: New class for caching message data to disk, and removing it when it's no longer relevant. Will eventually also support merging message parts together to save on files. Or maybe using a db instead of files? * providers/imap/camel-imap-private.h: Add a cache_lock to CamelImapFolderPrivate. This lock must be recursive, so make both locks EMutexes rather than GMutex. * providers/imap/camel-imap-folder.c (parse_fetch_response): "The only FETCH response parser you need!" Replaces the various almost-correct bits of code formerly scattered throughout this file with a single fully-correct function that can handle any FETCH response at any time, so we don't get confused by seeing a flags update when we were only expecting a message body, etc. (camel_imap_folder_fetch_data): FETCH a message body part either from the cache or the server (camel_imap_folder_changed): Remove expunged messages from the message cache. (camel_imap_folder_new): Change to take a directory instead of a summary file name. Create a CamelImapMessageCache for the folder. (imap_finalize): Unref the message cache. (camel_imap_folder_selected, imap_rescan, get_content, get_message, imap_get_message, imap_update_summary): Redone a bunch to use parse_fetch_data, CamelImapMessageCache, etc. * providers/imap/camel-imap-store.c (get_folder): Pass directory name to camel_imap_folder_new, not summary filename. Use e_path_to_physical to generate a path with /subfolders/ inserted between directory components. * providers/imap/camel-imap-wrapper.c (camel_imap_wrapper_new): Call camel_imap_folder_fetch_data (with cache_only TRUE) and if the data is cached, return an online datawrapper rather than an offline one. (write_to_stream): Use camel_imap_folder_fetch_data (with cache_only FALSE) here too * providers/imap/camel-imap-utils.c (imap_skip_list): Renamed from skip_list and made non-static. 2001-03-15 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.h: Uninclude prnetdb.h since it's not needed here. * providers/smtp/camel-smtp-transport.c: #include prnetdb.h here instead of making camel-tcp-stream-ssl.h include it. Prevents some header bloat. * providers/smtp/Makefile.am: Include the NSPR_CFLAGS and NSS_CFLAGS. 2001-03-14 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_connect): Update for using CamelTcpStreams and also for using SSL. * providers/smtp/libcamelsmtp.urls: defined "ssmtp" * providers/smtp/camel-smtp-provider.c: Defined the Secure SMTP provider. (camel_provider_module_init): Register the Secure SMTP provider. 2001-03-14 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (imap_connect): Check for "simap" instead of "imaps". * providers/pop3/camel-pop3-store.c (pop3_connect): Change the port to be 995 for spop. * providers/pop3/libcamelpop3.urls: Change to "spop". * providers/imap/libcamelimap.urls: Change "imaps" to "simap" to correspond to /etc/services. 2001-03-14 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-store.c (pop3_connect): Check to see if we are using "pops" and then set the appropriate SSL settings. * providers/pop3/libcamelpop3.urls: Add "pops" * providers/pop3/camel-pop3-provider.c: Defined the pops provider. (camel_provider_module_init): Register the pops provider. 2001-03-14 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (camel_tcp_stream_ssl_new): Ya know what? Lets not ref the service because otherwise we'll start having the same problems we had back with vtrash. Besides, since the store owns the stream, it's not like the store is going anywhere without the stream anyway. (camel_tcp_stream_ssl_finalize): No need to unref the service. 2001-03-14 Jeffrey Stedfast <fejj@ximian.com> * camel-remote-store.c (remote_connect): Pass the service into the SSL stream, not the session. * camel-tcp-stream-ssl.c (camel_tcp_stream_ssl_init): Set the service to NULL. (camel_tcp_stream_ssl_finalize): Unref the service. (camel_tcp_stream_ssl_new): Takes a CamelService arg now rather than a CamelSession arg. 2001-03-14 Jeffrey Stedfast <fejj@ximian.com> * camel.c (camel_init): So it turns out that NSS_Init *isn't* idempotent, so we have to protect against initializing it more than once(contrary to what their design specs suggest). * camel-session.c (camel_session_get_service): Use camel_exception_is_set() - Makes no difference but it's more consistant with how we normally do it. * camel-provider.h (CAMEL_URL_ALLOW_SSL): We don't need this. * providers/imap/camel-imap-provider.c: Define the imaps provider. (camel_provider_module_init): Register the imaps provider. * camel-provider.c (camel_provider_init): Only add the protocol to the hash table if it's non empty. Also, g_strdup() the filename into the hash table. * providers/imap/camel-imap-store.c (camel_imap_store_init): Eek! So the service's URL isn't set until after this is initialized. This means we can't check for SSL here. (imap_connect): Set the SSL options here instead. 2001-03-14 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c: Numerous fixes to get it to build correctly with NSS enabled. * camel-remote-store.c (remote_connect): Pass in the session and expected host args, oops. * camel-provider.h (CAMEL_URL_ALLOW_SSL): Defined. * providers/imap/camel-imap-store.c (camel_imap_store_init): Check to see if we are supposed to use SSL and set the options accordingly. (imap_connect): Return FALSE here instead of NULL. * providers/imap/camel-imap-provider.c: Add CAMEL_URL_ALLOW_SSL. * providers/imap/libcamelimap.urls: Add "imaps" which is the protocol for Secure IMAP. 2001-03-14 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-openssl.[c,h]: Added * Makefile.am: Added camel-tcp-stream-openssl.[c,h] to the build. * camel-types.h: Added CamelTcpStreamSSL and CamelTcpStreamOpenSSL 2001-03-14 Not Zed <NotZed@Ximian.com> * camel-mime-filter-charset.c (filter, complete): re-constify inbuf, to remove a warning. * camel-mime-parser.c (folder_scan_step): When we're out of data, run the filter_complete. For some reason the logic that was there was never being run, always try it now, i think it was to work around a buggy filter, rather than fix it the right way. * camel-folder-summary.c (summary_build_content_info): If indexing html parts, use the html filter to convert it to some indexable format. (summary_build_content_info): Reset the filters before adding them back to the stream, if they get re-used in a given instance (likely). * Makefile.am (libcamelinclude_HEADERS): Added camel-mime-filter-html.[ch]. (INCLUDES): Added xml clags 2001-03-05 Not Zed <NotZed@Ximian.com> * camel-folder-search.c (camel_folder_search_class_init): Setup a new function, "uid" which matches uids. (search_uid): Implement the "match uid" command. 2001-03-13 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c: Check for a new type of mailing list header. 2001-03-13 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-raw.c (stream_read): Made cancellable. (stream_write): Same. (stream_connect): Removed checks for DIVINE_INTERVENTION as the code worked fine. * camel-tcp-stream-ssl.c (stream_write): get rid of 'w' as it wasn't really needed. (set_errno): New function to set errno based on the NSPR error code. (stream_read): If an error occured, call set_errno(). (stream_write): Same. 2001-03-13 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Added camel-tcp-stream-ssl to the build. * camel-remote-store.h: Add a use_ssl member to the CamelRemoteStore object. 2001-03-13 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (camel_tcp_stream_ssl_get_type): Oops. Make this a subclass of CamelTcpStream, not CamelStream. * camel-types.h: Add the defs for CamelTcpStream and CamelTcpStreamRaw * Makefile.am: Add camel-tcp-stream and camel-tcp-stream-raw to the build. * camel-remote-store.c (remote_connect): Update to use CamelTcpStreams. * camel-tcp-stream-raw.c (stream_connect): Made cancellable by copying the currently used code in camel-remote-store.c. (stream_setsockopt): Oops, flags = fcntl (..., GET_FL); (camel_tcp_stream_raw_get_type): Oops. Make this a subclass of CamelTcpStream, not CamelStream. 2001-03-13 Jeffrey Stedfast <fejj@ximian.com> * camel.c (camel_init): Since all of the Mozilla libs (including NSPR and NSS) correctly handle reinitializations, we might as well init both NSPR and NSS in camel_init so we can be sure of that these libs have been initialized. (camel_shutdown): New function to call the NSS cleanup stuff. 2001-03-13 Jeffrey Stedfast <fejj@ximian.com> * camel-folder.c (camel_folder_copy_message_to): Don't allow the user to copy a message to or from the vtrash folder. Set an exception if they try. (camel_folder_move_message_to): This one is a bit more complicated: 1) If the user tried to move a message from the vtrash into the original folder, just undelete the message. 2) If the user tries to move the message to the vtrash folder, then just mark the message as deleted. 3) If the user tries to move the message from the vTrash to a folder other than the original, use the original folder as the source. As another optimization, I've made it so that if the user tries to move a message to the same folder, just no-op. (move_message_to): Unset the deleted flag before moving (assuming it's there). (copy_message_to): Same. * camel-vee-folder.c (camel_vee_folder_get_message_folder): New convenience function to get the folder from which the message uid is derived from. 2001-03-13 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.c (imap_connect): When trying again after a failed password attempt, check if the connection got dropped (which Courier will do) and deal accordingly. * providers/sendmail/camel-sendmail-transport.c (sendmail_send_to, sendmail_send): More fixing for exim: It can't deal with "-tif", you need to say "-t -i -f". 2001-03-12 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (init_trash): Rename the folder to "Trash" and i18nify it. 2001-03-09 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.h: Comment out everything unless HAVE_NSS is defined. * camel-tcp-stream-ssl.c (stream_read): Don't use errno, use nspr's error code stuff. (stream_write): Same. 2001-03-09 Jeffrey Stedfast <fejj@ximian.com> * camel-session.c (camel_session_query_authenticator): Created a new mode (CAMEL_AUTHENTICATOR_ACCEPT) which is a Yes/No prompt to the user. This will be needed by the SSL/TLS code to come. Also changed the return value to a gpointer rather than a char* to allow the returning of TRUE/FALSE values. * camel.c: Wrap stuff with HAVE_NSS 2001-03-09 Dan Winship <danw@ximian.com> * providers/sendmail/camel-sendmail-transport.c (sendmail_send, sendmail_send_to): Remove the "-U" for now, because it breaks exim's "sendmail" wrapper. (And it doesn't actually do all that much to sendmail's behavior, and it was working fine before anyway.) * camel-mime-utils.c (header_fold): Fix a bug that could make this walk off the end of a string. (The bug has been there since December... maybe this will fix some unreproduceable crashes?) 2001-03-08 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (camel_store_get_folder): Changes so that the folder gets added to the vtrash reguardless of whether or not the store is holding it's own folder lookup hash. (init_trash): Oops, lets try locking the cache before we go and add all the cached folders to the vtrash. 2001-03-08 Jeffrey Stedfast <fejj@ximian.com> These fixes make it so that the CamelStore does not actually hold a ref on it's vTrash folder. The vTrash folder is now created ONLY when it is requested. This solves the problem of not being able to unref a Store and have it disconnect/finalize like we expect it to. * camel-store.c (init_trash): When we create the vtrash, add all previously opened folders to it and hook on to the finalize event so that we can then set the store->vtrash to NULL. (get_trash): If a vtrash folder for the store doesn't exist, init_trash() and then try. (construct): No longer need this. (camel_store_class_init): Don't override the construct method. (camel_store_finalize): We no longer hold a ref on the vtrash so don't unref it. 2001-03-07 Jeffrey Stedfast <fejj@ximian.com> * providers/pop3/camel-pop3-store.c (init_trash): Implement. (get_trash): Implement. (camel_pop3_store_class_init): Override the default init_trash and get_trash virtual functions. Hopefully this should fix the problem of pop3 sotre's not disconnecting after the send&recv code finishes downloading mail. Wonder if I should override the default implementation for the mbox, mh, and maildir too? 2001-03-07 Dan Winship <danw@ximian.com> * camel-remote-store.c (camel_remote_store_connected): Return the return value of camel_service_connect rather than returning !camel_exception_is_set(), since that will be wrong if ex==NULL... Fixes a crash in the IMAP keepalive code, but this same problem probably exists in other places too... (remote_recv_line): Fix another problem... * providers/local/camel-mbox-provider.c: * providers/vee/.cvsignore: * providers/vee/libcamelvee.urls: Remove some unused stuff 2001-03-06 Jeffrey Stedfast <fejj@ximian.com> * camel-sasl-digest-md5.c (compute_response): Redone so as to save on memory allocations and speed. 2001-03-05 Jeffrey Stedfast <fejj@ximian.com> * camel-sasl-plain.c (plain_challenge): Don't return NULL if the token is non-NULL. This is why: sending : AUTH PLAIN received: 334 ok. go on. <-- this is why sending : ZGZPaQpAZ214Lm5ldBBnb29jYXI= received: 235 {mp005-rz3} go ahead * camel-sasl.c (camel_sasl_authtype): Add the PLAIN type here. (camel_sasl_authtype_list): And here too. * camel-sasl-plain.c: Initialize the camel_sasl_plain_authtype. * camel-sasl-plain.h: extern the camel_sasl_plain_authtype. 2001-03-05 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (imap_connect): i18n'd some strings in here. * providers/smtp/camel-smtp-transport.c (smtp_connect): Keep trying to authenticate until either we succeed or until the user cancels. 2001-03-04 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c: Wrote a new version of header_fold() that takes a 'force' option and uses another new function called header_fold_next_space() in place of strchr(inptr, ' ') to get the next whitespace char. The idea of header_fold_next_space() is to not treat spaces between a set of quotes as a space - this way it will be possible to fold (for example) the Content-Type MIME header without worrying about breaking up a boundary string. Note: This code is #if 0'd out until Zucchi approves of the patch. Another Note: We will probably still want to use the "don't fold this header" hash lookup for headers like the Message-Id and possibly a few others. 2001-03-04 Jeffrey Stedfast <fejj@ximian.com> * camel-internet-address.c (camel_internet_address_encode_address): Make sure there is a space between the name and the address tokens. * camel-mime-utils.c (rfc2047_decode_word): encoded string segments need to be at least 8 chars (7 doesn't leave room for an ecoding value. To be more realistic, even 8 chars isn't enough as this assumes there is no charset nor any encoded text. If the encoding value is not 'Q' or 'B', then return NULL. This fixes bug #1689. 2001-03-03 Not Zed <NotZed@Ximian.com> * camel-vee-folder.c (vee_folder_remove_folder): New function to remove a folder from a vfolder. (camel_vee_folder_remove_folder): New public function to remove a folder from a vfolder. 2001-03-02 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-provider.c: Doh! Set the CAMEL_URL_ALLOW_AUTH flag. * providers/smtp/camel-smtp-transport.c (get_smtp_error_string): Updated with the extended AUTH return codes. 2001-03-02 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-store.c (try_auth): Don't forget to unref the SASL object. * providers/smtp/camel-smtp-transport.c (query_auth_types): Implemented. (smtp_auth): Implemented. (smtp_helo): Don't bother parsing the authtypes if we already have them. (smtp_connect): call smtp_auth() here if we found any authtypes. (smtp_disconnect): Updated as I now use a hash table for the supported authtypes rather than a linked list. (esmtp_get_authtypes): modify to use a hash table instead of a linked list and also use isspace() rather than == ' '. 2001-03-02 Dan Winship <danw@ximian.com> * providers/sendmail/camel-sendmail-transport.c (sendmail_send_to, sendmail_send): Use "-f" to set the envelope from address so bounces go to the right place. Also, pass "-U" since the man page says we're supposed to... 2001-03-01 Jeffrey Stedfast <fejj@ximian.com> * camel-sasl.c: #include "camel-sasl-digest-md5.h" (camel_sasl_authtype_list): add DIGEST-MD5 stuff here. (camel_sasl_authtype): And here too. (camel_sasl_new): And here... * camel-sasl-digest-md5.[c,h]: new SASL class for DIGEST-MD5 (has been tested and proven to work). 2001-03-01 Dan Winship <danw@ximian.com> * camel-sasl.c (camel_sasl_challenge): Take a GByteArray as input as well. Comment that you can pass %NULL for @token to get the initial auth data for mechanisms that are client-initiated. (camel_sasl_challenge_base64): Convenience function for protocols that use base64-encoded SASL. (camel_sasl_authenticated): Implement this... (it was prototyped already) (camel_sasl_new): Function to take a service name, a mechanism name, and a CamelService, and return a CamelSasl for it. (camel_sasl_authtype, camel_sasl_authtype_list): Functions to return CamelServiceAuthType information about SASL mechanisms, to allow providers to deal with them generically. * camel-sasl-anonymous.c, camel-sasl-plain.c: Update/simplify for CamelSasl changes. Both of these are single-round (client-initiated) mechanisms, so they don't need to keep state. (camel_sasl_plain_new): Removed; use camel_sasl_new instead. (Can't get rid of camel_sasl_anonymous_new though...) * camel-sasl-cram-md5.c: Update/simplify for CamelSasl changes. (camel_sasl_cram_md5_new): Removed; use camel_sasl_new instead. (cram_md5_challenge): Use md5_get_digest where possible, and various other minor simplifications. CRAM-MD5 only has a single round, so there's no need to keep track of state. This code is now tested (against Cyrus IMAPd) and known to work. * camel-sasl-kerberos4.h: Update/simplify for CamelSasl changes. Make only a single #ifdef HAVE_KRB4. Remove stuff from priv that isn't needed between rounds. (camel_sasl_kerberos4_new): Removed; use camel_sasl_new instead (krb4_challenge): Fix up the logic I broke in my previous "at least make it compile" fixes, update to match other changes, and remove IMAP-isms that shouldn't be in the generic code. This still isn't tested, because we're stuck behind a NAT right now... * providers/imap/camel-imap-store.c: Add an "authtypes" hash table to CamelImapStore recording the supported authtypes. (connect_to_server): Record supported authtypes in the authtypes hash rather than the capabilities bitmask, since now the IMAP code is no longer responsible for keeping track of which authtypes we support. (query_auth_types): Use camel_sasl_authtype_list to get the SASL-supported authtypes. (try_auth): New function to try a SASL auth mechanism. Sort of formerly imap_try_kerberos_v4_auth. (imap_connect): Get rid of the krb4-specific bits and genericize them for any SASL authtype. * providers/imap/Makefile.am: Remove camel-imap-auth.[ch] (moved into camel-imap-store.c since it's now constant size) and the KRB4_CFLAGS and KRB4_LDFLAGS references. 2001-03-01 Jeffrey Stedfast <fejj@ximian.com> * camel-sasl-anonymous.c (anon_challenge): Don't base64 encode the data, leave that up to the provider to do if it needs to. Also save on some memory allocations and stuff. * camel-sasl-cram-md5.c (cram_md5_challenge): Same here. Also get rid of the need for the temp buf as we can just write it to the GByteArray. 2001-03-02 Not Zed <NotZed@Ximian.com> * camel-mime-utils.c (rfc2047_encode_word, rfc2047_decode_word): Reconstify variables to match iconv. * camel-search-private.c (camel_ustrstrcase): Change some assertions back into valid returns. 2001-03-01 Dan Winship <danw@ximian.com> * Makefile.am (libcamel_la_LIBADD, INCLUDES): camel depends on krb4 now (where available), for camel-sasl-kerberos4.c * camel-sasl-kerberos4.[ch]: Make this at least compile. Dunno if it works yet. 2001-03-01 Peter Williams <peterw@ximian.com> * camel-url.c (camel_url_new): Assert url_string != NULL. 2001-02-28 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Added camel-sasl-anonymous.[c,h] to the build. * camel-sasl-anonymous.[c,h]: new SASL class for ANONYMOUS * camel-sasl-plain.c (plain_challenge): Oops, have a state for setting sasl->authenticated = TRUE; * camel-sasl-cram-md5.c (cram_md5_challenge): Same here. 2001-02-28 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_decode_date): A fix for broken mailers that send 2-digit years (ie "Wed, 28 Feb 01 04:59:41"). Fixes bug #1633. * camel-filter-driver.c (camel_filter_driver_filter_folder): Check to make sure message isn't NULL. 2001-02-28 Dan Winship <danw@ximian.com> * camel-remote-store.c (remote_recv_line): If we get back 0 bytes from camel_stream_buffer_gets, that means the socket got disconnected, so set an exception. (Noticed because of a message that crashes our IMAP server when you try to look at it. :) 2001-02-28 Jeffrey Stedfast <fejj@ximian.com> * Makefile.am: Added camel-sasl*.[c,h] to the build. * camel-sasl.[c,h]: new "abstract" SASL class * camel-sasl-kerberos4.[c,h]: new SASL class for KERBEROS_V4 * camel-sasl-cram-md5.[c,h]: new SASL class for CRAM-MD5 * camel-sasl-plain.[c,h]: new SASL class for PLAIN * providers/imap/camel-imap-auth.c: Removed the base64 functions. * camel-mime-utils.c (base64_encode_simple): Moved here from camel-imap-auth.c (base64_decode_simple): Same. 2001-02-26 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (imap_sync): Group messages into sets with the same flags and use the IMAP message set notation rather than doing a separate STORE FLAGS for each one. This cuts the network traffic down to just a handful of commands per sync rather than one per changed message. Removed the progress notification stuff since it's no longer meaningful and should hopefully be unnecessary. (imap_copy_message_to): move the former sync_message function into here, since it's no longer shared with imap_sync. 2001-02-26 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c: Made thread-safe and moved to above the test code. * camel-mime-message.c (camel_mime_message_init): Set the message_id to NULL. (camel_mime_message_finalize): Free the message_id. (camel_mime_message_set_message_id): New function to set the Message-Id. (camel_mime_message_get_message_id): New function to get the Message-Id. (process_header): Decode the message-id. 2001-02-24 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (construct): init the vTrash folder here instead of in store_init. (camel_store_class_init): init service_class->construct. (camel_store_finalize): We only need to unref the vtrash folder. (init_trash): don't create a vstore since we don't need it. 2001-02-23 Jeffrey Stedfast <fejj@ximian.com> * camel-store.c (camel_store_class_init): init get_trash and init_trash. (camel_store_init): call the virtual init_trash function. (camel_store_finalize): unref the vtrash folder/store if present. (camel_store_get_folder): Add the new folder to the vtrash vfolder. (init_trash): default implementation of init_trash. (get_trash): default implementation of get_trash. * camel-vee-store.c (vee_init_trash): No-op. (vee_get_trash): return NULL 2001-02-23 Not Zed <NotZed@Ximian.com> * camel-folder-summary.c (camel_message_info_new_from_header): Doh, cut and paste problem, use mlist not cc. * camel-folder.c (move_message_to): (copy_message_to): If the source folder doesn't support a summary, dont try and get the message info from it. * camel-filter-search.c (check_header): Implement a pseudo-header "x-camel-mlist" which just looks up in the message info for a match. * camel-folder-search.c (check_header): Add pseudo-header "x-camel-mlist" which can be used to match on mailing list. * providers/imap/camel-imap-folder.c (imap_sync): Add some internal progress reporting. (imap_rescan): Do some progress reporting. * camel-mime-part.c (init_header_name_table): Add the message-id to the formatted table, so we dont try and wrap it, and break it. 2001-02-22 Jeffrey Stedfast <fejj@ximian.com> * camel-folder.h: Remove the 'FIXME: remove because this isn't required...' comment for copy_message_to since we now need and cherish its existance in the filter-driver code ;-) * camel-filter-driver.c (camel_filter_driver_filter_message): Now takes a UID and CamelFolder arguments (which can be NULL) to allow for possible optimizations when copying messages between folders on the same store. (camel_filter_driver_filter_mbox): Updated (passes NULL for both since this case can't be optimized anyway). (camel_filter_driver_filter_folder): Updated to pass in the uid and folder of the message. 2001-02-22 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-summary.h (CAMEL_IMAP_SERVER_FLAGS): #define this as the set of all IMAP system flags * providers/imap/camel-imap-folder.c (sync_message): after syncing flags, set server_flags to (flags & CAMEL_IMAP_SERVER_FLAGS) rather than just "flags", since any other flags won't have been synced. Prevents the CAMEL_MESSAGE_ATTACHMENTS flag from getting lost. 2001-02-22 Not Zed <NotZed@Ximian.com> * providers/vee/*: Removed all files. * camel-private.h: Move camel-vee-private.h stuff into here. * camel-session.c: Move the vee_provider here, and initialise it always as part of a session. The vfolder provider is always available. * providers/Makefile.am: Remove "vee" from the build list, now built as part of the main library. * Makefile.am (libcamelinclude_HEADERS): (libcamel_la_SOURCES): Added camel-vee-store.[ch], camel-vee-folder.[ch]. * providers/vee/camel-vee-folder.c (vee_folder_build): Removed, not used. (camel_vee_folder_new): Removed call to build_folder. (vee_get_message): Dont try and free the mi if we didn't get it. * camel-folder.c (camel_folder_change_info_new): Added a hash table of what we have in the lists, plus moved private things into private pointer. (camel_folder_change_info_add_source): Fixed up private accesses. (camel_folder_change_info_add_uid): Changed to use a hashtable to see if we already have the uid, etc. (camel_folder_change_info_remove_uid): Similar. (change_info_remove): Likewise. (camel_folder_change_info_clear): Fixes for privatisations. (camel_folder_change_info_free): " 2001-02-21 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (fetch_medium): Fixes for IMAP4-pre-rev1 from Torsten Schulz. 2001-02-20 Not Zed <NotZed@Ximian.com> * camel-mime-utils.c (mail_list_magic): Take the ^Header: part out of the expression, since we look that up ourselves. (header_raw_check_mailing_list): When getting the match, get match 1, not match 0, we dont want the full string match. * camel-folder-summary.c (CAMEL_FOLDER_SUMMARY_VERSION): Bumped for summary change. (message_info_new): Extract the mlist info from the headers. (message_info_load): Load mlist from summary. (message_info_save): Save mlist to summary. (message_info_free): Save the mlist entry. (camel_message_info_new_from_header): Extract mailing list header. (camel_message_info_dup_to): Copy the mlist entry. (camel_message_info_free): Free mlist. (message_info_dump): Dump the mlist. * camel-folder-summary.h: Add an mlist (mailing list) string to the summary. 2001-02-19 Not Zed <NotZed@Ximian.com> * providers/vee/camel-vee-folder.c (camel_vee_folder_new): Added flags argument. (message_changed): Add support for the auto-update flag, in which case we auto-match the content as it changes. (camel_vee_folder_init): Remoive "DYNAMIC" optionality (camel_vee_folder_finalise): " (vee_expunge): As we expunge folders, re-sync their content by rerunning the query. * providers/vee/camel-vee-store.c (vee_get_folder): Pass flags to folder new. 2001-02-18 Not Zed <NotZed@Ximian.com> * providers/vee/Makefile.am (noinst_HEADERS): Added camel-vee-private.h. * providers/vee/camel-vee-private.h: New file to add locking stuff for vee folders. * providers/vee/camel-vee-folder.c (camel_vee_folder_new): Add locking around unmatched setup. (camel_vee_folder_init): Setup lock. (camel_vee_folder_finalise): Free locks. (folder_changed): Add locking. (unmatched_finalise): " (message_changed): " (vee_folder_build): " 2001-02-17 Not Zed <NotZed@Ximian.com> * providers/vee/camel-vee-folder.c (vee_folder_add): Check uid doesn't exist before adding it. 2001-02-16 Not Zed <NotZed@Ximian.com> * providers/vee/camel-vee-folder.c (camel_vee_folder_new): Create a new virtual virtual folder UNMATCHED, to store all messages that dont match other vfolders. (unmatched_folder_changed): When a vfolder issues a changed event, use its info to update the unmatched folder info. (unmatched_finalise): When a vfolder is finalised, remove any uids it has as if we had a removed event for it. 2001-02-16 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_finalise): Thaw the defaultfolder. (camel_filter_driver_set_default_folder): Thaw the defaultfolder we're about to replace, and freeze the new defaultfolder. 2001-02-15 Jeffrey Stedfast <fejj@ximian.com> * providers/vee/camel-vee-folder.c (camel_vee_folder_new): Emit the folder_created signal here. * providers/local/camel-local-folder.c (camel_local_folder_construct): Emit the folder_created signal. * providers/local/camel-local-store.c (delete_folder): On error, free the 'name' string so we don't leak. Also emit the folder_deleted signal. * providers/imap/camel-imap-store.c (subscribe_folder): Emit the folder_created signal. (unsubscribe_folder): Emit the folder_deleted signal. * camel-store.c (camel_store_class_init): Added folder_created and folder_deleted signal defs. 2001-02-15 Dan Winship <danw@ximian.com> * camel-folder-thread.c (container_parent_child): Deal with out-of-order References headers so they don't cause messages to disappear in threaded view. * providers/imap/camel-imap-folder.c (fetch_medium): IMAP4 (pre-rev1) doesn't allow BODY.PEEK[], so use RFC822.PEEK instead in that case. 2001-02-14 Christopher James Lahey <clahey@ximian.com> * camel-mime-filter-charset.c (complete, filter), camel-mime-utils.c (rfc2047_decode_word, rfc2047_encode_word): Changed this code to match the glibc iconv signature. 2001-02-14 Dan Winship <danw@ximian.com> * camel-store.c (camel_store_create_folder): New method to create a folder when you don't know its full name. * providers/imap/camel-imap-store.c (create_folder): Implement. (create_folder): Oops, and fix bug involving namespaces 2001-02-12 Not Zed <NotZed@Ximian.com> * camel-mime-part.c (init_header_name_table): Add address headers to list that we dont fold when writing. The are properly formatted as we build them. (write_to_stream): DOH, lookup the header name in the formatted hash table, not the value, that would never have worked. * camel-internet-address.c (camel_internet_address_encode_address): Changed to take a parameter saying how much we've printed so far. We now fold the header as we format it. We dont fold addresses, even if they are too long, we simply put them on another line by themselves: this should make the result more parsable by mailers that can't handle split up addresses (which are legal). (internet_encode): Fix for changes to above. 2001-02-12 Jeffrey Stedfast <fejj@ximian.com> * providers/local/camel-local-provider.c: mbox, mh, and maildir camel-url's need absolute paths. * camel-provider.h (CAMEL_URL_PATH_IS_ABSOLUTE): New flaggy flag. 2001-02-08 Dan Winship <danw@ximian.com> * camel-store.c: Rewrite a bunch. Replace the existing folder cache stuff with much simpler code that still handles all the existing cases. Now the folder hash table is always created by the base class, using hash and compare functions provided by the class implementation. (If they are set to NULL, CamelStore won't cache folders.) lookup_folder, cache_folder, and uncache_folder are no longer class methods, and get_name is gone completely. (camel_store_get_inbox): Renamed from camel_store_get_default_folder, since that wasn't being used, and this is what we actually need. (camel_store_get_root_folder): Removed, since it's not needed for anything given get_folder_info. * camel-remote-store.c: * providers/local/camel-local-store.c: * providers/local/camel-mbox-store.c: * providers/local/camel-mh-store.c: * providers/local/camel-maildir-store.c: * providers/nntp/camel-nntp-store.c: * providers/pop3/camel-pop3-store.c: * providers/vee/camel-vee-store.c: Minor updates for CamelStore changes * providers/imap/camel-imap-store.c (camel_imap_store_class_init): Update for CamelStore changes. (hash_folder_name, compare_folder_name): treat INBOX case-insensitively, otherwise use g_str_hash and g_str_equal. * camel-service.c (camel_service_construct): Remove camel_service_new and create camel_service_construct (as a class method) in its place. * camel-session.c (camel_session_get_service): Use camel_object_new and camel_service_construct to replace camel_service_new. * providers/local/camel-local-store.c (construct): Append a '/' to the URL path if it doesn't end with one 2001-01-31 Jeffrey Stedfast <fejj@helixcode.com> * camel-tcp-stream-ssl.c: Oops, include the camel-tcp-stream-ssl header instead of the raw one. 2001-02-07 Not Zed <NotZed@Ximian.com> * camel-operation.c (camel_operation_start): Changed to push the operation into a status stack. (camel_operation_progress): Changed to only accept % complete. (camel_operation_reset): Free status stack as well. * providers/pop3/camel-pop3-folder.c (pop3_get_message): Get the octect count from the return line, and pass it to get_additional_data(). (pop3_refresh_info): Added status stuff. * providers/pop3/camel-pop3-store.c (camel_pop3_command_get_additional_data): Added a total bytes expected argument for progress reporting & fixed callers. (camel_pop3_command_get_additional_data): Added progress reporting. * providers/local/camel-mbox-summary.c (mbox_summary_sync_full): (mbox_summary_sync_quick): (summary_rebuild): Added progress reporting stuff. 2001-02-07 Jeffrey Stedfast <fejj@ximian.com> * camel-search-private.c (CAMEL_SEARCH_COMPARE): Macro for making comparisons easy. (camel_ustrcasecmp): Use CAMEL_SEARCH_COMPARE and check for end-of-string before the utf-8 error check. (camel_ustrncasecmp): Same. (camel_search_header_match): Strip leading spaces before doing anything else. Also use vlen and mlen so as to not need to do 500 strlen()'s. (camel_ustrcasecmp): Don't get_utf8() for the 2 strings in the comparison part of the loop because of short-circuit expression evaluation. (blame JPR if this is the wrong term!) (camel_ustrncasecmp): Same. 2001-02-06 Jeffrey Stedfast <fejj@ximian.com> * camel-search-private.c (utf8_get): Rewrote using unicode_get_utf8(). (camel_ustrcasecmp): Use unicode_char_t (camel_ustrncasecmp): Same. 2001-02-06 Christopher James Lahey <clahey@ximian.com> * camel-mime-filter-charset.c (complete, filter), camel-mime-utils.c (rfc2047_decode_word, rfc2047_encode_word): Made these pass char ** as the second parameter to iconv. 2001-02-06 Not Zed <NotZed@Ximian.com> * camel-session.c: Removed all the camel_cancel stuff. * camel-stream-fs.c (stream_read): Change to use camel_operation. (stream_write): " * camel-remote-store.c (socket_connect): Change to use camel_operation rather than camel_cancel stuff. Removed gal include, WTF is that doing there anyway? * Makefile.am (libcamel_la_SOURCES): (libcamelinclude_HEADERS): Added camel-operation.[ch], stuff to handle notification and cancellation. * camel-search-private.c: Removed unwanted header. It was never put in for a reason. Stop fixing irrelevant warnings. (camel_ustrstrcase): Our own strstrcase impl for utf8 strings. (camel_ustrcasecmp): Ditto for strcasecmp. (camel_ustrncasecmp): And strncasecmp. (utf8_get): Simpler interface to utf8 string processing. (camel_search_header_match): Use the new things. 2001-02-05 Not Zed <NotZed@Ximian.com> * camel-folder.c (get_summary): Removed some old variables/a small memleak. (free_summary): Removed old variables. * camel-mime-utils.c (header_raw_check_mailing_list): New utility function to get the mailing list (if any) that a set of headers came from. 2001-02-05 Christopher James Lahey <clahey@helixcode.com> * camel-stream-fs.c: Added a missing #include. * camel-search-private.c: Added a missing #include. 2001-02-05 Jeffrey Stedfast <fejj@ximian.com> * camel-remote-store.c (remote_query_auth_types): Oops. Change this too. * providers/smtp/camel-smtp-transport.c (query_auth_types): Updated. * providers/nntp/camel-nntp-store.c (nntp_store_query_auth_types): Updated. * providers/pop3/camel-pop3-store.c (query_auth_types): Updated. * providers/imap/camel-imap-store.c (query_auth_types): Updated. * camel-service.c (camel_service_query_auth_types): Now takes a boolean value to specify whether or not to connect when constructing a supported authtype list. 2001-02-05 Not Zed <NotZed@Ximian.com> * camel-stream-filter.c (do_write): Revert jeff's earlier change, the change below is the more-correct fix. All his was doing was ignoring the return code & looping actually :) * camel-stream-fs.c (stream_write): Since we are non-blocking on the outgoing fd, keep looping if writing out large blocks, so we can keep checking for cancel, etc. 2001-02-02 Not Zed <NotZed@Ximian.com> * camel-folder-thread.c (camel_folder_thread_messages_new): When we drop a fake node, dont skip checking the next in the list. (camel_folder_thread_messages_new): Add an assertion check just to make sure we dont get any pseudo nodes, and spit a warning if we do. 2001-02-04 Jeffrey Stedfast <fejj@ximian.com> * camel-stream-filter.c (do_write): Make do_write loop until the entire buffer is written. 2001-02-02 Not Zed <NotZed@Ximian.com> * camel-stream-fs.c: Add missing header for cancel check stuff. * camel-session.c (camel_cancel_cancel): Fix a wrong cast. * camel-mime-part.c (init_header_name_table): Setup a new table header_formatted_table, that lists headers that we dont want to fold (they've already been folded). (write_to_stream): Check for already formatted headers, and dont try and fold them. This is a fix for bug #1097. 2001-02-01 Not Zed <NotZed@Ximian.com> * camel-mime-utils.c (header_fold): If we are folding, drop the space at the end of the line. 2001-01-30 Not Zed <NotZed@Ximian.com> * camel-remote-store.c (timeout_cb): Remvoe the stream locking here. I think this was leading to a deadlock when a keepalive is being sent, in imap. 2001-01-29 Not Zed <NotZed@Ximian.com> * providers/imap/camel-imap-command.c (imap_command_strdup_vprintf): Include the terminating NUL in the calculated string length. This hit memcheck. 2001-01-25 Not Zed <NotZed@Ximian.com> * tests/folder/test3.c: Changed the subject search to handle case sensitive when it is mixed case. * Makefile.am (libcamel_la_SOURCES): Added camel-ssearch-private.c (noinst_HEADERS): Added camel-search-private.h * camel-folder-search.c (check_header): New function to do the work of the various header checks. (search_header_matches): (search_header_starts_with): (search_header_ends_with): (search_header_contains): Use check_header to do the work. (build_match_regex): Removed. * camel-search-private.c (header_soundex): New function to match words to phrases using soundex algorithm. * camel-filter-search.c (soundexcmp): Removed. (check_match): Moved to search-private.h 2001-01-24 Not Zed <NotZed@Ximian.com> * camel-search-private.c (camel_search_build_match_regex): Added extra flags, so the same function can be used for start/end/whole/partial matches. (camel_search_header_match): Convenience function to check a single header against all sorts of different matches. * providers/imap/camel-imap-search.c (imap_body_contains): Fix for e_sexp api changes. * camel-folder-search.c: Fix for e_sexp api changes. (search_header_contains): Free args/quit on unknown header. (search_header_matches): " (search_header_starts_with): " (search_header_ends_with): " (match_message): Add an exception argument. (search_body_contains): Free args/quit on fatal error. (message_body_contains): Removed (moved to camel-search-private.c), fixed callers. * camel-filter-search.c: Fix for e_sexp api changes. (build_match_regex, message_body_contains): Moved into camel-filter-private.c Fixed callers. (check_header): moved guts to camel-search-private, and changed to use regex's for everything. Just calls that with the right args. (check_header): GEts the header, decodes it, and checks for failure, and whatnot. (check_match): Removed. (header_soundex): Changed significantly. Now it soundexes each word in the header separately, and compares it to the first argument. * tests/folder/test9.c (main): Fix for api changes. (main): Added tests to see that invalid match and action rules are properly detected. * camel-filter-driver.c (camel_filter_driver_filter_mbox): Remove the 'finished message' bit. (camel_filter_driver_filter_message): Remove an accidentally checked in debug. (camel_filter_driver_filter_message): Fix for e-sexp api changes. 2001-01-23 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-wrapper.c: Add a lock around the delayed loading, so two threads won't try to do it at the same time. 2001-01-22 Dan Winship <danw@ximian.com> * providers/vee/camel-vee-folder.c (camel_vee_folder_new): Set the CamelFolder::name to just the base part of the vfolder name (the non-query part) 2001-01-23 Not Zed <NotZed@Ximian.com> * camel-filter-search.c (camel_filter_search_match): Perform error checking on parsing/execution. * camel-folder-search.c (camel_folder_search_execute_expression): Perform error handling on search expression. (CamelFolderSearchPrivate): Add a camelexception for error returns. (camel_folder_search_execute_expression): Setup exception pointer. (search_match_all): Quit on error. * providers/imap/camel-imap-summary.c (message_info_load): Removed some debug 'warnings', as they should now be displayed at the toplevel loader, and just made the code match similar code elsewhere. * providers/local/camel-mbox-summary.c (message_info_load): Error handling. (message_info_save): more error handling. * camel-folder-summary.c (message_info_load): Add error handling and sanity checking. (camel_folder_summary_load): Add error checks. (perform_content_info_load): Error + sanity checks. (content_info_load): error + sanity checks. * camel-filter-driver.c (close_folder): avoid /0 by updating after we've done the sync. (close_folders): Setup the first progress report to start it off. (camel_filter_driver_filter_message): Fix a fixme, check errors in e_sexp parsing, etc. 2001-01-22 Christopher James Lahey <clahey@helixcode.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_selected): Initialize the value of the variable exists here. 2001-01-22 Not Zed <NotZed@Ximian.com> * .cvsignore */*/.cvsignore: Added profiling temp files. * camel-folder-thread.c (camel_folder_thread_messages_new): When removing phantom nodes, check for the end node too. * camel-filter-driver.[ch]: Changed status vars to be CAMEL_FILTER_STATUS_ etc. 2001-01-21 Jeffrey Stedfast <fejj@ximian.com> * camel-mime-utils.c (header_param_list_format_append): Make the output a little prettier. Okay, so I'm anal... 2001-01-21 Not Zed <NotZed@Ximian.com> * camel-filter-driver.h: Added an exception to CamelFilterGetFolderFunc. * camel-filter-driver.c (camel_filter_driver_filter_message): Remove source parameter, as it is determined elsewhere now. (camel_filter_driver_filter_folder): Same here. (camel_filter_driver_filter_mbox): And here. (do_move): If we cannot open a folder, quit, rather than ignoring it. (do_copy): Here too. (open_folder): Pass an exception into get_folder callback. (camel_filter_driver_filter_folder): Report that we're syncing, when we are. (camel_filter_driver_filter_mbox): And here too. (close_folder): And here. 2001-01-19 Jeffrey Stedfast <fejj@ximian.com> * camel-filter-driver.c (camel_filter_driver_filter_message): Set the deleted bit *before* copying to the default folder, this way if they are importing mail and have a rule to delete it, it doesn't get copied to the default folder un-deleted. 2001-01-19 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-store.c (get_subscribed_folders_by_hand): If LISTing a subscribed folder doesn't return a LIST response, just ignore that folder. Yay RFC2060. (subscribe_folder): add a missing UNLOCK 2001-01-19 Kjartan Maraas <kmaraas@gnome.org> * providers/smtp/camel-smtp-provider.c: Mark strings for translation. 2001-01-19 Not Zed <NotZed@Ximian.com> * camel-filter-driver.c (report_status): Add a percentage complete indicator to the status thing. (camel_filter_driver_filter_mbox): Add percentage complete to reporting. (camel_filter_driver_filter_folder): Same. 2001-01-18 Dan Winship <danw@ximian.com> * camel-store.c (camel_folder_info_build): More tweaking... skip separator characters after the namespace character. (Gets rid of the shell folder registration warning some people have had with IMAP) * providers/imap/camel-imap-store.c (imap_connect): I'm sure there was some clever reason I was storing the flags of the folder in the hash table rather than just "1", but I don't remember what it was now. Anyway, since we only ever test NULL/non-NULL, store 1, since flags is sometimes 0. * providers/imap/camel-imap-folder.c (imap_update_summary): Oops. Somewhere in one of the reorgs, the code to add new messages to the CamelFolderChangeInfo structure got removed. Fix that. 2001-01-17 Jeffrey Stedfast <fejj@ximian.com> * providers/imap/camel-imap-summary.c (message_info_load): Add some g_warnings() to help debug later if I ever get the segfaults I was getting earlier. Can't seem to reproduce them after my previous "fix". * camel-folder-summary.c (camel_folder_summary_load): Undid my change here, NotZed said it was wrong and that if it gets a NULL mi then it should abort and not continue. 2001-01-17 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.c (camel_folder_summary_load): Make sure that the message-info is non-null before we go accessing inner parts of it and/or adding it to the summary array. I may not have handled the error correctly, but it does seem to work correctly. If NotZed could double-check this it'd be great. * providers/smtp/camel-smtp-transport.c (get_smtp_error_string): New smtp error-code to string mapping function. (smtp_connect): Use the new error->string function for reporting useful errors. (smtp_helo): Same. Also use a case-insensitive strstr for 8BITMIEM as it seems Exchange sends back lowercase for that one single SMTP extension. Everything else is uppercase. I'm seeing a lack of love for that extension by Microsoft dudes... (smtp_mail): Use get_smtp_error_string. (smtp_rcpt): Same. (smtp_data): Same. (smtp_quit): Same. (smtp_rset): Same. 2001-01-16 Jeffrey Stedfast <fejj@ximian.com> * camel-folder-summary.c (camel_folder_summary_array): Add an int i so this code will compile. Should I be worried about how well this code will work? ;-) 2001-01-17 Not Zed <NotZed@Ximian.com> * camel-folder.c (free_summary): Call camel_folder_summary_array_free() to do the work. (get_summary): Use camel_folder_summary_array() to get the array atomically. These fixes allow folder/test8 to work again, and fix a sort of race where the summary size can change while we were making a copy of it. * camel-folder-summary.c (camel_folder_summary_array): Get the summary array atomically, so it can't contain empty records. (camel_folder_summary_array_free): And free it. * tests/lib/camel-test.c (die): If we are verbose & in threads, then goto sleep so we can debug. * tests/folder/test8.c (worker): Add a missing pull() for comnparing content. * camel-filter-search.c: Fix the symbol table, so match-all is an immediate function, as it should be. * tests/folder/test9.c (main): New test, tests some filtering things. * tests/message/test3.c (main): Dont use a boundary string with spaces in it. Folding can corrupt it. Maybe the folding isn't working entirely right, but anyway. * camel-session.c: Debug out the debug. * camel-filter-driver.c (camel_filter_driver_filter_folder): Plug a messageinfo leak. 2001-01-16 Dan Winship <danw@ximian.com> Delayed loading of IMAP message parts. * camel-types.h: typedef CamelMessageInfo and CamelMessageContentInfo here * camel-folder-summary.h: Add a "size" field to CamelMessageContentInfo. * camel-folder-summary.c (camel_folder_summary_content_info_new, camel_folder_summary_content_info_free): Renamed and made non-static for providers that construct their own content info. (content_info_load, content_info_save): load/save size * camel-data-wrapper.c (camel_data_wrapper_is_offline): New function to return if a data wrapper's contents are "offline". (So that, for example, we don't make thumbnails of images that haven't been loaded off the IMAP server yet.) Defaults to FALSE. * providers/imap/camel-imap-folder.c (camel_imap_folder_selected): Fix a bug in re-selecting a folder when messages have been expunged from it by another client in the meantime. (imap_get_message): Rewrite. If the message is larger than a certain size, just create a skeleton message containing CamelImapWrappers that will read parts as needed. This way, large attachments only need to be downloaded if the user looks at them, and multipart/alternative alternatives that aren't used will never be downloaded at all. (imap_update_summary): Rewrite this a bunch too to make the parsing more robust. * providers/imap/camel-imap-summary.c (CAMEL_IMAP_SUMMARY_VERSION): bump. (camel_imap_summary_new): Set build_content to TRUE. (content_info_load, content_info_save): Only save/load the content for messages that have it. (The content info gets created as a side effect of imap_get_message.) * providers/imap/camel-imap-utils.c (imap_parse_body): New routine (and helpers) to parse an IMAP 'body' FETCH response and fill in a CamelMessageContentInfo from it. * providers/imap/Makefile.am (libcamelimap_la_SOURCES, libcamelimap_la_HEADERS): add camel-imap-wrapper. 2001-01-16 Dan Winship <danw@ximian.com> * camel-tcp-stream-raw.c (stream_getsockopt, stream_setsockopt): * camel-stream-fs.c (stream_read, stream_write): * camel-remote-store.c (socket_connect): fcntl(fd, F_GETFL) returns the flags as the return value, not via a passed in pointer. And F_SETFL looks for an int, not a long, and you have to pass it what it's expecting because it's a va_arg parameter. (Yes, the man page lies on Linux. But check the UNIX98 spec or the glibc source.) Also, fix another bug in socket_connect: if we manage to connect right away, unset O_NONBLOCK so it doesn't mess us up later. Fixes a bunch of problems with non-blocking I/O being done in the allegedly-blocking case and then returning EWOULDBLOCK. 2001-01-16 Chris Toshok <toshok@ximian.com> * providers/Makefile.am (NNTP_DIR): set to nntp if ENABLE_NNTP (SUBDIRS): use $(NNTP_DIR) 2001-01-16 Jeffrey Stedfast <fejj@ximian.com> * providers/smtp/camel-smtp-transport.c (smtp_rcpt): Don't check errno as it's not being set. Fixes bug #1150. 2001-01-16 Jeffrey Stedfast <fejj@ximian.com> Ugh, this design is ugly like my butt. * camel-session.c (camel_session_query_cert_authenticator): New function which will be useful when we integrate SSL. * camel-tcp-stream-ssl.c (ssl_auth_cert): Callback to authenticate a certificate. (ssl_bad_cert): Callback that gets the reason the certificate is bad and then calls camel's cert-authenticator callback to notify the user and to allow the user to override the check. (stream_connect): Set the URL we expect to connect with and setup the auth_cert and bad_cert callbacks. (camel_tcp_stream_ssl_new): Now takes a CamelSession and a expected_host argument that it will use for certificate authentication. (camel_tcp_stream_ssl_finalize): Unref the session and free the expected_host. 2001-01-16 Not Zed <NotZed@Ximian.com> * camel-filter-search.c (header_exists): Changed to support multiple args (or'd together). (header_contains): Cleaned up to match the search code. Why did fejj change it? I'll never know. (header_matches): (header_starts_with): (header_ends_with): Big cleanup of fejj's "i'm the cut & paste king" code. Also properly handle or'ing of additional args to match what the folder-search code should do. (check_match): New function which does the annoying matching stuff (for header matches). (check_header): Similarly, handles or'ing of the matches together. (header_contains): (header_matches): (header_starts_with): (header_ends_with): Call check_header to do the actual work. (header_soundex): And here too. (match_all): Yeah like match-all isn't passed expression results, its passed expression terms. Fix this so match-all works like it should, by executing the contained expression. (message_body_contains): Copied directly from camel-folder-search.c, a more robust/faster/simpler body search code. (mime_part_matches): Removed entirely. (handle_multipart): Removed entirely. (build_match_regex): Copied from camel-folder-search. Builds a set of simple strings into a regex pattern that matches any of them (for faster & simpler matching). Expanded to accept regex patterns itself, so it can merge them together. (body_contains): Use build match/match message to match using a built regex. (body_regex): Likewise, this time we tell it we're building a regex though. (header_full_regex): Use build_match_regex to take the drudgery out of it, and expand it to handle multiple regex's at once. (get_full_header): slightly cleaner (well i dunno, the sprintf stuff just got to me). (header_regex): Cleaned up to use build_match_Regex too, and to properly check types. (filter_message_search): Just allocate 'fms' on the stack. * camel-filter-driver.c (camel_filter_driver_finalise): (camel_filter_driver_init): (camel_filter_driver_class_init): (camel_filter_driver_get_type): Changed from gtk object to camel object. (camel_filter_driver_add_rule): New function to add a rule to be processed in sexp form. (camel_filter_driver_init): Init the rules list. (camel_filter_driver_finalise): Clear the rules/rules list. (camel_filter_driver_filter_message): Scan rules list directly rather than creating on the fly. * Makefile.am (libcamelinclude_HEADERS): Added camel-filter-driver.h (libcamel_la_SOURCES): Added camel-filter-driver.c, code taken from filter-driver, which can drive, uh, filters based on sexp's. (libcamelinclude_HEADERS): (libcamel_la_SOURCES): Added camel-filter-search.[ch] * camel-folder-summary.c (camel_folder_summary_decode_string): Chganged len back to be unsigned. And do a simple range check on the string value to try and detect corrupted summary files. * providers/imap/camel-imap-command.c (imap_read_untagged): Handle cancelled stream reads with an appropriate exception. * providers/imap/camel-imap-private.h: Fix the include-once macro. Doh, confliced with camel-private.h. * providers/imap/camel-imap-store.c (imap_store_refresh_folders): A copy of camel_remote_store_refresh_folders. We avoid locking each folder when we call it though. This should be removed when i can work out how to remove the folder lock from this function easily. * camel-stream-fs.c (stream_write): Fix n' argument of select. (stream_read): Likewise. * camel-remote-store.c (socket_connect): Bump the connect timeout upto 4 minutes. (socket_connect): Oops, fix the 'n' argument of select. * camel-session.c (camel_cancel_cancel): If we are given no cancellation node, then do it for all active ones. 2001-01-15 Jeffrey Stedfast <fejj@ximian.com> * camel-stream-fs.c (stream_close): If the close() is successful, set the fd to -1. * camel-tcp-stream-raw.c: Removed the disconnect() method. (stream_close): If the close() is successful, set the sockfd to -1. * camel-tcp-stream-ssl.c: Removed the disconnect() method. (stream_close): If the close() is successful, set the sockfd to NULL. * camel-tcp-stream.c (camel_tcp_stream_disconnect): Removed, easier to just use the close() method as it did the same thing anyway. 2001-01-15 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-raw.c (stream_getsockopt): Updated to be able to get the CAMEL_SOCKOPT_NONBLOCKING socket opt. (stream_setsockopt): Updated to be able to set the CAMEL_SOCKOPT_NONBLOCKING socket opt. 2001-01-15 Peter Williams <peterw@ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_new): Lock around the imap_rescan, which needs it. 2001-01-15 Not Zed <NotZed@Ximian.com> * camel-remote-store.c (socket_connect): A cancellable connection routine. (remote_send_string): Return cancelled exception if we were. (remote_send_stream): " (remote_recv_line): " * camel-stream-fs.c (stream_read): First cut at cancellation stuff. Its looking a bit ugly. 2001-01-15 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (stream_connect): Uses an SSL socket now although there's still a few things missing (like certificate handling and such). * camel.c (camel_ssl_init): A replacement function for camel_init() that also initializes SSL. 2001-01-14 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream-ssl.c (stream_getsockopt): Implemented. (stream_setsockopt): Implemented. * camel-tcp-stream-raw.c (stream_getsockopt): Implemented. (stream_setsockopt): Implemented. * camel-tcp-stream.c (camel_tcp_stream_getsockopt): New method. (camel_tcp_stream_setsockopt): Another new method. I think you get the idea of what these are for so I won't explain them. 2001-01-14 Jeffrey Stedfast <fejj@helixcode.com> * camel-tcp-stream-ssl.[c,h]: New CamelTcpStream class that implements nspr sockets and eventually will use nss for SSL/TLS. Currently doesn't do any SSL/TLS but it should still work. It's functionally equivalent to CamelTcpStreamRaw at the moment only it uses nspr i/o. * camel-tcp-stream-raw.[c,h]: New CamelTcpStream class that implements native sockets. Should be usable but may have some bugs yet. 2001-01-13 Jeffrey Stedfast <fejj@ximian.com> * camel-tcp-stream.[c,h]: New abstract class for TCP streams. The next step is to write child classes (CamelTcpStreamBSD and CamelTcpStreamSSL). The BSD stream will use normal BSD sockets while SSL will use Mozilla's NSPR and NSS libraries to do Secure Socket Layers which wille ventually replace CamelStreamSSL which I just commit'd. Oh well. 2001-01-13 Jeffrey Stedfast <fejj@ximian.com> * camel-stream-ssl.[c,h]: New stream for handling SSL/TLS connections. 2001-01-12 Not Zed <NotZed@Ximian.com> * camel-folder-thread.c (camel_folder_thread_messages_new): Perform a final pass, removing any pseudo nodes we added earlier. Quick patch to test the idea. (camel_folder_thread_messages_new): Added casts to rmeove some warnings for the analites out there. 2001-01-11 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_selected): Fix a bug in previous commit: don't check for deleted messages if there are no known messages in the folder (because it would end up sending "FETCH 0 ..."). (imap_copy_message_to): Fix this up: sync flags to the server before copying so that they end up correct in the remote folder. And poke the destination folder after doing the copy so it notices the new message right away. (imap_move_message_to): Call imap_copy_message_to for most of the work rather than duplicating the code (since it's much more complicated now). 2001-01-11 Dan Winship <danw@ximian.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_selected): New function to check for added/deleted messages when re-selecting a folder. (imap_refresh_info, imap_rescan, imap_update_summary): Various locking tweaks that turn out to be irrelevant since command_lock is recursive, but whatever. * providers/imap/camel-imap-command.c (camel_imap_command): When selecting a new folder, call camel_imap_folder_selected on it. * camel-folder.c (camel_folder_copy_message_to): Fix a locking problem. (One branch of an if statement was returning with the lock still locked.) Also remove the deprecation comments, to match move_message_to. 2001-01-10 Jeffrey Stedfast <fejj@helixcode.com> * providers/local/camel-local-provider.c: Changed the descriptions for mh, mbox, and maildir to be more "user-friendly". 2001-01-10 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-types.h: New header to prevent recursive #include problems * providers/imap/*.h: Replace some #includes with #include "camel-imap-types.h", remove typedefs that were moved to camel-imap-types.h * providers/imap/*.c: Add #includes to make up for #includes removed from .h files * providers/imap/Makefile.am (libcamelimapinclude_HEADERS): Add camel-imap-types.h 2001-01-10 Dan Winship <danw@helixcode.com> * camel-store.c (camel_store_sync): New class function, parallel to camel_folder_sync. (The default implementation just calls camel_folder_sync on each cached folder.) * providers/imap/camel-imap-store.c (get_folder_info): Call camel_store_sync before doing anything else so that the IMAP server and Camel are working from the same data. Don't ask the server for the unread message count of the current folder, since UW will return often-incorrect cached data, and we can calculate it without talking to the server anyway. 2001-01-09 Dan Winship <danw@helixcode.com> Mostly IMAP changes. Use the NAMESPACE extension (where available). Deal with servers that don't return LIST flags in response to LSUB (like UW) to get rid of the "not a selectable folder" error messages in the UI. Take advantage of the \Marked and \Unmarked flags to try to speed up the folder scan by not doing STATUS on unmarked folders. Some further tweaks on the shape of the resulting folder tree in various situations... * camel-store.h: Remove the (read) message_count, since nothing uses it, and we can speed up IMAP a bit this way. * camel-store.c (camel_folder_info_build): Redo this a bit to make it more useful for IMAP since that's the only thing that uses it. * camel-remote-store.c (camel_remote_store_connected): Public function to check if the store is connected, and try to connect it if it's not. (remote_send_string, remote_send_stream, remote_recv_line): Use that. * providers/imap/camel-imap-store.c (camel_imap_store_finalize): fix up for changes. (camel_imap_store_init): Initialize subscribed_folders to NULL rather than an empty hash table. (imap_connect): Get the list of subscribed folders here. If the server doesn't claim that any of them are either Marked or Unmarked, then assume that it doesn't do that for LSUB and remember that for later. If the server supports the NAMESPACE extension and the user didn't specify a namespace, use the server-provided one. (imap_disconnect): Free the list of subscribed folders, and the namespace. (get_folder): check camel_remote_store_connected (get_folder_info): check camel_remote_store_connected. Add a bunch of new cleverness. If we learned that the server doesn't do LSUB usefully, do a bunch of LISTs by hand. Then, if we're getting unread counts, only do it for folders that weren't listed as Unmarked. Also, deal with namespaces that end with the separator character, and update for changes to camel_folder_info_build. (folder_subscribed): Add a g_return_val_if_fail. (subscribe_folder, unsubscribe_folder): check camel_remote_store_connected. * providers/nntp/camel-nntp-store.c (build_folder_info, build_folder_info_from_grouplist, nntp_store_get_folder_info): Don't fill in message_count since it doesn't exist any more. 2001-01-09 Dan Winship <danw@helixcode.com> Kill off a long-hated Camel kludge: "empty" URLs and query_auth_types_generic. * camel-url.c: Remove "empty" from CamelURL. (camel_url_new): No longer set it. (camel_url_to_string): Treat "" as equivalent to NULL for authmech. (Unrelated change, but it simplifies some stuff with the new config dialog.) * camel-service.c (camel_service_new): Remove url->empty check: if the URL isn't valid, we don't create the service. (camel_service_query_auth_types): No longer need to switch between generic and connected variants. * providers/smtp/camel-smtp-transport.c (query_auth_types): * providers/pop3/camel-pop3-store.c (query_auth_types): * providers/nntp/camel-nntp-store.c (nntp_store_query_auth_types): * providers/imap/camel-imap-store.c (query_auth_types): * camel-remote-store.c (remote_query_auth_types): Remove generic version, rename connected version. 2001-01-06 Not Zed <NotZed@HelixCode.com> * providers/vee/camel-vee-folder.c (vee_folder_build_folder): Dont try and unref a messageinfo that wasn't there, and use the right free function too. Modified patch from Iain. (vee_move_message_to): New method, call the real folders' move_message_to method. 2001-01-04 JP Rosevear <jpr@helixcode.com> * tests/lib/Makefile.am: fix includes for compilation * tests/message/Makefile.am: Build test3 again * tests/message/test2.c: Kill warnings with header includes * tests/message/test3.c: ditto * tests/lib/streams.c: ditto * tests/lib/camel-test.c: ditto * tests/lib/messages.c: ditto * tests/lib/folders.c: ditto 2001-01-04 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-folder.c (imap_rescan): Fix two problems in figuring out server-expunged messages. 2001-01-04 Not Zed <NotZed@HelixCode.com> * camel-folder.c (thaw): If we have a lot of messages changed, promote it to a folder changed event. 2000-01-03 Jeffrey Stedfast <fejj@helixcode.com> * camel-folder-search.c (search_header_matches): Implemented. (search_header_exists): Implemented. (search_header_starts_with): Implemented. (search_header_ends_with): Implemented. 2001-01-03 Christopher James Lahey <clahey@helixcode.com> * providers/imap/camel-imap-folder.c (camel_imap_folder_new): Initialize the exists variable so that we don't do random things if the imap server misbehaves. 2000-12-29 Not Zed <NotZed@HelixCode.com> * camel-folder.c (camel_folder_move_message_to): REmove warning, this isn't really deprecated, sigh. * camel-mime-utils.c (header_fold): Comment out some debug. 2001-01-02 Christopher James Lahey <clahey@helixcode.com> * providers/maildir/, providers/mbox/, providers/mh/: Removed these directories since they're unused. 2001-01-02 Dan Winship <danw@helixcode.com> IMAP randomness. * providers/imap/camel-imap-stream.[ch]: Remove. To be replaced. * providers/imap/Makefile.am (libcamelimapinclude_HEADERS): Remove camel-imap-stream.h * providers/imap/camel-imap-utils.c (imap_parse_flag_list): Take a char ** instead of char *, to return the position at the end of parsing like the string parsing functions. (imap_parse_string_generic): New function to parse a string, nstring, or astring. (imap_parse_nstring, imap_parse_astring): Now macros (imap_parse_string): Added * providers/imap/camel-imap-folder.h: Remove the "exists" field from CamelImapFolder. * providers/imap/camel-imap-folder.c: Remove unused include of camel-imap-stream.h. (camel_imap_folder_init): Remove no-longer-relevant summary initialization. (camel_imap_folder_new): Update for imap_parse_flag_list change, exists removal, and imap_rescan. (imap_rescan): New function that does most of the work of the old imap_refresh_info, but taking "exists" as an argument instead of getting it from the folder. Also calls camel_imap_folder_changed to do the summary updating and signalling, rather than duplicating that code. (imap_refresh_info): Just call imap_rescan (using the size of the folder summary as "exists"). (imap_update_summary): Update for imap_parse_flag_list change (camel_imap_folder_changed): Update for "exists" change. 2001-01-02 Dan Winship <danw@helixcode.com> * camel-mime-utils.c (header_content_type_simple): New function to return just foo/bar with no parameters. * camel-data-wrapper.c (get_mime_type): Use header_content_type_simple rather than header_content_type_format. 2000-12-28 Dan Winship <danw@helixcode.com> * gmime-content-field.[ch]: Remove this. It was only a thin wrapper around struct _header_content_type anyway, and didn't match the naming scheme of anything else. * Makefile.am: Remove gmime-content-field.[ch] * camel.h: Remove gmime-content-field.h * camel-types.h: Add CamelContentType as a typedef for struct _header_content_type (especially for use outside of camel). * camel-multipart.c: * camel-mime-part.c: * camel-mime-message.c: * camel-folder-summary.c: * camel-folder-search.c: * camel-data-wrapper.[ch]: Use CamelContentType and header_content_type_* functions rather than the GMime stuff. * camel-mime-part-utils.c: * camel-medium.c: Remove unused gmime-content-field.h include. 2000-12-27 Dan Winship <danw@helixcode.com> * providers/pop3/camel-pop3-store.c (connect_to_server): Fix the APOP check to not crash on servers that don't return any information on the +OK greeting line. 2000-12-24 Not Zed <NotZed@HelixCode.com> * providers/imap/camel-imap-search.c (imap_body_contains): Lock the command channel while searching. (imap_body_contains): If performing a whole uid search, then add references to our own summary items, dont look it up in the folder. This way they can't vanish unexpectedly. * providers/imap/camel-imap-folder.h (CamelImapFolder): Added a private field. (CamelImapFolder): Removed summary again from here. * providers/imap/camel-imap-private.h: Added lock for imap searches. * Merge from camel-mt-branch. * providers/imap/camel-imap-folder.c (imap_update_summary): Merge fix, use the folder->summary. (imap_get_message_flags, imap_set_message_flags, imap_get_message_user_flag, imap_set_message_user_flag): Removed again. (camel_imap_folder_init): Setup private data/lock. (imap_finalize): Free private data/search lock. (imap_search_free): Lock the search_lock. (imap_search_by_expression): Lock the search lock when using the search object. Also copy/ref hte summary, rather than getting it directly. (imap_refresh_info): Free any info lookups. Use folder->summary not imap_folder->summary. And lock around commands. (camel_imap_folder_changed): Use folder->summary again. (camel_imap_folder_changed): Make sure we unref the summary lookup. 2000-12-22 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-store.c (get_folder_info): When getting the top-level folder list, include INBOX even if it's not subscribed. Don't show subscribed folders outside of the given namespace. Do a better job of pruning the namespace from the returned folder tree. 2000-12-21 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-search.c: New CamelFolderSearch subclass that just reimplements body_contains (using the IMAP SEARCH command). All other kinds of searching are done against the local summary. * providers/imap/camel-imap-folder.c (imap_search_by_expression): Use a CamelImapSearch to do searching. * providers/imap/camel-imap-utils.c (imap_translate_sexp, etc): No longer needed. * camel-folder-search.h: Add missing CAMEL_FOLDER_SEARCH_TYPE #define 2000-12-21 Dan Winship <danw@helixcode.com> * camel-stream-buffer.c (camel_stream_buffer_gets): Update the doc comment: since it always NUL-terminates the buffer, it reads at most @max-1 bytes, not @max. * camel-remote-store.c (remote_recv_line): Fix the "did camel_stream_buffer_gets fill the whole buffer" check. Fixes a bug when reading lines longer than 1024 characters (eg, IMAP SEARCH responses in very large folders). 2000-12-19 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-folder.c (imap_refresh_info): Oops. Don't do "FETCH 1:0" when the folder is empty. (imap_protocol_get_summary_specifier): Request RFC822.SIZE too. (imap_update_summary): Parse RFC822.SIZE and add it to the summary. 2000-12-18 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-summary.c: Add "guint32 server_flags" to CamelImapMessageInfo to keep track of the last known message flag state on the server. (message_info_save, message_info_load): Save/load the server_flags. * providers/imap/camel-imap-folder.c: Make this use CamelFolderChangeInfo and emit folder_changed notifications as it gets them rather than only on refresh_info. (imap_refresh_info): Notice flags that get cleared on the server as well as flags that get set. (imap_update_summary): Remove a comment that never actually applied to the committed code. 2000-12-15 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-command.c (camel_imap_command): Change the semantics of fmt: Now %S (capital S) means an IMAP "string", (which can be sent as either a quoted string or a literal). If the server supports LITERAL+, these will be sent as extended literals (which don't require any special escaping). Otherwise they'll be sent as quoted strings (and it now properly deals with " or \ in the string). (imap_command_strdup_vprintf): Utility routine that does the real work for the functionality mentioned above. * providers/imap/camel-imap-utils.c (imap_quote_string): Turns a string into a proper IMAP "quoted string". * providers/imap/camel-imap-store.c: * providers/imap/camel-imap-folder.c: Use %S instead of "%s" where appropriate. 2000-12-22 Not Zed <NotZed@HelixCode.com> * camel-folder-summary.c (GLOBAL_INFO_UNLOCK): Doh, cut and paste victim again, call unlock instead of lock! (summary_assign_uid): Unlock around trying to get a new uid, if we have a clash. 2000-12-21 Not Zed <NotZed@HelixCode.com> * tests/folder/Makefile.am (TESTS): reenabled the tests here. * providers/nntp/camel-nntp-newsrc.c (CamelNNTPNewsrc): Add a lock. Made completely mt-safe. Sigh, this is all so i can support snooping of the set_flags stuff inside camel-nntp-folder, since i've removed the global folder lock from all of the other folder summary operations. (camel_nntp_newsrc_read_for_server): Setup lock. (camel_nntp_newsrc_get_highest_article_read): Lock internal access. (camel_nntp_newsrc_get_num_articles_read): (camel_nntp_newsrc_mark_article_read): (camel_nntp_newsrc_mark_range_read): (camel_nntp_newsrc_article_is_read): (camel_nntp_newsrc_group_is_subscribed): (camel_nntp_newsrc_unsubscribe_group): (camel_nntp_newsrc_subscribe_group): (camel_nntp_newsrc_get_subscribed_group_names): (camel_nntp_newsrc_get_all_group_names): " (camel_nntp_newsrc_write_to_file): Lock internal accesses. (camel_nntp_newsrc_write): Lock around diry stuff. * providers/local/camel-maildir-summary.c (maildir_summary_check): Lock around our internal poking of the summary hashtable/array. (maildir_summary_sync): And the same here too. * camel-folder-summary.c: Moved the summaryprivate data into camel-private.h. Only needed for the locks really. (camel_folder_summary_init): Setup locks. (camel_folder_summary_set_filename): Lock. (camel_folder_summary_index): lock (camel_folder_summary_uid): Lock (camel_folder_summary_next_uid): " (camel_folder_summary_set_uid): " (camel_folder_summary_load): Lock around whole load. Hrm, probably not required. (camel_folder_summary_save): Lock around whole save, this is reqwuired, unless we ref/copy the whole list first, and i couldn't be bothreed. (camel_folder_summary_add): lock (summary_assign_uid): Lock around internal accesses. (camel_folder_summary_info_new_from_parser): Lock around filtr stuff. (camel_folder_summary_info_ref): Add locking. Could probably use its own lock? (camel_folder_summary_touch): Add locking, likewise. (camel_folder_summary_clear): Lock. If the preiovus two are changed, then so this will need mroe locking. (camel_folder_summary_info_free): Lock around refcounting. (camel_folder_summary_remove): Locking. (camel_folder_summary_remove_uid): Lock ref, unlock, unref. Also fixed a bug that would have made this never work, freeing the key, which isn't allocated. (camel_folder_summary_remove_index): lock, ref, unlock unref also. (camel_message_info_ref): Lock using a separate global lock, since we have no context here. (camel_message_info_free): Here too. (content_info_alloc): Use an alocation lock here. avoids races with setting up content_info_chunks, etc. (camel_folder_summary_info_new): And here too. (camel_folder_summary_load): Changed to use a differnet lock for loading/saving. (camel_folder_summary_init): Doh, i do need a reflock afterall, set one up. (camel_folder_summary_finalize): Free it. (camel_folder_summary_index): Add a reflock. (camel_folder_summary_uid): And here too. (camel_folder_summary_info_free): Use reflock instead of summary_lock. (camel_folder_summary_info_ref): Likewise. (camel_folder_summary_remove_uid): Add reflocking. (camel_folder_summary_remove_index): And here too. * providers/vee/camel-vee-folder.c (camel_vee_folder_new): Setup a summary object. (vee_folder_build): Build each item into the real summary object. (vee_folder_add, vee_folder_add_uid, vee_folder_add_change): New functions to add a single record to the vfolder summary. (vee_folder_build_folder): Use a real summary. (vee_get_uids): Removed, use default impl. (vee_free_message_info): " (vee_get_message_info): " (vee_get_summary): " (vee_get_unread_message_count): " (vee_get_message_count): " (vee_sync): Make vee-folder-sync sync all subordinate folders. Well i dont think this is right, but enough peopl ehave aksed for it. (vee_expunge): Same for expunge. (vee_set_message_flags): Call parent class, then mirror flags to subfolder. (vee_get_message_user_flag): Removed, we just use the default impl, and our flags should always match. (vee_set_message_user_flag): Snoop the set and set on subfolder. (vee_get_message_flags): Removed, jsut use parent impl, assume our copy matches the real ones. (get_real_message): Removed. We sort ofneed to keep the summaryitem refed while we use it, so this doesn't make as much sense as it did once. (camel_vee_folder_finalise): Unhook listening events as we unref folders. (camel_vee_folder_new): Set the summary messageinfo data size properly. (vfolder_remove_match): Fixed for summary change. (folder_changed): ditto. Also stopped changed items from being re-searched and possibly removed. Might have to resort to the old whole-search idea again. (message_changed): Similarly. Might be easier just not to bother with a whole search. * providers/vee/camel-vee-folder.h: Removes summary hashtable/array, and use a real summary object. Sigh, more work. * providers/nntp/camel-nntp-folder.c (nntp_folder_sync): Move summary to folder object. (nntp_folder_get_message_count): Removed, use default impl. (nntp_folder_set_message_flags): Call parent impl to do the work, but snoop for newsrc. (nntp_folder_get_uids): Removed, use default impl. (nntp_folder_get_summary): " (nntp_folder_get_message_info): " (nntp_folder_free_message_info): " (camel_nntp_folder_new): moved summary to folder class * providers/imap/camel-imap-folder.c (camel_imap_folder_init): Move summary to folder object. (camel_imap_folder_new): " (imap_finalize): No longer unref the summary (or infact, do anything else either). (imap_refresh_info): move summary to folder object. (imap_sync): " (imap_get_message_count): Removed, use default impl. (imap_get_unread_message_count): " (imap_get_uids): " (imap_get_summary): " (imap_free_summary): " (imap_get_message_info): " (imap_free_message_info): " (imap_get_message_flags): " (imap_set_message_flags): " (imap_get_message_user_flag): ", this changes functionality, but only by providing non-permanent flags. (imap_set_message_user_flag): " (imap_update_summary): move summary to folder object, and use camel_folder_summary_info_new_from_header, instead of calling the virtual function directly. * providers/local/camel-maildir-folder.c (maildir_append_message): move summary to folder object, again. (maildir_get_message): " * providers/local/camel-mh-folder.c (mh_append_message): move summary to folder object. (mh_get_message): " * providers/local/camel-mbox-folder.c (mbox_append_message): Move summary to folder object (mbox_get_message): " (mbox_set_message_user_flag): We need our own copy of this, sigh, so we can track if the xev needs a rewrite (probably, its only a guide anyway). (mbox_set_message_user_tag): Same same with this one. (camel_mbox_folder_class_init): Link in these new virtual functions. * providers/local/camel-local-folder.h (CamelLocalFolder): removed summary. * providers/local/camel-local-folder.c (local_get_message_count): (local_get_unread_message_count): (local_get_uids): (local_get_summary): (local_free_summary): (local_get_message_info): (local_free_message_info): (local_get_message_flags): (local_set_message_flags): (local_get_message_user_flag): (local_set_message_user_flag): (local_get_message_user_tag): (local_set_message_user_tag): Removed, all now use default implementation. (camel_local_folder_class_init): Removed overrides. (local_init): Clear folder->summary. (local_finalize): move summary to folder. (camel_local_folder_construct): " (local_sync): " * camel-folder.c (free_summary): Provide a real impl. (get_summary): " (free_uids): " (get_uids): " (free_message_info): " (get_message_info): " (set_message_user_tag): " (get_message_user_tag): " (set_message_user_flag): " (get_message_user_flag): " Sigh, the last so far. (get_message_flags): Sigh, 1 more afterall. (get_unread_message_count): And and this. (get_message_count): and this! (set_message_flags): Sigh, and this. (camel_folder_finalize): Unref the summary if we have it. (camel_folder_free_uids): Remove locking. (camel_folder_get_message_flags): Remove locks, since the summary is now mt-safe. (camel_folder_set_message_flags): " (camel_folder_get_message_user_flag): " (camel_folder_set_message_user_flag): " (camel_folder_get_message_user_tag): " (camel_folder_set_message_user_tag): " (camel_folder_get_message_info): " (camel_folder_free_message_info): " (camel_folder_get_uids): " (camel_folder_free_summary): " (camel_folder_get_unread_message_count): " (get_unread_message_count): Check we got info, no big deal, but the summary might change while we're counting, and upset the count slightly. (camel_folder_get_message_count): Remove locks. * camel-folder.h (struct _CamelFolder): Added CamelFolderSummary to the base folder object. Sigh, this is going to be a lot of work ... * camel-service.c (camel_service_init, finalise): changed connect_lock to be recursive e_mutex. * camel-remote-store.c (camel_remote_store_init, finalise): Changed stream_lock to recursive e_mutex. * camel-private.h (RemoteStorePrivate, ServicePrivate): Use recursive mutexes for connect_lock & stream_lock. Ick. Because of the imap code. * providers/imap/camel-imap-private.h: Change the command_lock to be an EMutex. * providers/imap/camel-imap-store.c (connect_to_server): Removed the command_lock stuff. If we are just connected again, it should be safe to assume we have exclusive access at this point without it (and yes, removes a deadlock). (imap_create): Remove a lock that wasn't even supposed to be there. (camel_imap_store_finalize, init): g_mutex->e_mutex(E_MUTEX_REC). Use a recursive mutex since the imap code is multiply recursivly calling some functions (sigh, it would be nice to fix this). 2000-12-20 Not Zed <NotZed@HelixCode.com> * camel-folder.c (folder_changed): Change the locks to inside the if. Stops a deadloc,k and we dont need a lock to read a value. (message_changed): Likewise. * camel.c (camel_init): Blah, comment out g_thread_init, leave it to the caller, since it can only ever be done once. * camel-folder-thread.c (camel_folder_thread_messages_new): Ref and save the folder in the thread return. Also get the summary once, and access that for the messageinfo's. Added bonus that now threads should come out in a similar order. We need to do this so that the messageinfo's dont get unref'd under us, and using the summary directly simplifies memory management. (camel_folder_thread_messages_destroy): Free the summary/unref the folder. * camel-folder-thread.h: Add a folder to CamelFolderThread struct. 2000-12-19 Not Zed <NotZed@HelixCode.com> * providers/imap/camel-imap-utils.c (imap_translate_sexp): Unref using e_sexp_unref, since it is no longer an object. * tests/lib/camel-test.c: Added a global lock, and made it thread-aware. * camel-exception.c: Add a global lock. Not sure its really needed here, but it doesn't really hurt. (camel_exception_clear): Lock around access. (camel_exception_set): Same. (camel_exception_setv): " (camel_exception_xfer): " * providers/local/Makefile.am (noinst_HEADERS): Added, for camel-local-private.h * camel-folder.c (camel_folder_search_by_expression): Removed locking here, now upto the implementing class to lock. (camel_folder_search_free): Removed the folder lock here too, for consistency really, and the locking probably wasn't needed. * providers/local/camel-local-folder.c (local_get_summary): Changed to copy/ref the summary items. (local_free_summary): Unref the summary items/free the array. (local_get_unread_message_count): Free info lookups. (local_search_by_expression): Perform more fine-grained locking, and dont lock the whole folder while searching, just the search object. Dum de dum, it *would* be possible to not even have to lock the search, by using our own copy of the search object each time. Could be investigated. (local_init): Init priaate data/lock. (local_finalize): Free private data/lock. (local_search_free): Also lock around the search object, because it accesses global data inside. * README.mt: Some info about the mt code development. * camel-folder-search.c (match_message): Put in a warning, this plain deadlocks so far, but i'm going to bed now. (camel_folder_search_finalize): Use e_sexp_unref() since its no longer an object. * camel-session.c (camel_session_init): Init private data/lock. (camel_session_finalise): Free private data/lock. (camel_session_register_provider): Lock hash insert. (service_cache_remove): Lock. (camel_session_list_providers): " (camel_session_get_service): Lock the whole function. * camel-session.h (struct _CamelSession): Added private data. * providers/imap/camel-imap-store.c (camel_imap_store_init): Setup private data. (camel_imap_store_finalize): Free private data. (connect_to_server): Lock around command calls. I dont think its needed, but it shouldn't hurt either. (imap_connect): Lock around command calls. Again, I don think it is needed, but wont hurt. (imap_disconnect): ditto for this one. (imap_create): Lock around command. (get_folder): " (get_folder_info): " (subscribe_folder): " (unsubscribe_folder): " (imap_keepalive): " * providers/imap/camel-imap-folder.c (camel_imap_folder_new): Lock around calling a command. (imap_refresh_info): Likewise. (imap_sync): " (imap_append_message): " (imap_copy_message_to): " (imap_move_message_to): " (imap_get_message): " (imap_update_summary): " (imap_search_by_expression): ", also minor logic cleanup. (imap_get_summary): Copy/ref the summary items. (imap_free_summary): And free it now too. * providers/imap/camel-imap-auth.c (imap_try_kerberos_v4_auth): Lock the command stream for the command session. * providers/imap/camel-imap-private.h: Private (locking info) stuff for imap provider. (CAMEL_IMAP_STORE_UNLOCK): Fix to call unlock, rather than lock! * camel-service.c (camel_service_query_auth_types): Lock also for the connection here, incase the service needs to connect. * camel-remote-store.c (camel_remote_store_init): Init private data/lock. (camel_remote_store_finalise): Readded, free private data/lock. (remote_send_stream): Changed to return the number of bytes written, like a good little stream function should. (camel_remote_store_send_string): Lock for stream access. (camel_remote_store_send_stream): Same here. (camel_remote_store_recv_line): And here. (camel_remote_store_refresh_folders): Lock the store for cache access. (timeout_cb): Lock for stream access. 2000-12-18 Not Zed <NotZed@HelixCode.com> * camel-service.c (camel_service_init): Init private data. (camel_service_finalise): Free private data. (camel_service_connect): Lock around connection access. (camel_service_disconnect): " * camel-service.h (struct _CamelService): Added private field. * camel-store.c (camel_store_init): Init private data/lock. (camel_store_finalize): Free private data/lock. (camel_store_get_folder): Lock internal calls. (camel_store_delete_folder): " (camel_store_rename_folder): " (camel_store_get_root_folder): " (camel_store_get_default_folder): " (camel_store_get_folder_info): " (camel_store_folder_subscribed): " (camel_store_subscribe_folder): " (camel_store_unsubscribe_folder): " (lookup_folder): Lock accesses to the folder cache. (cache_folder): " (uncache_folder): And here too. * camel-store.h (struct _CamelStore): Added a private field. * Makefile.am (noinst_HEADERS): Add camel-private.h. There is a comment in that header as to why it ins't installed. * camel-private.h: New header for private data, that subclasses can get to. * tests/lib/camel-test.c (camel_test_init): If we do have threads enabled already, dont call g_thread_init(). * tests/lib/folders.c (test_folder_counts): Add free of info. (test_folder_message): And here too. * providers/local/camel-maildir-summary.c (remove_summary): Free info. (maildir_summary_check): Free info lookups. * providers/local/camel-mh-summary.c (message_info_new): Removed, didn't add value. (remove_summary): free info after being used. (mh_summary_check): Free info lookups. (mh_summary_sync): and here too. * providers/local/camel-mbox-summary.c (summary_update): Free info lookups. (mbox_summary_sync_full): Likewise. (mbox_summary_sync_quick): And here. (mbox_summary_sync): ... and here. * providers/local/camel-maildir-folder.c (maildir_get_message): Free messageinfo lookups. * providers/local/camel-mh-folder.c (mh_get_message): Free messageinfo lookups. * providers/local/camel-mbox-folder.c (mbox_get_message): Free messageinfo lookups. * providers/local/camel-local-summary.c (message_info_new_from_message): Removed, since it wasn't doing anything useful. (message_info_new_from_parser): Same. And some other dead code. * providers/local/camel-local-folder.c (local_get_message_info): deconstify. (local_free_message_info):new func to free info. (local_get_uids): (local_get_message_flags): (local_set_message_flags): (local_get_message_user_flag): (local_set_message_user_flag): (local_get_message_user_tag): (local_set_message_user_tag): Free all info lookups. (local_expunge): Dont call camel_folder_* functions, to avoid a deadlock. * providers/nntp/camel-nntp-folder.c (nntp_folder_get_message_info): deconstify. (nntp_folder_free_message_info): Free summary item. (nntp_folder_get_message_flags): Free summary lookup. (nntp_folder_set_message_flags): and here. (nntp_folder_get_uids): And here. * providers/imap/camel-imap-auth.c: Make base64_*_simple optional with the HAVE_KRB4 flag. * providers/imap/camel-imap-folder.c (imap_free_message_info): Added a free of the message info. (imap_refresh_info): (imap_sync): (imap_refresh_info): (imap_get_unread_message_count): (imap_get_uids): (imap_get_message_flags): (imap_set_message_flags): Free any summary lookups. (imap_get_message_flags): Get the message info directly from the summary, not via the folder interface. (imap_move_message_to): Dont call folder functions directly (delete_message), as it will deadlock since we already have the lock. * providers/vee/camel-vee-folder.c (vee_free_message_info): Free/unref the messageinfo. (vee_get_message_info): First ref the internal copy, then return it. (folder_changed): Free all got message info's. (message_changed): And here. (get_real_message): And here. (vee_folder_build): and here. (vee_folder_build_folder): ... and here. * camel-folder-summary.c (camel_folder_summary_info_new): Initialise the messageinfo refcount to 1. (camel_folder_summary_info_free): Down the refcount till we need to free. (camel_folder_summary_info_ref): New function to up the refcount. (camel_message_info_dup_to): Sewt the refcount of the dest to 1. (camel_message_info_new): Set refcount to 1. (camel_message_info_new_from_header): Dont allocate the mi ourselves. (camel_message_info_free): Handle refcounting. (camel_message_info_ref): New function. (camel_folder_summary_index): Ref the messageinfo before returning it. (camel_folder_summary_uid): Likewise. (camel_folder_summary_save): Access the messageinfo array directly, rather than through accessors (saves ref/unref). (camel_folder_summary_clear): Likewise. (camel_folder_summary_remove_index): Likewise. (main): Free lookups. * camel-folder-summary.h (CamelMessageInfo): Added a refcount value. * camel-folder.c (camel_folder_free_message_info): New function to 'free' a message info. (camel_folder_get_message_info): Deconstify return. (camel_folder_lock): New (internal) function to thread-lock the folder. (camel_folder_unlock): Likewise for unlocking. (freeze): (thaw): Lock around access to folder->frozen/changed_frozen. (folder_changed): Likewise. (message_changed): Likewise. (camel_folder_init): Init private data, locks and moved frozen info. (camel_folder_finalize): Free new private data, locks and moved frozen info. (copy_message_to): Free the messageinfo after we've used it. (move_message_to): And here too. (camel_folder_sync): Lock around virtual method invocation. Just locking around any virtual method that is known to need it. This is the first cut at locking. (camel_folder_refresh_info): " (camel_folder_expunge): " (camel_folder_get_message_count): " (camel_folder_get_unread_message_count): " (camel_folder_append_message): " (camel_folder_set_message_flags): " (camel_folder_get_message_flags): " (camel_folder_get_message_user_flag): " (camel_folder_set_message_user_flag): " (camel_folder_get_message_user_tag): " (camel_folder_set_message_user_tag): " (camel_folder_get_message): " (camel_folder_get_uids): " (camel_folder_free_uids): " (camel_folder_get_summary): " (camel_folder_search_by_expression): " (camel_folder_free_summary): " (camel_folder_search_free): " (camel_folder_move_message_to): " (camel_folder_copy_message_to): " (copy_message_to): Dont call any of our own folder functions directly, otherwise we will deadlock. (move_message_to): Likewise. * camel-folder.h (CamelFolder): Added free_message_info() function & deconstify get_message_info()'s return. (CamelFolder): Add a private field (for locks). (struct _CamelFolder): Moved changed_changed/frozen into private data, since it really is private and now needs locked access. 2000-12-15 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-utils.c (header_fold): When checking to see if we need to fold the header, when we come accross a \n, make sure to start at p + 1 the next time through the loop or else we get into an infinite loop. 2000-12-13 Dan Winship <danw@helixcode.com> * Namespace cleanup: change a bunch of IS_CAMEL_* macros to CAMEL_IS_* 2000-12-13 Chris Toshok <toshok@helixcode.com> * providers/imap/camel-imap-auth.c: add #include <sys/types.h> for freebsd. 2000-12-12 Christopher James Lahey <clahey@helixcode.com> * camel-folder-summary.c, camel-folder-summary.h: Added #ifndef NO_WARNINGS around a #warning. Added (void) to the prototype and declaration of camel_message_info_new. * camel-mime-message.h: Added an include for camel-mime-filter-bestenc.h. Added a prototype for camel_mime_message_set_best_encoding. Reformatted prototypes to line up. * camel-mime-parser.c: Added #if d(!)0 around the states string lookup table since it's only used in debugging output. * camel-seekable-substream.c (stream_flush): Added a cast. * providers/imap/camel-imap-auth.c: Added #include <netinet/in.h>. * providers/imap/camel-imap-folder.c (imap_refresh_info): Made uid and flags const to fix warnings here. * providers/imap/camel-imap-store.c (get_folder_info): Made p const to fix warnings here. 2000-12-12 Dan Winship <danw@helixcode.com> * tests/lib/Makefile.am: Add missing .h file. (From campd.) 2000-12-11 Not Zed <NotZed@HelixCode.com> * providers/smtp/camel-smtp-transport.c (smtp_data): Remove use of linewrap filter. Headers are now wrapped. encode_8bit already enforces a 998 octet line limit. (smtp_data): Also fixed a memleak, we always have to unref our own copy of the filters. We also dont need to remove them manually, so dont bother. The type's an int too ... * camel-internet-address.c (internet_unformat): When scanning past quotes, remove them also. (camel_internet_address_format_address): If the name contains "'s, or ','s then strip and quotes and wrap the whole lot in one set of quotes. * Makefile.am (noinst_HEADERS): We dont want to install camel-charset-map-private.h, ever. There are probably other similar files ..? * camel-mime-part.c (write_to_stream): Fold header lines appropriately as we're writing them out. * camel-mime-utils.c (header_fold): Add a new argument, headerlen, tells it how long the associated header token is. (header_fold): Also,k check to see if we need to fold first, using a better algorithm, and also accept already-folded lines, and re-process accordingly. (rfc2047_decode_word): Add a little buffer space to iconv output for shifting overheads? (rfc2047_decode_word): finish the iconv with a null call, to flush shift state, etc. (rfc2047_encode_word): Attempt to break up long words into appropriately sized, independent, chunks. See rfc2047, section 2. (header_decode_mailbox): Dont add in extra spaces into the output if we are decoding adjacent encoded words. We can only guess this case, as some broken mailers put encoded words inside quoted words. (header_encode_phrase): Dont merge words if they are going to end up too long. Also change back ot only merge consecutive words of the same type. e.g. 'foo. blah fum.' -> "foo." blah "fum." or 'iam an. idiot' -> iam "an." idiot * camel-medium.c (camel_medium_set_header): Hrm, we actually want to call set_header, not add_header here, probably explains some duplicate X-Evolution headers i was trying to track down. Also changed the api to handle a NULL value == remove the header. * providers/local/camel-maildir-summary.c (maildir_summary_decode_x_evolution): Always return error, we dont use x-evolution for maildir. (maildir_summary_encode_x_evolution): Always return a NULL string, likewise. (maildir_summary_add): Hook in here, since the _new function doesn't have access to any flags from the caller. If we have flags, then update the filename again. Not ideal, but should work. 2000-12-08 JP Rosevear <jpr@helixcode.com> * tests/message/Makefile.am: Remove test3 from build until the files show up 2000-12-09 Not Zed <NotZed@HelixCode.com> * camel-mime-message.c (camel_mime_message_set_date): Change the sign of the default date offset when none is supplied. (camel_mime_message_set_date): Also do dst if its dst (forward 1 hour). Fixes #928 + some. 2000-12-06 Not Zed <NotZed@HelixCode.com> * tests/lib/camel-test.h (check): Change line no format so that emacs can detect it. 2000-12-06 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-folder.c (imap_refresh_info): We want to bitwise-or the server and cached flags here so that we keep the flags that we have set locally and set any additional flags that a parallel connection may have set. 2000-12-05 Jeffrey Stedfast <fejj@helixcode.com> * providers/smtp/camel-smtp-transport.c (smtp_data): If the data wrapper fails to be written to the stream, unref it and the filters before returning. 2000-12-05 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-folder.c (imap_get_message): use BODY.PEEK[] rather than RFC822, so the message doesn't get marked as \Seen. (imap_refresh_info): Fix a really really really dumb bug. 2000-12-05 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-folder.c (imap_set_message_flags): Fix the check for "flags aren't actually changing". * providers/local/camel-local-folder.c (local_set_message_flags, local_set_message_user_flag, local_set_message_user_tag): Don't emit message_changed unless the flags actually changed. * providers/nntp/camel-nntp-folder.c (nntp_folder_set_message_flags): Don't emit message_changed unless the flags actually changed. Fix the check for marked as seen. 2000-12-05 Not Zed <NotZed@HelixCode.com> * camel-seekable-substream.c (stream_flush): stream_flush does make sense for a substream afterall (if you have a stream_write). (stream_write): Implement this. (stream_seek): Change the STREAM_END behaviour to be more sane. if bounded go from the end of the bound, if unbounded, go from the end of the parent stream. * camel-stream-mem.c (stream_read): Dont return error if reading past the end of data, just return 0. * camel-stream-fs.c (camel_stream_fs_init): Initialise the stream to be unbound. (stream_seek): Fix the logic when seeking from the end of an unbounded stream. (camel_stream_fs_new_with_fd): If the fd is invalid (-1), then return NULL immediately. (stream_seek): Range check a SEEK_END so it fits within bound_start. 2000-12-01 Not Zed <NotZed@HelixCode.com> * tests/lib/folders.c (test_folder_basic): New test to perform basic store operations on folders (taken from folders/test1). (test_folder_message_ops): Tkane the guts out of folders/test2. 2000-12-04 Jeffrey Stedfast <fejj@helixcode.com> * providers/smtp/camel-smtp-transport.c (smtp_connect): i18n-ize exception strings. (get_name): And here. (_send_to): Here too. (smtp_helo): And here and there... (smtp_mail): And around the square... (smtp_rcpt): Saying catch me if you can... (smtp_data): And here three. (smtp_rset): And here. (smtp_quit): And finally here. 2000-12-01 Jeffrey Stedfast <fejj@helixcode.com> * camel-folder-search.h: Added system_flag to CamelFolderSearchClass * camel-folder-summary.c (camel_system_flag_get): Convenience function to return whether or not a flag is set using a string as the flag name. (camel_system_flag): Return the integer value of the flag string. * camel-folder-search.c (search_system_flag): New ESExp callback for allowing vfoldering on CamelMessageInfo flags. 2000-12-01 Dan Winship <danw@helixcode.com> * camel-service.c (camel_service_disconnect): Don't g_return_if_fail if the service is already disconnected. Just return. * providers/pop3/camel-pop3-store.c (pop3_try_authenticate): Return FALSE (don't try again) if we get CAMEL_POP3_FAIL. (pop3_connect): If we don't succeed, disconnect. 2000-11-30 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-store.c (parse_list_response_as_folder_info): Fix a possible bug where 'name' could be uninitialized. * camel-folder-summary.c (camel_message_info_new): New convenience function, doesn't do much but it sure makes code cleaner to read. (camel_message_info_new_from_header): This one makes my life heaven. 2000-11-30 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-summary.c (camel_imap_summary_new): Handle the case where the summary failed to load - clear the summary and then set the dirty bit so that it is sure to save later. Is this the right fix? 2000-11-30 Dan Winship <danw@helixcode.com> * camel-exception.c (camel_exception_setv): Remove unused variable. 2000-11-30 Not Zed <NotZed@HelixCode.com> * providers/local/camel-mbox-folder.c (mbox_get_message): Remove assertion that content is there, when it no longer can be. * camel-folder-summary.h: Removed pos/bodypos/endpos from camelmeessagecontentinfo. (CamelMessageFlags): Added an attachments flag. * providers/local/camel-local-summary.h: Added load virtual function. * tests/lib/folders.c (test_message_info): Accessors. (test_folder_message): " * camel-folder-thread.c (get_root_subject): Fix accessors. (dump_tree_rec): " * camel-folder-search.c (camel_folder_search_execute_expression): Accessors for messageinfo. (search_match_all): " (search_header_contains): " (search_header_contains): " (search_body_contains): " (camel_folder_search_execute_expression): Use mepool_strdup. * providers/local/camel-mbox-summary.c (summary_update): Accessors for messageinfo. (mbox_summary_sync_full): " * providers/local/camel-mh-summary.c (remove_summary): Accessors for messageinfo. (mh_summary_check): " (mh_summary_sync_message): " (mh_summary_sync): " * providers/local/camel-mh-folder.c (mh_append_message): Use accessor for uid. * providers/local/camel-local-summary.c (local_summary_decode_x_evolution): Use accessor to uid. (local_summary_encode_x_evolution): Likewise. (message_info_new): And here. (camel_local_summary_load): Call virtual load function. (local_summary_load): Default load function, load summary. (camel_local_summary_load): Check file exists before trying to load. (camel_local_summary_construct): Turn off building content info! (CAMEL_LOCAL_SUMMARY_VERSION): Bump, since we dont build content info anymore. (camel_local_summary_load): After a successful load/check, do a save too so we dont have to go through it again randomly. * providers/nntp/camel-nntp-utils.c (get_XOVER_headers): Use accessors for messageinfo. * providers/nntp/camel-nntp-folder.c (nntp_folder_get_uids): Use accessors for uid. * providers/imap/camel-imap-folder.c (imap_refresh_info): Use accessor for uid. (imap_sync): Likewise. (imap_get_uids): Likewise. (imap_update_summary): And here. * providers/vee/camel-vee-folder.c (vfolder_remove_match): Use accessor for uid. (vfolder_add_match): Handle estrv stuff. (vfolder_change_match): Accessor for uid. (get_real_message): " (vee_get_uids): " (vee_folder_build): " + estrv. (vee_folder_build_folder): " * providers/local/camel-maildir-folder.c (maildir_append_message): Use acccessors for uid's. (maildir_get_message): Here too. * providers/local/camel-maildir-summary.c (camel_maildir_summary_init): Setup the string count for us. (message_info_new): Access the string array directly. (message_info_free): No need to free string if using array. (camel_maildir_summary_info_to_name): Use accessor to get to uid. (remove_summary): And here. (maildir_summary_check): Likewise. (maildir_summary_sync): And here. (maildir_summary_load): Load up a cache of uid->filename mappings before loading the actual summary file. This saves us having to waste the diskspace storing the filenames in the summary itself, and also helps us sync the summary better on load. (message_info_load): If we have the load_map setup, and the uid exists, then set the filename cache from it, and update the flags from the name, incase our summary mismatches it. * camel-folder-summary.c (camel_folder_summary_init): Setup string count for compressed info record. An optional compile mode which stores all strings for a given messageinfo into a packed array, which should save 36-50 bytes/record. (camel_folder_summary_info_new): Init the string array. (message_info_new): Set the string array items, as required. (message_info_load): And here too. (message_info_save): Use accessors to get to strings. (message_info_free): Free strings as one. (camel_message_info_dup_to): Handle packed array case. (camel_folder_summary_add): Use accessors. And pack the strv before storing it. (summary_assign_uid): New function to assign a unique uid to a message, if it doesn't have one. (camel_folder_summary_add): Call assign_uid instead of doing it ourselves. (camel_folder_summary_info_new_from_parser): " (camel_folder_summary_info_new_from_message): " (camel_folder_summary_encode_string): constify. (camel_folder_summary_encode_token): " (summary_build_content_info_message): Fix accessors to messageinfo. (CAMEL_FOLDER_SUMMARY_VERSION): Bumped, for removal of contentinfo->pos data. (camel_folder_summary_info_new_from_parser): Calculate the size based on the parser position, not the removed contentinfo stuff. (camel_folder_summary_info_new_from_message): Remove size stuff. (camel_folder_summary_offset_content): Removed, no longer means anything. (content_info_new): (content_info_load): (content_info_save): (summary_build_content_info): Remove stuff for contentinfo->pos*. (summary_build_content_info): Take a msginfo argument, set attachments flag if we find any attachments. (summary_build_content_info_message): set attachments flag if we find any attachments. (camel_folder_summary_info_new_from_parser): Always scan the content info, even if we dont save it. (camel_folder_summary_info_new_from_message): And here too. (summary_build_content_info): Only create the contentinfo stuff if we have it turned on, otherwise just parse and discard. (summary_build_content_info_message): Likewise. 2000-11-29 Jeffrey Stedfast <fejj@helixcode.com> * camel-store.c (camel_store_get_folder): Updated the gtk-doc comment. 2000-11-29 Dan Winship <danw@helixcode.com> * camel-mime-utils.c: * camel-mime-filter-charset.c: Use iconv instead of unicode_iconv. * providers/imap/camel-imap-command.c (imap_read_untagged): Fix some off-by-one-ness. * camel-stream-buffer.c (stream_read): Fix another bug found in previously-unused code here. 2000-11-29 Ettore Perazzoli <ettore@helixcode.com> * tests/lib/Makefile.am (INCLUDES): `$(top_srcdir)' for builddir != srcdir compilation. 2000-11-29 Not Zed <NotZed@HelixCode.com> * tests/message/test2.c (main): Add a missed unref. * camel-stream-mem.c (camel_stream_mem_set_buffer): We must set ourselves as the owner of the byte-array. Weird, someone has modified this file (its been reindented), but i can't see any changelogs ... * tests/lib/messages.c (content_finalise): Fix memleak in tester, free byte array when our content object is deleted. * camel-folder-search.c (camel_folder_search_finalize): Yeah great, so the sexp is a gtk object, not a camel object. Isn't that going to be fun to fix? * camel-session.c (camel_session_finalise): Free the storage path. * providers/local/camel-local-store.c (camel_local_store_init): If store->folders is setup, free it first, then overwrite. Hmm, this seems a bit crappy to me. * camel-store.c (camel_store_init): Dont setup store->folders if its already setup. * camel-exception.c (camel_exception_setv): Removed a memleak. no need to strdup after a strdup_printf!!! * camel-address.c (camel_address_finalize): Free the address ptrarray, once finished. * providers/local/camel-local-folder.c (local_finalize): Make sure we dont leave the folder locked on close. (local_finalize): Free summary/search. * providers/local/camel-mh-summary.c (mh_summary_next_uid_string): Small memleak, always free name after using it. * camel-mime-part.c (set_content_object): Free txt after setting the header. * providers/local/camel-maildir-summary.c (maildir_summary_check): Fix a memleak, close the dir after scanning new. (message_info_free): Added so we can free the filename cached in the messageinfo. (camel_maildir_summary_finalise): Free the hostname. * tests/folder/test[12].c (main): Clear out camel-test before starting. * providers/local/camel-mbox-summary.c (mbox_summary_sync_quick): Because encode_x_evolution folds the line (sigh, because encode_param does, unlike every other function in camel-mime-utils), unfold the encoded result before comparing. (mbox_summary_sync_quick): Another small memleak, free xevnew once finished with it. * camel-mime-utils.c (header_decode_quoted_string): Removed a redundant check for c=0. (header_unfold): New function to un-fold headers. * providers/local/camel-local-summary.c (local_summary_encode_x_evolution): some problems with encoding tags, using the wrong output strings. (local_summary_encode_x_evolution): We dont need to append a ; either, param_list_format_append() will do it for us. 2000-11-28 JP Rosevear <jpr@helixcode.com> * camel-lock.c: No longer hard code the enabled lock types 2000-11-28 Dan Winship <danw@helixcode.com> * camel-remote-store.c (remote_recv_line): Don't set exception to g_strerror (errno) when nread == 0, because it won't have been set. * providers/pop3/camel-pop3-folder.c (pop3_finalize): Don't try to free things if they haven't been set yet. 2000-11-28 Jeffrey Stedfast <fejj@helixcode.com> * providers/smtp/camel-smtp-transport.c (smtp_connect): Clear the exception if EHLO fails before trying HELO in the cases where the SMTP provider doesn't announce its ESMTPness. 2000-11-28 Not Zed <NotZed@HelixCode.com> * providers/local/camel-maildir-summary.c (camel_maildir_summary_init): Set the info size's properly, oops! * tests/lib/folders.[ch]: Folder testing helpers. * tests/folder/test2.c: Test basic message ops on folders. * tests/folder/test1.c (main): Test basic folder ops on (local) stores. * providers/local/camel-local-provider.c (camel_provider_module_init): Removed some debug. * providers/local/camel-maildir-folder.c (camel_maildir_folder_class_init): fix parent class. * providers/local/camel-mh-folder.c (camel_mh_folder_class_init): Fix parent class (damn cut & paste). * providers/local/camel-maildir-store.c (get_folder): Call parent impl. (camel_maildir_store_class_init): Fix parent class setup. (delete_folder): Check the folder exists before trying to delete it. (delete_folder): Try and make the delete operation atomic/rollback failures. e.g. if one directory isn't empty, then create the other empty ones back. Also clear the tmp directory fully first. * providers/local/camel-mbox-store.c (get_folder): Call parent impl. (camel_mbox_store_class_init): parent class is camel_local_store, not camel_folder, oops. (delete_folder): Return an error if it doesn't exist, rather than covering it up. * providers/local/camel-mh-store.c (get_folder): Call parent impl. (camel_mh_store_class_init): fix parent class setup. (delete_folder): Error if it doesn't exist now. * camel-folder.c (camel_folder_move_message_to): (camel_folder_copy_message_to): Added warnings as these functions are going to be removed later. * camel-store.c (camel_store_get_root_folder): Fix for an early api change. We want CAMEL_STORE_FOLDER_CREATE, not TRUE, since its a flag. (camel_store_get_default_folder): And here too. * providers/local/camel-local-store.c (xrename): Handle renaming folders differently to renaming files. (get_default_folder_name): local stores dont have a default folder, so make it so. Or at least, it doesn't seem to make sense to have one. (get_root_folder_name): Same for root. (get_folder): Added parent implementation, that makes sure the service path exists, if we are creating a new folder (but doesn't create the folder). 2000-11-27 Not Zed <NotZed@HelixCode.com> * providers/local/camel-local-store.c (xrename): Fixed races. Use link/unlink, rather than rename, to properly detect overwriting another file. And allow some files to be missing. 2000-11-28 Radek Doulik <rodo@helixcode.com> * providers/local/camel-local-summary.c (local_summary_decode_x_evolution): add scan = scan->next; to avoid infinite loop 2000-11-27 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-command.c (imap_read_untagged): Make this gratuitously more complicated. No wait, I mean, fix bugs. Now fully handles NULs in the data stream (which "can't happen" but do) and also handles responses containing multiple literals. Also does less copying than the original code. * camel-stream-buffer.c (stream_read): Fix a bug that could make it lose sync and/or overrun buffers. 2000-11-27 JP Rosevear <jpr@helixcode.com> * providers/local/.cvsignore: shush 2000-11-27 Not Zed <NotZed@HelixCode.com> * providers/Makefile.am: Removed mh, mbox, added local, to the default. 2000-11-24 Not Zed <NotZed@HelixCode.com> * tests/data/getaddr.pl: little util to scan mailboxes for any and every address they contain. * tests/message/test2.c (main): Added a bunch of stuff to test decoding/reencoding/etc of internationalised addresses. * tests/message/lib/address-data.h: Copy of some unicode/other testing data. **Beware** of editing this file in emacs, it'll probably try and convert all the characters to something unusable. * tests/lib/camel-test.c (camel_test_break): Add a debugger hook point. * camel-mime-utils.c (quoted_encode): Check for space and convert to _ separately. (header_decode_mailbox): Fixed the 'check comments for realname' code, problem was the domain getting code was skipping all whitespace/comments before we could get a look-in. This is approximate but fairly robust. (header_decode_text): Dont use the c-type isspace func here, we want a specific whitespace only. (header_decode_text): If we have decoded words next to each other, do not insert whitespaces between them, which is what rfc2047 requires. (header_decode_text): Make c unsigned too. * tests/README: Added at least some explanation of all this stuff. * tests/lib/camel-test.h (check_msg): Added a non-gcc version of the fail command, we dont get the expression that failed, but no matter. Should be (more) portable now. (check, check_msg): Put the file/lineno in the default message. * Makefile.am (SUBDIRS): Add tests. * camel-mime-filter-basic.c (filter): Well, I'll add the extra bytes here too, lathough not strictly needed, might save a re-malloc when we get to complete(). * camel-mime-filter-charset.c (filter): Make sure we have room if we only convert very short data. (complete): and here too. * tests/Makefile.am: Initial test harness & tests. Requires gcc for this. * camel-internet-address.c (d): Turn off debug. * camel-charset-map.c (camel_charset_step): Oops, & masks for set intersection, not | them. Dunno how this got even close to working. 2000-11-23 Not Zed <NotZed@HelixCode.com> * camel-mime-filter-basic.c (filter): For base64 encoding, the output size for 0, 1, or 2 bytes of input can exceed input*2, so make sure we account for that as well. (complete): And here. (complete): Similarly for qp encoding, if we have a trailing space, we need some extra bytes (not needed for 'filter()', as any such bytes are stored in state/save). * camel-mime-utils.c (quoted_decode_step): Removed fixme not required. (quoted_encode_close): Dont append a trailing afterall. Otherwise a pass through the encode/decode will grow the message each time. 2000-11-22 Radek Doulik <rodo@helixcode.com> * camel-mime-utils.c (header_msgid_generate): check for return values 2000-11-21 Dan Winship <danw@helixcode.com> * camel-lock.c: * camel-movemail.c: add #ifdef HAVE_ALLOCA_H 2000-11-21 Not Zed <NotZed@HelixCode.com> * providers/local/camel-mh-summary.c (mh_summary_sync_message): Shite, -1 on error, >=0 on success. So i've just been truncating all the messages I touched, good one zed. (mh_summary_sync_message): Sigh, and write to the right damn fd as well. (mh_summary_sync_message): Argh, and we need to compare the length of the old xev -1 to the new xev, to check if we can optimise it. * camel-folder.c (camel_folder_change_info_new): Init the pool. (camel_folder_change_info_add_source): Allocate string in the pool. (camel_folder_change_info_add_source_list): (camel_folder_change_info_add_update): No longer free the key, as it cannot be yet. (change_info_add_uid): Add a new arg, copy, telling it whether to copy the uid argument or not, and copy using mempool_strdup. (change_info_cat): Tell add_uid to copy the string. (camel_folder_change_info_add_update): Call add_uid directly. (change_info_remove): Call add_uid directly, with no copy, and dont free the key. (change_info_free_update): No longer required since we dont malloc the keys. (camel_folder_change_info_add_uid): Fix for add_uid change. (camel_folder_change_info_remove_uid): (camel_folder_change_info_change_uid): (change_info_clear): No longer needed, just set the size to 0 on the array directly. (camel_folder_change_info_clear): Empty the arrays directly, and flush the mempool too, and also clear uid_source, incase anyone was silly enough to call us in the wrong order. (camel_folder_change_info_free): Dont bother clearing the array's contents, just free the pool and throw away all the indexes. * camel-folder.h: Added a mempool to CamelFolderChangeInfo to store the uid's we get. * camel-folder-search.c (search_match_all): If we are only matching a single info, just use that/do the search. (camel_folder_search_match_expression): New function. Matches a single message info against an expression. (camel_folder_search_init): Init a hash table used to map the returned gptrarrays' to mempools. (camel_folder_search_execute_expression): Store all of the string data in a mempool, slightly faster, less wasted space (usually),. (camel_folder_search_free_result): Check for the mempool that stores the data for the list, and free that if we have it, otherwise assume we need to use g_free() (which should only happen if the list is empty at the moment). : commented out the debugging prints. Got sick of 'executing header search' crap. * providers/vee/camel-vee-folder.c (camel_vee_folder_init): Init changes. (camel_vee_folder_finalise): Free changes. (vfolder_add_match): Simple helper to add a new matching info record. (camel_vee_folder_add_folder): Only trigger a changed event if we have changes. (vfolder_change_match): New function, changes our local vfolder info to match the source. (vfolder_add_match): Add a new info to the vfolder list. (vfolder_remove_match): Remove a no-longer matching info from the vfolder summary. (message_changed): check if the message still matches, and remove/etc as required. (camel_vee_folder_finalise, init): init/free search object. (vee_folder_build_folder): Build the changes to the folder into the changes data, as we go. (folder_changed): If the folder gave us an explicit list of changes, then process each one separately (unless there's a lot added/changed). * providers/vee/camel-vee-folder.h: Added a changes field to the folder. * Makefile.am (libcamel_la_SOURCES): Added camel-folder-thread.[ch]. * camel-folder-thread.c: message-threading algorithm, taken from evolutions' mail component, as it is generally more useful than just for evolution itself. Changed to use e-memchunks as well to speed it up a little and use less mem. 2000-11-20 Jeffrey Stedfast <fejj@helixcode.com> * camel-remote-store.c (remote_recv_line): Fixed to return the correct bytecount in all cases which is the real fix to imap_parse_nstring. * providers/imap/camel-imap-command.c (imap_read_untagged): Again, don't use strlen for the post-data, use 'n'. * providers/imap/camel-imap-utils.c (imap_parse_nstring): Undo my previous temp-fix. 2000-11-20 Not Zed <NotZed@HelixCode.com> * providers/nntp/camel-nntp-utils.c (get_XOVER_headers): Fixes for the summary messageid changes. Hash the messageid and store it. (get_XOVER_headers): Use camel_folder_summary_info_new() to create the summary item before adding it. * camel-folder-summary.h (CamelMessageInfo): Changed the messgae-id to be an 8 byte md5 hash, and the references list to be an array of these. * providers/local/camel-mh-summary.c (mh_summary_sync_message): New function, sync out the message info stuff. Only updates the X-Ev header if it can get away with it, otherwise writes out a whole new message. (mh_summary_sync): Added more functionality. All summary info is now written to the X-Ev header, etc, and new messages re-written if required during the sync process. * providers/local/camel-local-folder.c (local_set_message_user_flag): Set the XEVCHANGE flag. (local_set_message_user_tag): And here too. * providers/local/camel-local-summary.h: New flag CAMEL_MESSAGE_FOLDER_XEVCHANGE to indicate the XEV header has probably changed size and needs to be rewritten in whole. * camel-folder-summary.c (next_uid_string): Want this static, not const. (message_info_new): Store the references and message-id values as 64 bit, binary hashes. (message_info_load): fix for message-id/references changes. (message_info_save): Likewise. (camel_message_info_dup_to): And here. (camel_message_info_free): And here too. No longer free message_id, and simple free for references array. (CAMEL_FOLDER_SUMMARY_VERSION): Bumped file revision. (camel_folder_summary_init): Init memchunk allocators to empty. (camel_folder_summary_finalize): Free memchunk allocators if there. (message_info_new): Use the chunk allocator to allocate message info's. (camel_folder_summary_info_new): New helper to allocate the message info, and setup the memchunk if required. (content_info_alloc): Likewise for content info's. (message_info_load): Use summary_info_new_empty. (content_info_new): Use content_info_alloc. (content_info_load): " (content_info_free): Free the content info as a memchunk. (message_info_free): Free everything directly and the base as a memchunk, rather than calling camel_message_info_free(), which assumes a malloc'd array. * providers/local/camel-local-summary.c: Include ctype.h, kill a warning. (local_summary_decode_x_evolution): If we get a NULL message info, then dont try and set anything, just check for validity. (camel_local_summary_write_headers): New function to write a set of headers to an fd. (camel_local_summary_check): Added some statistic generation stuff for memory profiling. * providers/local/camel-mbox-summary.c (header_write): Changed to use stdoi functions to write out the header to a buffered stream, instead of using writev, which is apparently slow (and writing each line separately is slow anyway). (mbox_summary_sync_full): New implementation. Does things differently, doesn't use or require the content info stuff. (summary_rebuild): Dont return an error if we start scanning at the end of file. (mbox_summary_sync_full): If we are not writing out new headers, make sure we copy the From line as we go, and update frompos appropriately. (mbox_summary_sync_full): Always copy the From line from the existing one, rather than trying to make one up ourselves. (mbox_summary_sync): If we can get by with a quick-sync, then try it, if that fails, then try a full sync anyway. (mbox_summary_sync_quick): Quick sync. Only update system flags, etc. (mbox_summary_sync_full): Use the proper local summary encode_xev function. (header_evolution_decode): Removed, no longer needed. (header_evolution_encode): Same. (copy_block): No longer needed, removed. (header_write): Removed, replaced with camel_local_summary_write_headers. (mbox_summary_sync_full): Fixed for header_write change. * camel-mime-parser.c (folder_scan_step): Implement the new optional parser state HSCAN_PRE_FROM, that returns the (currently unfiltered) input data. (folder_scan_drop_step): Do the right thing for the PRE_FROM state. (camel_mime_parser_scan_from): Update the doco. (camel_mime_parser_scan_pre_from): Ok, make this behaviour optional, it simplifies a lot of loops that dont otherwise need to know about it. (folder_scan_step): Made the PRE_FROM state optional. (struct _header_scan_state): Made the bool vars 1 bit. (folder_pull_part): Free the from_line buffer if it is there. (folder_scan_skip_line): Added a new arg, can save the skpped data to a byte_array, as we go. (folder_scan_step): Fixed calls to skip_line approrpiately. Now we save the from line as we parse it. (camel_mime_parser_read): New function to read from the mime parser buffer directly. Useful if you use the parser to read the first/some headers, then need to scan the rest of the data, without needing to use a seek(), or allocate your own buffers. * camel-mime-parser.h (struct _header_state): Added a new parser state, pre-from which returns any data found before a from line during parsing (all other data can be retrieved by the caller except this). 2000-11-17 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-utils.c (imap_parse_nstring): When extracting a literal string, capture up until the end of the last line - this we we don't lose any data if the byte count is off. * providers/imap/camel-imap-command.c (imap_read_untagged): Use the byte-read count to decrement the number of bytes left to read rather than using strlen. Not only does this protect against a DoS (embedded NUL chars in the literal string would make strlen inaccurate) but it also improves performace a little. * camel-remote-store.c (remote_recv_line): *Sigh* Return the number of bytes read on success rather than 0. Also don't use camel_stream_buffer_read_line since we can't get an accurate octet count. 2000-11-17 Not Zed <NotZed@HelixCode.com> * providers/local/camel-local-summary.c (local_summary_add): Clear the NOXEV/FLAGGED bits, since we do have an xev header. um m, maybe this is right, this assumes a write is following. Maybe this should be done in folder::append() instead ... * camel-stream-buffer.c (camel_stream_buffer_gets): We should always terminate the string. No need to check outptr is in range, its already been checked. * providers/local/camel-mbox-summary.c (mbox_summary_sync): When we update the summary, do it from mbox_summary->folder_size, not the content info endpos (which isn't any good anymore anyway). * providers/local/camel-mbox-folder.c (mbox_append_message): Set the frompos from the current folder size, since summary_add wont have initialised it to anything useful. 2000-11-16 Not Zed <NotZed@HelixCode.com> * providers/local/camel-local-summary.c (local_summary_encode_x_evolution): Check the uid string is all digits before trying to write a 'standard' x-ev header. * providers/local/camel-maildir-summary.c (camel_maildir_summary_info_to_name): Convert an info into a maildir name:info filename. (camel_maildir_summary_name_to_info): Convert a name:info filename into an info, and tell us if it didn't match it. (message_info_new): When creating a new filename, gets its info from the flags field. Likewise if creating from an existing file, extract the flags. (maildir_summary_sync): Remove a small memleak. Also, if our flags and that requires a filename change, perform that here. (message_info_new): Get the received date from the filename. Also, dont overwirte the uid if we have one. (maildir_summary_check): Sort the summary in received order before completion. (maildir_summary_next_uid_string): Test the name for collusions before we give it out. Retry, and if that fails, well, I guess we collide :( * providers/local/camel-mbox-folder.c (mbox_lock): Implement mbox locking. (mbox_unlock): And unlocking. (mbox_append_message): Lock the folder for write before doing anything. (mbox_get_message): Lock the folder for read before doing anything. * providers/local/camel-local-folder.c (camel_local_folder_lock): Implement something here. We handle the recursive ability but pass the locking to the folder itself. (camel_local_folder_unlock): Likewise for unlocking. (local_lock): Default - do nothing, return success. (local_unlock): Same. (local_sync): Changed slightly for locking api changes, and also, only lock around the sync process itself. * camel-lock.c: New file - utility functions for locking using different strategies and/or for locking folders safely. * Makefile.am (libcamel_la_SOURCES): Adde camel-lock.[ch] 2000-11-15 Radek Doulik <rodo@helixcode.com> * camel-mime-utils.c: mime_guess_type_from_file_name moved back to composer as it introduced unwanted VFS dependency removed #include <libgnomevfs/gnome-vfs.h> 2000-11-15 Not Zed <NotZed@HelixCode.com> * providers/Makefile.am: Removed local again, not quite ready. 2000-11-14 Not Zed <NotZed@HelixCode.com> * camel-folder-summary.c (message_info_new_from_message): Use message_info_new to create the summary from headers, instead of getting directly from the message. (format_recipients): No longer required. * providers/Makefile.am (SUBDIRS): Added local. 2000-11-11 Not Zed <NotZed@HelixCode.com> * camel-mime-parser.c (folder_push_part): Make sure 'atleast' is at least 1, always. This is possibly a temporary fix for a bad failure mode on bad multipart messages. (folder_scan_content): Go until inend, not inend+1. Changed the continuation and retry logic to be simpler and more robust. If we can't find a marker within the atleast we need, just set it to 1, and try again, rather than just going to the next smaller limit (boundary check checks the length anyway). (header_append): streamline the empty line case. And strip trailing \r's if there are any (\n's already stripped). (folder_scan_header): Reordered and cleaned up a lot. Check for continuation only once, and compress lwsp then. Assume the header buffer already has end of line's stripped, and simplify some things: Only check for end of headers once and easier. Dont check to remove end of line character Dont bother testing inptr-start if we get a boundary match - it is always zero. (folder_scan_header): Removed the unused part variable, and a few pointless assignments. (folder_scan_header): Change the end limit to be 1 higher, to make sure we get all content. (folder_scan_content): And here too. (folder_scan_header): Killed a warning. (folder_push_part): Removed a bad comment. Actually boundarylenfinal can be zero for a new message not in a multipart. So we fix that here. 2000-11-09 Not Zed <NotZed@HelixCode.com> * camel-mime-utils.c (header_decode_param_list): Renamed from header_param_list_decode. (header_param_list_decode): New external function to decode a parameter list. (header_param_list_format_append): Made public. (header_param_list_format): Another new public helper function for formatting just a param list. * camel-folder-summary.c (next_uid_string): Default implementation is the same as before. (camel_folder_summary_class_init): And set it up. * camel-folder-summary.h: Make next_uid_string a virtual function. * camel-folder.c (camel_folder_change_info_changed): New function to return true if the changeset contains any changes. 2000-11-14 Jeffrey Stedfast <fejj@helixcode.com> * camel-stream.c (camel_stream_printf): Don't cast an int to a ssize_t. 2000-11-10 Dan Winship <danw@helixcode.com> * providers/mbox/camel-mbox-summary.c (camel_mbox_summary_build_from): put a space before a single-digit day of the month since it seems some mailers are fantastically picky about this. (bugs.gnome.org #27232) 2000-11-10 Radek Doulik <rodo@helixcode.com> * camel-mime-utils.c (header_msgid_generate): fix ids ending with '.' 2000-11-09 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-auth.c (imap_try_kerberos_v4_auth): Fix error handling. (Only send a "*" to bail out of authentication if the server hasn't already bailed on us.) 2000-11-08 Radek Doulik <rodo@helixcode.com> * camel-mime-utils.c (mime_guess_type_from_file_name): new function, moved from composer 2000-11-08 Not Zed <NotZed@HelixCode.com> * camel-mime-message.c (find_best_encoding): Keep track of the caller bestenc flags that make sense. * camel-mime-filter-bestenc.c (filter): Added code to detect when we have "^From " lines in the sequence of text. (camel_mime_filter_bestenc_get_best_encoding): Added a new flag CAMEL_BESTENC_NO_FROM: if set, it will not allow any lines matching "^From " to appear in the output - currently forcing base64 encoding to achieve this. * camel-mime-parser.c (folder_scan_step): Call camel_mime-filter_complete() once we're done, rather than filter_filter(). (folder_scan_content): Some fixes for state changing; so that when we do find another boundary it is properly flagged. Since we strip the last \n off all data, we must take that into account too. Sigh. Fixes a rather nasty set of bugs where multipart messages could start including following messages as parts, etc. (struct _header_scan_stack): Added new parameter, boundarylenfinal, which holds the length of the final boundary, if it is different (e.g. for From lines, whihc aren't) (folder_scan_step): Setup teh boundarylenfinal value when creating a new boundary. (folder_scan_content): Hmm, if we hit the end-of-buffer sentinal, reset the scanner back to leave 'atleast' chars in the buffer still, dump that content, and retry again. Stops us losing a check for a boundary on some data we haven't really looked at yet! (folder_scan_content): Use boundarylenfinal to calculate 'atleast'. (folder_scan_header): And here too. (folder_boundary_check): Use the atleast value directly, dont truncate it. Use the boundarylen/boundarylenfinal values directly too. (struct _header_scan_stack): Add an atleast parameter to cache the atleast info. (folder_push_part): Determine/set 'atleast', every time we add a new part. (folder_scan_header): Get the cached atleast info from the current part. (folder_scan_content): And here too. (folder_scan_header): Fix a problem where a part starting with " text" would be interpreted as a followon header wrongly. * camel-mime-filter-charset.c (complete): Add some assertions to find a bug. 2000-11-07 Ettore Perazzoli <ettore@helixcode.com> * camel-service.c (get_name): Strdup the "???" string you return, so that the expected semantics for `::get_name()' is respected. 2000-11-07 Dan Winship <danw@helixcode.com> * camel-stream-filter.c (d): Kill debugging, as it causes lots of evolution-mail spewage. 2000-11-07 Not Zed <NotZed@HelixCode.com> * camel-mime-filter-bestenc.c (complete): Implement a complete() function, now we need one. (filter): Upgraded to match rfrc2045 properly. Checks also for length of line and valid CRLF sequences. (camel_mime_filter_bestenc_get_best_encoding): Do the work of working out what is the best encoding given what we found about the stream. * camel-mime-part.c (camel_mime_part_encoding_to_string): Use a lookup table to get the encoding naem, and add the binary type. (camel_mime_part_encoding_from_string): Likewise for the reverse. * camel-mime-part.h: Added the binary encoding type, see rfc2045. * camel-mime-utils.c (header_param_list_format_append): Dont put a space before ;'s in parameter lists, makes them more readable/consistent. * camel-mime-message.c (multipart_has_8bit_parts): Cleaned up the old stuff, well removed it. (camel_mime_message_set_best_encoding): Added another argument that lets you select what you want to set the best of. i.e. for smtp transport we only need 7 bit, and dont need to optimise the charset (although of course, we should always). (find_best_encoding): Implement this feature, if we are not getting the best charset, use the one we have. (best_encoding): Set the charset on the part appropriately. Sigh, the interfaces for this are nonexistant. (find_best_encoding): Tell the bestenc filter that lf should be treated as crlf for the purposes of determining encodings. 2000-11-06 Not Zed <NotZed@HelixCode.com> * camel-charset-map.c (camel_charset_init): Init function for an iterative charset determinator. (camel_charset_step): Iterate another buffer. (camel_charset_mask): Removed, since it couldn't have worked. (camel_charset_best): Use the iterative interface to do the work. (camel_charset_best_name): Get the best name for a charset so far. * camel-mime-filter-bestenc.c: New class, a stream filter that can be used to memory-efficiently determine the best encoding and/or charset to use for a given stream of bytes. * Makefile.am (libcamelinclude_HEADERS): Added stream-null*. (libcamel_la_SOURCES): Added bestenc* * camel-stream-null.c: New class, a null-stream, that always succeeds, and never has any contents. * camel-stream.c: Minor pointless changes. Was going to do something else but changed my mind. Added trivial default implementations for all callbacks. * camel-mime-message.h: Cleaned up some old cruft. * camel-folder-summary.c (camel_folder_summary_format_address): address_list_format() no longer encodes, so we dont need to decode it. * camel-address.c (camel_address_unformat): New function, attempts to reverse the formatting process on display addresses. (camel_address_length): New function to get the number of addresses, without having to peek the structure. * camel-mime-message.c (camel_mime_message_set_from): Fix a typo. (camel_mime_message_finalize): Only unref from/reply_to if we have it. (camel_mime_message_set_recipients): New function - set the recipients as a CamelInternetAddress. This function effectively deprecates the older recipient setting functions. (camel_mime_message_add_recipient): What the hell, i'll bite the bullet. Terminate this function. The old api was ambiguious and inefficient and didn't work right anyway. (camel_mime_message_remove_recipient_address): And this one. (camel_mime_message_remove_recipient_name): And this one too. (camel_mime_message_set_recipients): If we set an empty header, then remove it from the header list. Allow a null receipient object to clear a header. (camel_mime_message_set_from): Likewise, if setting an empty from address. (camel_mime_message_encode_8bit_parts): Eeek!! camel_stream_mem_new_with_byte_array owns the byte_array we give it, so make sure we dont free any of it! (camel_mime_message_encode_8bit_parts): Infact, i'll just rewrite the whole lot, its a bit of a mess. Should really rename it and make it a little more useful too, lets see ... (best_encoding): This has a string interface? Oh boy. (camel_mime_message_foreach_part): New experimental function to iterate over all message parts. Might not remain. (camel_mime_message_has_8bit_parts): New implementation using foreach_part. Fixed a couple of problems. (find_best_encoding): New function, that finds the best encoding for a given part (will probably be moved to camel-mime-part), and also the best charset to use if it is a text part. Since one affects the other it is a two pass process, but uses streams and not memory to achieve this. (camel_mime_message_set_best_encoding): Uses the function above to configure an entire message for the best encoding possible given transport constraints. (camel_mime_message_encode_8bit_parts): Reimplemented to use the function above to perform the work. * camel-internet-address.c (camel_internet_address_format_address): Dont put <> around a lone address with no real name. (camel_internet_address_encode_address): Similarly. (internet_decode): Actually return the count of decoded addresses. (internet_unformat): Implement the unformatting routine. 2000-11-05 Not Zed <NotZed@HelixCode.com> * providers/smtp/camel-smtp-transport.c (_send_to): Changed to get the internetaddress directly, rather than having to parse it itself. * camel-address.c (camel_address_format): Added a new function which will format address, suitable for display. (camel_address_cat): Concatentate 1 camel address onto another. It is upto the caller to ensure the addresses are of compatible types. (camel_address_new_clone): New function to create a new address by copying an existing one of the same type. (camel_address_copy): New helper function to copy an address. * camel-mime-message.h (struct _CamelMimeMessage): Removed cached copy of date string. (struct _CamelMimeMessage): Added date_received info. * camel-mime-message.c (camel_mime_message_get_date_string): Removed. Nothing uses it anyway, and it is redundant. (camel_mime_message_finalize): No more date_str. (camel_mime_message_init): No more date_str, initialise date_received* (write_to_stream): Change the check for a date header. (process_header): No longer track the date_str. (camel_mime_message_get_received_date): Removed. totally invalid anyway. (camel_mime_message_get_sent_date): Removed. Redundant. The only 'date' is the sent date, the received date is just made up. (camel_mime_message_get_date): Args changed to be more consistent with utility functions. (camel_mime_message_get_date): Dont set the date when we're asked for it (if its not set by the time its written, it'll be set then). (camel_mime_message_get_date_received): Actually do 'the right thing' here, if we have a received header, use that to determine the received date. And return the data in the same format as get_date. (camel_mime_message_set_from): Changed the api to better match what we should be doing. Pass a camelinternetaddress, etc. (camel_mime_message_set_reply_to): Cahnged similarly to take an internetaddress. (camel_mime_message_get_reply_to): Likewise. (camel_mime_message_finalize): Unref the from/reply_to objects. (format_address): Removed, no longer needed. (process_header): Changed to store the from/reply_to as internetaddress's. (write_to_stream): Set the from header directly to empty, if we dont have one. Maybe we should just abort, and/or create one based on the current user. * camel-mime-utils.c (header_address_list_format): Renamed to header_address_list_encode, which is what it is actually doing. (header_address_list_format_append): Similarly. (encoding_map[]): Removed, no longer used. (header_address_list_encode_append): Take another arg, do we encode the address (for internet), or not (for display - utf8 only). (header_address_list_format): Re-added this function, but now it generates a display version only. Surprise surprise, that is all anythign needs to generate anyway. Sigh. * camel-internet-address.c (camel_internet_address_get): Return false if we get an invalid index only. (camel_internet_address_encode_address): Helper function to encode a single address for mailing. (internet_encode): Use the above function to format it. (camel_internet_address_format_address): Format a single address for display. (internet_format): Implement the display version. (camel_internet_address_class_init): Init the internet_format virtual function. (internet_cat): Implement virtual function to concatenate addresses. * camel-folder-summary.c (camel_folder_summary_info_new_from_header): new function, only build the summary info, dont add it. (camel_folder_summary_info_new_from_parser): Likewise, for new info from parser. (camel_folder_summary_add_from_parser): Cahnged to call function above to build info. (camel_folder_summary_add_from_header): Changed to call function above, to build info. (camel_folder_summary_info_free): New function to free the summary message info. (camel_folder_summary_clear): Changed to clal above to free info. (camel_folder_summary_remove): Likewise. (camel_folder_summary_add): Cleaned up the clashing uid re-assignment logic a little bit. (camel_folder_summary_decode_uint32): Fixed a typo, 01 != -1. (camel_folder_summary_decode_time_t): Return -1 on error. (camel_folder_summary_encode_off_t): New function to encode an off_t type. (camel_folder_summary_decode_off_t): And likewise for the reverse. (CAMEL_FOLDER_SUMMARY_VERSION): Bumped the summary version, since we're now encoding time/off_t's right. (summary_header_save): Use time_t encoder to save the timestamp. (summary_header_load): Likewise for decoding the timestamp. (content_info_load): Decode off_t types directly, now we can. (content_info_save): And likewise for encoding. (camel_folder_summary_add_from_message): New function, create a summary item from an existing message and add it. (camel_folder_summary_info_new_from_message): New function, create a summary item from an existing message. (summary_build_content_info_message): New function to do the dirty work of building the conent info/indexing, from a message source. (format_recipients): Format an internetaddress suitable for the summary. (message_info_new_from_message): Build a new summary item from a mime message. (content_info_new_from_message): Build a new conent info from a mime part. (camel_folder_summary_class_init): Init the new class functions. (message_info_new_from_message): Fixed for message api change. Added documentation to the functions. 2000-11-03 Radek Doulik <rodo@helixcode.com> * camel-mime-utils.c (header_msgid_generate): new function, generates simple message/content id 2000-11-04 Not Zed <NotZed@HelixCode.com> * camel-mime-part-utils.c (camel_mime_part_construct_content_from_parser): Set the preface/postface from the parser into the multipart object. * camel-multipart.c (camel_multipart_set_postface): Function to set the postface text on a multipart. (camel_multipart_set_preface): Similarly for preface text. * camel-mime-parser.c (folder_scan_content): If we scan until a boundary, then we do not include the \n that starts the boundary line in the content. (struct _header_scan_stack): Added a ByteArray to store the multipart pre/post-text as we're scanning. (folder_pull_part): Free pre/posttext if they are allocated. (folder_scan_step): Build into the pre/posttext arrays as we encounter data. (camel_mime_parser_preface): REturn the multipart preface text, if there is any scanned. (camel_mime_parser_postface): Likewise for postface text. (byte_array_to_string): helper function for above. * providers/mbox/camel-mbox-folder.c (mbox_append_message): Change the from line to be "\nFrom ..." always, so no need to check/append a \n to messages. (mbox_append_message): Open the output stream with append mode [assuming this is more efficient than seeking to the end] And dont prepend \n on the From line if its the first in the mbox. (mbox_append_message): Pass the offset of the real start of the "From " line when we perform the update (which may != 'seek') * camel-mime-filter-charset.c (complete): Removed the terminating NUL 'fix'. * camel-stream-filter.c (do_read): Added some debug. (do_flush): And here. (do_write): And here too. (do_write): ARGH!!! ARGH! Ok, so the filter stream was writing a different number of bytes than the requester was asking it to write (because of filtering, of course!). So instead of returning the true number of written bytes, we'll return what they asked us to write - unless there is an error in which case we return -1. * camel-mime-utils.c (base64_encode_close): Sigh, forgot to make it unsigned. I think this is actually a gcc bug as (48 >> 2) somehow ended up negative, when it obviously should not, even if the data load was signed. 2000-11-03 Jeffrey Stedfast <fejj@helixcode.com> * camel-url.c: Undo my incorrect fix - I misunderstood danw - sorry! 2000-11-03 Dan Winship <danw@helixcode.com> * providers/pop3/camel-pop3-store.c (camel_pop3_store_expunge): Unbreak this. 2000-11-03 Jeffrey Stedfast <fejj@helixcode.com> * camel-url.c (camel_url_to_string): If show_pass, then base64 the password before writing it to the output string. (camel_url_new): Assume password has been base64 encoded and decode accordingly. 2000-11-03 Dan Winship <danw@helixcode.com> * camel-provider.h: Add an "url_flags" field to CamelProvider. Move the CAMEL_SERVICE_URL_* defines here and remove the SERVICE_ part of the name. * camel-service.h: Remove CAMEL_SERVICE_URL_* flags and service->url_flags field. * camel-service.c (check_url, get_path): Get URL flags from service->provider, update for changed flag names. * providers/*/camel-*-provider.c: Add URL flags to provider structures. * providers/*/camel-*-{store,transport}.c, camel-remote-store.c: Remove service->url_flags initialization. 2000-11-02 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-store.c (get_folder_info): Quote foldernames when sending to the IMAP server because the folder name might contain spaces. 2000-11-02 Not Zed <NotZed@HelixCode.com> * Merged in camel-incremental-branch. 2000-11-01 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-store.c (get_folder_info): Remove old subbed folders from hash table after freeing them. * providers/imap/camel-imap-folder.c (imap_get_full_name): Deal correctly with namespace == "" 2000-11-01 Gediminas Paulauskas <menesis@delfi.lt> * camel-charset-map.c, camel-charset-map-private.h: synced all charsets with ones found in libunicode. Just include more "iso/*.h" and rebuild ...private.h with this. 2000-10-31 Dan Winship <danw@helixcode.com> * camel-service.c (service_disconnect): Add a new argument, clean, that says whether or not to try to disconnect cleanly. * camel-remote-store.c (remote_send_string, remote_send_stream, remote_recv_line): disconnect uncleanly on failure to prevent infinite loops when providers would normally send commands from disconnect(). Remove some unneeded CamelException goo. * providers/smtp/camel-smtp-transport.c (smtp_disconnect): * providers/pop3/camel-pop3-store.c (pop3_disconnect): * providers/nntp/camel-nntp-store.c (nntp_store_disconnect): * providers/imap/camel-imap-store.c (imap_disconnect): Don't send QUIT/LOGOUT if !clean. 2000-10-30 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-auth.c: New file with code for IMAP authentication mechanisms. (Currently just krb4, and without integrity/privacy protection). * providers/imap/Makefile.am: Add camel-imap-auth.[ch] and krb4 CFLAGS/LDFLAGS * providers/imap/camel-imap-store.c (connect_to_server): Split out from imap_connect. Just does the basic connect and CAPABILITY check. Redo the CAPABILITY code more robustly. (query_auth_types_connected): Do this right rather than punting to query_auth_types_generic. Check for KERBEROS_V4 if compiled with krb4 support. (query_auth_types_generic): Mention KERBEROS_V4 if compiled with krb4 support. (imap_connect): Use connect_to_server(). * camel-mime-utils.c (base64_encode_step, base64_encode_close): Take an additional argument, "break_lines", saying whether or not to add '\n's to the output. * camel-multipart.c (set_boundary): * camel-mime-filter-basic.c (filter, complete): Update for base64 api change. 2000-10-30 Peter Williams <peterw@helixcode.com> * providers/imap/camel-imap-store.c (unsubscribe_folder): Correct parameters to g_hash_table_lookup_extended. 2000-10-29 Dan Winship <danw@helixcode.com> Improved IMAP namespace handling: leave the namespace in the folder names rather than constantly prepending it and stripping it off. Also some subscription fixes. * camel-store.c (camel_folder_info_build): Fix for the case where @top isn't in @folders. * providers/imap/camel-imap-folder.c (camel_imap_folder_new): Add a "short_name" argument rather than figuring it out ourselves. (imap_get_full_name): Implementation of CamelFolder::get_full_name that strips off namespace so the user doesn't have to see it. (imap_append_message, imap_copy_message_to, imap_move_message_to): Use folder->full_name rather than calling camel_imap_store_get_folder_path. * providers/imap/camel-imap-utils.c (imap_parse_list_response): Update this: make @flags a bitmask and @sep a char rather than a string. Make all of the out arguments optional. Handle literals in the server response. * providers/imap/camel-imap-store.c (imap_connect): Do a better job of getting the correct dir_sep for the namespace we're using. Construct a base_url here that will be used by get_folder_info. (camel_imap_store_folder_path): Removed (imap_folder_exists): Add an argument to return the short name of the folder (parsed out of the LIST response). Update for imap_parse_list_response change. (get_folder): Update for the various other changes. (get_folder_info): Update for the various other changes. Be more consistent about the returned layout: put everything underneath the "namespace" directory, including INBOX, even if it doesn't belong there. Don't destroy the list of subscribed folders until we've actually gotten the new list. (folder_subscribed, subscribe_folder, unsubscribe_folder): Use folder_name directly rather than camel_imap_store_folder_Path. * providers/imap/camel-imap-command.c (camel_imap_command): Update for folder name changes. 2000-10-29 Dan Winship <danw@helixcode.com> * camel.h: Remove md5-utils.h include since it's not part of Camel any more. * camel-charset-map.c: Kill some warnings. * providers/nntp/camel-nntp-grouplist.c (camel_nntp_get_grouplist_from_file, camel_nntp_grouplist_save): Clean up warnings about time_t casts. * providers/smtp/camel-smtp-transport.c: Remove unused md5-utils.h include. * providers/pop3/camel-pop3-store.c: Undefine the "_" macro defined by krb4's des.h when compiling with krb support. Fix md5-utils.h include. 2000-10-27 Dan Winship <danw@helixcode.com> * camel-mime-utils.c (header_param_list_format_append): Only quote Content-type parameters when the quoting is mandatory, and deal with embedded quotes/backslashes when quoting. 2000-10-27 <jpr@helixcode.com> * providers/pop3/Makefile.am: Tidy up build * providers/smtp/Makefile.am: ditto * Makefile.am: Move md5-utils.[hc] to e-util because the addressbook is going to use md5 hashes for pilot syncing. Maybe the calendar conduits as well because this is a good idea Chris had. 2000-10-26 Kjartan Maraas <kmaraas@gnome.org> * camel-exceptions-list.def: Add newline to kill warnings. 2000-10-25 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-store.c (get_folder_info): Fix folder listing code infinite loop. * camel-store.h: Add a "parent" field to CamelFolderInfo. * camel-store.c (camel_folder_info_build): Deal with "parent" (camel_store_folder_subscribed, camel_store_subscribe_folder, camel_store_unsubscribe_folder): Add g_return_if_fails checking that the folder supports subscriptions. * providers/imap/camel-imap-store.c (folder_subscribed, subscribe_folder, unsubscribe_folder): Remove "+ 1"s since the mail subscribe UI won't prepend / to the folder names now. (get_folder_info): Clear the "parent" field of the folderinfos when removing an empty top level. * providers/nntp/camel-nntp-store.c (nntp_store_folder_subscribed, nntp_store_subscribe_folder, nntp_store_unsubscribe_folder): Remove "+ 1"s since the mail subscribe UI won't prepend / to the folder names now. 2000-10-24 Chris Toshok <toshok@helixcode.com> * providers/imap/camel-imap-store.h: add subscribed_folders. * providers/imap/camel-imap-store.c (camel_imap_store_class_init): fill in vtable entries for subscription functions. (camel_imap_store_finalize): new function, so we can free up our subscribed_folders hashtable. (camel_imap_store_init): set CAMEL_STORE_SUBSCRIPTIONS in the CamelStore flags, and create our subscribed_folders hashtable. (camel_imap_store_get_type): camel_imap_store_finalize is our finalize function. (get_folder_info): if we're looking at subscribed_only, clear out the subscribed_folders hashtable, use LSUB instead of LIST, and insert folder paths (prepended by their namespace if there is one) into subscribed_folders. INBOX subscription support needs work, since we always show it, regardless of it's subscribed state. (folder_subscribed): new function. just look up the folder_path in the hashtable. (subscribe_folder): new function. use the imap SUBSCRIBE command, and if successful add it to the hashtable. (unsubscribe_folder): new function. use the imap UNSUBSCRIBE command, and if successful remove it from the hashtable. 2000-10-24 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-store.c (get_folder_info): Fill in the message_count and unread_message_count flags (if !fast). 2000-10-23 Dan Winship <danw@helixcode.com> * camel-object.h: #include gnome-i18n.h (and gnome-defs.h since the former depends on it.) * *: Add lots of _() and N_(). 2000-10-23 Chris Toshok <toshok@helixcode.com> * providers/nntp/Makefile.am (libcamelnntpinclude_HEADERS): add camel-nntp-types.h. 2000-10-23 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-command.c (imap_read_untagged): Fix a "how could this have been working before" memory overrun bug found by Vlad. * camel-op-queue.[ch], camel-thread-proxy.[ch]: These should have gone away a long time ago. 2000-10-20 Chris Toshok <toshok@helixcode.com> * providers/Makefile.am (SUBDIRS): re-enable the nntp provider. 2000-10-20 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-store.c (build_folder_info): add function to build a tree from the nntp group names (using '.' as a heirarchy separator.) #ifdefed INFO_AS_TREE only. (build_folder_info_from_grouplist): if INFO_AS_TREE is selected, call build_folder_info instead of appending a new CamelFolderInfo to our list. (ensure_news_dir_exists): use e_mkdir_hier instead of failing if a parent directory (~/evolution/news generally) isn't there. 2000-10-19 Chris Toshok <toshok@helixcode.com> * camel-folder-search.c: #include <sys/types.h> before <regex.h> 2000-10-19 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-utils.c (imap_parse_list_response): Fix a bug with "INBOX" (or anything else with NIL hierarchy separator) as the namespace. * providers/imap/camel-imap-folder.c (imap_refresh_info): Emit message_changed and folder_changed as appropriate. 2000-10-19 Dan Winship <danw@helixcode.com> * camel-folder.c (camel_folder_refresh_info): Update comment here so refresh_info isn't just for reconnects any more. Make the default implementation a no-op rather than an error. * providers/nntp/camel-nntp-folder.c: Move refresh_info impl into camel_nntp_folder_new, since it would have leaked memory and not done anything useful if it was called later. * providers/mbox/camel-mbox-folder.c: Remove no-longer-necessary refresh_info impl. * providers/imap/camel-imap-folder.c (camel_imap_folder_changed): Update imap_folder->exists, but don't actually load the new messages. This is a temporary workaround to deal with the IMAP provider stealing the message list focus at annoying times. (imap_copy_message_to, imap_move_message_to): Emit a folder_changed by hand, for now. 2000-10-19 Ettore Perazzoli <ettore@helixcode.com> * providers/imap/Makefile.am (libcamelimapinclude_HEADERS): Add `camel-imap-stream.h'. * Makefile.am (libcamelinclude_HEADERS): Add `camel-charset-map.h' and `camel-charset-map-private.h'. 2000-10-18 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-part.c (camel_mime_part_set_description): The correct fix this time - the description should be encoded here and not in the camel-medium layer. 2000-10-30 Not Zed <NotZed@HelixCode.com> * camel-folder.c (camel_folder_change_info_add_update): Fixed the changeset logic, which was completely wrong. 2000-10-27 Not Zed <NotZed@HelixCode.com> * camel-mime-utils.c (quoted_encode_step): Removed is_blank() stuff. (header_decode_init): When we set the 'space' bit, dont clear all the others. (quoted_encode): Put back the safemask. Yes we totally need it, see rfc 2047 section 5, parts (1) and (3). (CHARS_PSPECIAL): Remove '=' and '_' from the list of allowed characters (this is an allowed list, not a not-allowed list, like the ESPECIAL). (camel_mime_special_table): Updated for fixes to definitions. (CHARS_ESPECIAL): Added '_' to list of characters that should be encoded. 2000-10-26 Not Zed <NotZed@HelixCode.com> * providers/mbox/camel-mbox-summary.c (camel_mbox_summary_update): Use the new camel_folder_change_info_* stuff to build the update diff, rather than doing it ourselves. * camel-folder.c (camel_folder_change_info_add_source): Add a 'source' list, used to create change lists. (camel_folder_change_info_add_source_list): Add a list of uid's, convenience function. (camel_folder_change_info_add_update): Add a uid to the list of uid's in the new updated list. (camel_folder_change_info_add_update_list): Add a bunch of uid's at once for the same purpose. (camel_folder_change_info_build_diff): Take the source list, the update list, and find the differences, building on the added or removed list as appropriate. 2000-10-20 Not Zed <NotZed@HelixCode.com> * providers/mbox/camel-mbox-summary.c (removed_uids): REnamed from add_uid. (camel_mbox_summary_update): Oops, fix the reversed logic for determining the uid changesets. * camel-folder.c (message_changed): Oops, we want to change the uid, not add it. 2000-10-19 Not Zed <NotZed@HelixCode.com> * providers/mh/camel-mh-summary.c (camel_mh_summary_sync): Added a change list argument, and add expunged uid's to it. * providers/mh/camel-mh-folder.c (mh_init): Setup the change list. (mh_finalize): And free it. (mh_sync): Track changes, and send a folder_changed signal as appropriate. (mh_expunge): Likewise. (mh_append_message): Add the new uid to the change list and trigger a folder_changed event. * providers/mbox/camel-mbox-folder.c (camel_mbox_folder_new): Setup a mbox list of changes structure. (mbox_finalize): And free it. (mbox_expunge): Include the change list with the folder_changed event, and clear it off. (mbox_sync): And the same for when we are just syncing the folder. (mbox_append_message): And do the same here, after we've updated the folder. * providers/mbox/camel-mbox-summary.c (camel_mbox_summary_build_from): Use gmtime_r to get the time thread-safely. (camel_mbox_summary_sync): Added a changeinfo argument. Add any removed or changed messages to the changelists as appropriate. (camel_mbox_summary_update): Added a changeinfo argument. Genereate a list of added/removed uid's based on the difference before and after rebuilding the summary. * camel-folder.c (camel_folder_change_info_new): (camel_folder_change_info_add_uid): (change_info_add_uid): (camel_folder_change_info_remove_uid): (camel_folder_change_info_change_uid): (change_info_clear): (camel_folder_change_info_clear): (camel_folder_change_info_free): (change_info_cat): (camel_folder_change_info_cat): Bunch of utility functions for working with change info lists. (camel_folder_init): Init the change info list. (camel_folder_finalize): And free it. (thaw): Changed to pass through a list of changes, or to get the changed message uids from the camelfolderchangeinfo struct, and reset it. (folder_changed): Add the changed lists to the frozen change list if we are frozen. (message_changed): Add the message to the changed list if we are in the frozen state. * camel-folder.h (CamelFolderChangeInfo): New structure to hold information for the folder_changed event. 2000-10-18 Ettore Perazzoli <ettore@helixcode.com> * camel-mime-filter-charset.c (complete): Put a zero at the start of the outbuf. 2000-10-18 Not Zed <NotZed@HelixCode.com> * camel-mime-part.c (add_header): No, we must not encode the headers here. These interfaces ARE raw interfaces as they are defined in camel_medium. Also removed a bogus/meaningless FIXME. (set_header): Likewise here, we must not. (process_header): Removed another bogus comment. * camel-object.c (shared_is_of_type): Comment out the spitting of a big warning when we're trying to determine types from code. * providers/mbox/camel-mbox-summary.c (message_info_new_from_parser): Only call ibex funcitons if we have an index. * providers/mh/camel-mh-summary.c (camel_mh_summary_add): Only call ibex functions if we have an index. (remove_summary): Likewise. (camel_mh_summary_check): Likewise. * providers/nntp/camel-nntp-store.c (nntp_store_get_folder): get_folder -> flags argument. * providers/vee/camel-vee-store.c (vee_get_folder): create->flags. * providers/pop3/camel-pop3-store.c (get_folder): Changed create -> flags. * providers/imap/camel-imap-store.c (get_folder): Added flags argument. * providers/mh/camel-mh-folder.c (camel_mh_folder_new): Added flags argument, and fixed code appropriately. * providers/mh/camel-mh-store.c (get_folder): Added flags argument. * camel-folder-search.c (message_body_contains): Perform a regex match on the contents of messages. This wont quite work yet as message contents are encoded when written to a stream. (build_match_regex): Converts a number of strings into a regex matching pattern, escaping special chars. (match_message): match a single message from a folder, by uid. Slow. (search_body_contains): Changed to support matching where no index is supplied. Matches are performed by retrieving message contents, etc. () WTF? camel should not be including any widget headers. * providers/mbox/camel-mbox-folder.c (camel_mbox_folder_new): Added flags argument. (mbox_refresh_info): Changed into a NOP, the refresh info code moved into the new function. (camel_mbox_folder_new): If we have an index requested, build one, otherwise, remove an old one, or just dont do anything. * providers/mbox/camel-mbox-store.c (get_folder): Changed create to flags, changed code to suit. * camel-store.c (camel_store_get_folder): Changed create to flags. (get_folder_internal): And here. (get_folder): And here too. * camel-store.h (camel_store_get_folder): Change the create argument to be a flags argument. 2000-10-17 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-folder.c (imap_refresh_info): Remove cached info at the end of the summary when the folder shrinks between sessions. Also remove an untrue comment. (camel_imap_folder_new): Move the summary creation to after the folder selection again, since it depends on the uidvalidity having been set. * providers/imap/camel-imap-store.c (get_folder): Fix up summary_file to not include the namespace twice. 2000-10-17 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-part.c (set_header): Encode the header value. (add_header): Same. 2000-10-17 Not Zed <NotZed@HelixCode.com> * camel-mime-filter.c: Added some malloc check debugging stuff. * camel-mime-parser.c (struct _header_scan_state): Removed top_part, top_start, and pending. I can't even remember why they were there, and they're not used anymore. * camel-mime-filter-basic.c (filter): Forgot to up the space here too. 2000-10-14 Not Zed <NotZed@HelixCode.com> * camel-mime-filter-basic.c (complete): Ok, so we hit a fixme, 3x just wasn't enough for some sequences. 2000-10-16 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-command.c (imap_read_response): Don't imap_next_word(respbuf + 2), instead use imap_next_word(respbuf) or else we'll skip over the second token. 2000-10-16 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-command.c (camel_imap_response_extract): Don't free 'resp' as it doesn't point to the beginning of the allocated data, instead free response->untagged->pdata[i]. Also, if '*resp' is equal to a space character, then set resp = imap_next_word (resp) rather than expecting resp++ to work (there's a list broken IMAP daemons that like to put extra spaces between tokens). (imap_read_response): Don't expect 'respbuf+2' to be where the untagged number response to start (see above fix for an explanation). 2000-10-16 Chris Toshok <toshok@helixcode.com> * camel-service.c (get_path): when using the construct (flags & CAMEL_SERVICE_URL_NEED_*) make sure to do ((flags & CAMEL_SERVICE_URL_NEED_*) == CAMEL_SERVICE_URL_NEED_*) (check_url): same. 2000-10-16 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-store.c (imap_disconnect): Only send the LOGOUT command if the store is connected. (imap_connect): Set the 'connected' state to TRUE when we successfully connect. (get_folder_info): if (!topfi), 'topfi' was allocated but then 'fi' was set. I think Dan meant to set topfi since fi is an uninitialized value at this point. * providers/imap/camel-imap-command.c (imap_read_response): Check for the untagged BYE response and set the 'connected' state to FALSE if we receive the BYE response. Return NULL if we get a BYE response. 2000-10-16 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-store.c (get_folder_info): Deal with the possibility of not getting a LIST response back for the top level. 2000-10-12 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-summary.c: Simple subclass of CamelFolderSummary that also keeps a UIDVALIDITY value (and doesn't, for the moment, build content info). * providers/imap/camel-imap-folder.c: (various): Use a CamelImapSummary to store/fetch summary info. (camel_imap_folder_new): Take a path to a file to use for the summary. Set the folder's permanent_flags correctly according to the server response. Read in the summary (checking the UIDVALIDITY) and update it if it's out of date. (imap_refresh_info): Just fetch UIDs and flags. If the UIDs all match, update the flags as needed and be done with it. Otherwise, delete messages that have been expunged from the server and fetch full summary info for any new messages. (imap_sync): Save the summary to disk. (imap_update_summary): Renamed from imap_get_summary_internal. Can now be told to get summary for only a subset of messages. Use camel-mime-utils functions rather than rolling our own header parsing. (imap_get_message_info_internal): Merged into imap_update_summary. (imap_set_message_flags): Don't marked the message FOLDER_FLAGGED if we're not actually changing the value of any of the flags. (camel_imap_folder_changed): Deal with EXISTS rather than RECENT. * providers/imap/camel-imap-store.c (imap_connect): Call camel_session_get_storage_path and save the value. (get_folder): Create a local directory to store summary information and pass a summary file name to camel_imap_folder_new. Don't call camel_folder_refresh_info from here any more since camel_imap_folder_new does it again. * providers/imap/camel-imap-command.c (camel_imap_command): Add a special case to this to make it possible to get the repsonses from a SELECT and still have store->current_folder be updated correctly. (imap_read_response): parse EXISTS rather than RECENT * camel-session.c (camel_session_get_storage_path): Use e_mkdir_hier. * camel-folder-summary.c (camel_folder_summary_remove_index): New function. * camel-mime-utils.c (header_raw_append_parse): fix this. (camel-mime-parser.c doesn't use this code because of the MEMPOOL optimization, so nothing was ever actually calling it before.) 2000-10-11 Not Zed <NotZed@HelixCode.com> * camel-mime-part.h (struct _CamelMimePart): Removed temp_message_buffer, and content_input_stream fields which seem to have come from nowhere, and are unused. * camel-mime-utils.c: Added a note about touching this file. Nobody is to touch it without asking me first. That goes for you too Jeff. (header_decode_text): In what way is this broken? 2000-10-10 Not Zed <NotZed@HelixCode.com> * providers/imap/camel-imap-folder.c (imap_get_summary_internal): Fix camel_summary_* function rename (imap_get_message_info_internal): Likewise. * camel-mime-parser.c (camel_mime_parser_finalise): Fixed a spelling mistake. * camel-folder-summary.c (camel_summary_format_address): Uh, why do we encode and then decode here ... sigh. This is not the way to fix this. (camel_folder_summary_format_address): Renamed to a proper name, this was only supposed to be a private function. (camel_folder_summary_format_string): Likewise. Oh i see why it was made public, code reuse by cut & paste. Joy. 2000-10-11 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-store.c (nntp_store_unsubscribe_folder): remove the leading '/'. (nntp_store_subscribe_folder): same. (nntp_store_folder_subscribed): same. 2000-10-11 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-folder.c (imap_move_message_to): Quote the mailbox name as it may contain spaces. (imap_copy_message_to): Same. 2000-10-10 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-store.c (build_folder_info_from_grouplist): fill in message_count and unread_message_count properly. * providers/nntp/camel-nntp-newsrc.h: reformat. * providers/nntp/camel-nntp-grouplist.c (camel_nntp_get_grouplist_from_file): remove spew. * providers/nntp/camel-nntp-newsrc.c (camel_nntp_newsrc_group_get_num_articles_read): new function. (camel_nntp_newsrc_group_get_highest_article_read): robustification. 2000-10-10 Joe Shaw <joe@helixcode.com> * providers/imap/camel-imap-store.c (parse_list_response_as_folder_info): Check before dereferencing the sep pointer. 2000-10-10 Jacob "Ulysses" Berkman <jacob@helixcode.com> * camel-*.c: teach camel about "its" vs. "it's" 2000-10-09 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-store.c (finalize): write out the newsrc. (nntp_store_get_name): if @brief, just return host. * providers/nntp/camel-nntp-newsrc.c: robustification and bug fixes. 2000-10-06 Jeffrey Stedfast <fejj@helixcode.com> * camel-folder-summary.c (camel_summary_format_address): Decode the resulting string. 2000-10-06 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-grouplist.c: new file. * providers/nntp/camel-nntp-grouplist.h: new file. * providers/nntp/camel-nntp-types.h: new file. 2000-10-06 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-utils.c (quoted_encode): Fix so that we don't encode every single char in the word. Also, do we need a safemask? I don't see why we would. (header_encode_string): Don't strip off the last char!! 2000-10-06 Chris Toshok <toshok@helixcode.com> * providers/nntp/Makefile.am (libcamelnntpinclude_HEADERS): add camel-nntp-grouplist.h (libcamelnntp_la_SOURCES): add camel-nntp-grouplist.c * providers/nntp/camel-nntp-provider.c: add our own hash functions for nntp urls. * providers/nntp/camel-nntp-newsrc.c (camel_nntp_newsrc_group_is_subscribed): new function. (camel_nntp_newsrc_subscribe_group): new function. (camel_nntp_newsrc_unsubscribe_group): new function. * providers/nntp/camel-nntp-newsrc.h: add prototypes for _group_is_subscribed, _subscribe_group, and _unsubscribe_group. * providers/nntp/camel-nntp-store.c (build_folder_info_from_grouplist): new function. (nntp_store_get_folder_info): add subscribed_only_parameter. if it's FALSE, load the grouplist and call build_folder_info_from_grouplist. (nntp_store_folder_subscribed): implement. (nntp_store_subscribe_folder): implement. (nntp_store_unsubscribe_folder): implement. (camel_nntp_store_init): add CAMEL_STORE_SUBSCRIPTIONS to the store's flags. * providers/mh/camel-mh-store.c (get_folder_info): add subscribed_only parameter. * providers/mbox/camel-mbox-store.c (get_folder_info): add subscribed_only parameter. * providers/imap/camel-imap-store.c (get_folder_info): add subscribed_only parameter. * camel-store.c (camel_store_supports_subscriptions): new function. (camel_store_folder_subscribed): new function. (camel_store_subscribe_folder): new function. (camel_store_unsubscribe_folder): new function. * camel-store.h: add prototypes and virtual functions for the subscribe implementation. also, add a subscribed_only argument to camel_store_get_folder_info. 2000-10-05 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-utils.c (header_address_list_format_append): Encode the name part of the address and don't quote the name. (header_decode_text): Rewrote from scratch, the old code was badly broken. 2000-10-05 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-message.c (camel_mime_message_set_reply_to): Use the camel_address_encode function again. (camel_mime_message_set_from): Same. 2000-10-05 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-utils.c (quoted_encode_step): Modified to not encode space chars in the middle of a line. (isblank): New macro if we're not on a system with the GNU isblank extension. * camel-mime-message.c (camel_mime_message_set_from): Reversed my changes, don't header_encode_phrase - it generates broken headers. (camel_mime_message_set_reply_to): Same. 2000-10-04 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-utils.c (camel_nntp_get_headers): revert to old method (only use XOVER if OVER is supported.) * providers/nntp/camel-nntp-store.c (camel_nntp_store_get_overview_fmt): handle the case where the OVER extension isn't listed but LIST OVERVIEW.FMT works (again, INN 2.2). enable the OVER extension in this case. 2000-10-04 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-utils.c (get_XOVER_headers): return a gboolean so we can tell if this command worked. we can't key off the OVER extension being present because at least one server (INN 2.2) doesn't report the OVER extension but implements the XOVER command. This could of course just be because I'm a loser for thinking they were related in the first place. (camel_nntp_get_headers): always try XOVER first, and if it fails revert to the slow method. 2000-10-04 Dan Winship <danw@helixcode.com> * providers/mbox/camel-mbox-store.c (get_folder): Fix a case where a variable was free'd and then possibly used in an error message. 2000-10-04 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-provider.c (camel_provider_module_init): news: -> nntp:. 2000-10-04 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-store.c (nntp_store_get_folder_info): use "nntp:" instead of "news:" since "news:" urls aren't supposed to have host/user/port info in them. also, if there's a user defined in the url, put it in the urls for our folders. 2000-10-04 Chris Toshok <toshok@helixcode.com> * providers/nntp/camel-nntp-auth.c (camel_nntp_auth_authenticate): borrow some code from the imap provider to query the user for their password, and pass the user/passwd to nntp. be extra paranoid and zero out the password before freeing it. * providers/nntp/camel-nntp-store.c (camel_nntp_store_init): add ALLOW_USER/ALLOW_PASSWORD/ALLOW_AUTH to the url flags. (nntp_store_query_auth_types_generic): return our list of auth_types. (nntp_store_query_auth_types_connected): broken, return same as in query_auth_types_generic. 2000-10-04 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-store.c (imap_connect): IMAP4 (pre-rev1) doesn't support the 'LIST "" ""' idiom, so don't use it. Just assume the dir_sep is '/'. Shrug. 2000-10-04 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-message.c (camel_mime_message_set_reply_to): Use header_encode_phrase instead. (camel_mime_message_set_from): Same. 2000-10-04 Not Zed <NotZed@HelixCode.com> * camel-mime-utils.c (header_content_type_is): Handle the case where ct != NULL, but type and subtype are, and also match that against text/plain. * camel-folder-summary.c: Bump summary file version. (message_info_save): Save the size from the messageinfo. (message_info_load): Load the size from the summary file. (message_info_load): Fixed up the time_t saving/loading. There was a reason the warning was left there ... obviously nobody could read the comment "/* warnings, leave them here */", why do i even bother. (camel_folder_summary_decode_time_t): Decode a time_t value from the summary file. (camel_folder_summary_encode_time_t): Encode a time_t value to the summary file. 2000-10-03 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-command.c (camel_imap_command): Quote the mailbox name when sending a SELECT request otherwise mailboxes with spaces in their names will cause problems. * camel-mime-message.c (camel_mime_message_set_reply_to): encode before setting. (camel_mime_message_set_from): Same. 2000-10-03 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-command.c: New file containing camel_imap_command and friends. Major camel_imap_command rewrite to remove duplicated code, make the parsing of literals be more safe/correct, deal with RECENT/EXPUNGE responses more consistently, and make it possible to implement the AUTHENTICATE command. * providers/imap/camel-imap-utils.c (imap_parse_nstring): New function, to parse an IMAP "nstring". * providers/imap/camel-imap-store.c: Move command stuff to camel-imap-command.c. Update for camel_imap_command changes. * providers/imap/camel-imap-folder.c: Update for camel_imap_command changes. (imap_append_message): CRLF filter the message before sending it. * providers/imap/Makefile.am: Add camel-imap-command.[ch], remove camel-imap-stream.[ch] for now. 2000-10-02 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-message.c (camel_mime_message_has_8bit_parts): New convenience function to determine if there are any 8bit mime parts in a mime message. (camel_mime_message_encode_8bit_parts): New convenience function to recursively reencode all 8bit mime parts to either quoted-printable or base64 depending on which would be the best encoding for that part. * providers/smtp/camel-smtp-transport.c (smtp_data): If the mime message contains 8bit parts and the server doesn't support 8bit transfers, reencode those parts before proceding with the send. (smtp_mail): If the mime message contains 8bit parts and the server supports the 8BITMIME extension to SMTP, notify the server that we'll be sending it 8bit mime parts. (_send_to): Find out if the message contains 8bit parts. 2000-10-02 Jeffrey Stedfast <fejj@helixcode.com> * providers/smtp/camel-smtp-transport.c (_send_to): Use the CamelInternetAddress parser. (smtp_get_email_addr_from_text): deprecated. 2000-10-02 Dan Winship <danw@helixcode.com> * camel-provider.h: Remove default_ports. * camel-remote-store.c (remote_connect): Get default_port from CamelRemoteStore rather than CamelProvider. * providers/{imap,nntp,pop3}/camel-*-store.c: Initialize CamelRemoteStore::default_port * providers/*/camel-*-provider.c: Remove default_ports. 2000-10-02 Dan Winship <danw@helixcode.com> * camel-folder.[ch]: Remove camel_folder_{get,free}_subfolder_info, as we want to be able to scan the whole subfolder tree without having to open any folders, so this needs to be in CamelStore. Remove can_hold_folders and can_hold_messages flags; things that don't hold messages are no longer considered CamelFolders. * camel-folder-summary.[ch]: Remove CamelFolderInfo stuff. * camel-store.[ch]: Add camel_store_{get,free}_folder_info, as well as camel_store_free_folder_info_full and ..._nop for default implementations, and camel_folder_info_free and camel_folder_info_build as convenience functions. Turn CamelFolderInfo into a tree structure and also add an "url" member. * providers/*/camel-*-folder.c: Remove subfolder_info and can_hold stuff. * providers/*/camel-*-store.c: Add folder_info stuff. * providers/imap/camel-imap-folder.c (imap_summary_free): Free the summary elements with camel_message_info_free, not camel_folder_info_free. Oops. * providers/imap/camel-imap-utils.c: const poison 2000-09-28 Jeffrey Stedfast <fejj@helixcode.com> * providers/smtp/camel-smtp-transport.c: Fixed some memory leaks. (camel_smtp_transport_init): Initialize supports_8bit to FALSE. (smtp_helo): If server supports 8bit, set supports_8bit to TRUE. * camel-transport.h (struct _CamelTransport): Added variable gboolean supports_8bit (we'll need this eventually? - see bugzilla bug #53) * providers/smtp/camel-smtp-transport.c (smtp_get_email_addr_from_text): Ugh, no wonder people were getting illegal seek warnings *sigh*. I guess I can only blame myself for this one though :-( 2000-09-28 Jeffrey Stedfast <fejj@helixcode.com> * providers/smtp/camel-smtp-transport.c (_send_to): Don't send the recipient data through smtp_get_email_addr_from_text - this is a complete waste. In fact, we don't want to have to use that function ever. * camel-internet-address.c, camel-address.c: Added some gtk-doc comments. 2000-09-28 Jeffrey Stedfast <fejj@helixcode.com> * camel-mime-utils.c (header_encode_string): Make sure to add the space char after an encoded word when the encoding is iso-8859-1. 2000-09-28 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-folder.c (imap_get_message): When getting a literal string response, don't include the \r\n after the closing } (as in: "... {798}\r\n...") * providers/imap/camel-imap-stream.c (stream_read): Same. 2000-09-28 Not Zed <NotZed@HelixCode.com> * camel-mime-utils.c (header_fold): New function to fold headers. 2000-09-27 Not Zed <NotZed@HelixCode.com> * camel-mime-parser.c (folder_scan_header): If we had an empty header, then it must be end of the headers too. (folder_scan_init): No we dont need to init the outbuf with a nul terminator. * camel-folder-summary.c (camel_folder_summary_set_uid): New function to reset the uid to a higher value. * providers/mbox/camel-mbox-summary.c (camel_mbox_summary_sync): "something failed (yo!)" what sort of crap is this? Fixed all the indenting again, what wanker keeps running stuff through indent? (message_info_new): Check the uid we loaded off the disk, if it existed already, assign a new one. If it didn't then make sure the nextuid is higher. * camel-charset-map.c: New file, used to build a large unicode decoding mapping table, and use it to determine what is the lowest charset a given word can be encoded with. Uses tables from libunicode's source. * camel-internet-address.c (internet_encode): Use header_phrase_encode to properly encode the fullname, as required. refixed indenting. Who keeps doing that? (camel_internet_address_find_address): Changed fatal return/warnings into assertions. * camel-mime-utils.c (header_raw_append_parse): Check : explicitly (removed from is_fieldname() macro). (camel_mime_special_table): Changed to short, so we can represent more bit types. (quoted_encode): Take a mask of the safe chars for this encoding. (header_address_decode): Removed a #warning that makes no sense anymore. (header_decode_date): Fixed the 'broken date' parser code, if it ever decoded it it just threw away the result. (header_encode_string): Use better charset matching for encoding strings as well. 2000-08-31 Not Zed <NotZed@HelixCode.com> * providers/mh/camel-mh-summary.c (camel_mh_summary_sync): Save the index if we do a sync. (camel_mh_summary_check): Save the index here too. Probably. 2000-09-27 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-store.c (camel_imap_command_extended): Return untagged data in a GPtrArray rather than a string, since it saves processing time and is much easier to deal with for several commands. Update for camel_imap_folder_changed change. (camel_imap_fetch_command): Update for camel_imap_folder_changed change. (imap_connect, imap_folder_exists): Update for camel_imap_command_extended change. * providers/imap/camel-imap-folder.c (imap_get_message_count_internal, imap_get_subfolder_info_internal, imap_search_by_expression): Update for camel_imap_command_extended change. (imap_get_summary_internal, imap_get_message_info_internal): Use camel_imap_fetch_command here now to get around the camel_imap_command_extended change. (camel_imap_folder_changed): turn expunged into a GArray of ints rather than a GPtrArray of strings representing ints. 2000-09-26 Jeffrey Stedfast <fejj@helixcode.com> * providers/imap/camel-imap-folder.c (imap_get_message_count_internal): Quote the folder name as it may contain spaces. (imap_get_subfolder_info_internal): Same. * providers/imap/camel-imap-utils.c (imap_parse_list_response): Do proper unquoting for folder names. (func_get_current_date): Implemented. * providers/imap/camel-imap-store.c (imap_folder_exists): Quote the folder name as it may have spaces. (imap_create): Same. (check_current_folder): Same. 2000-09-22 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-folder.c (imap_get_subfolder_info_internal): The root folder's name is "", not the namespace. (camel_imap_folder_new): constify folder_name. * providers/imap/camel-imap-store.c (get_folder): Create the folder with folder_name, not folder_path. (camel_imap_command_preliminary): Don't free cmdid here. 2000-09-21 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-utils.c (imap_create_flag_list): New function to convert Camel flags to an IMAP flag_list. (imap_parse_flag_list): Contrariwise. * providers/imap/camel-imap-store.c (camel_imap_command_*): Make the @ret arg actually optional, as (mostly) documented. (various): Don't pass "&result" to camel_imap_command_* if we're just going to immediately free it. Don't record status if we're not going to look at it. * providers/imap/camel-imap-folder.c: Likewise. (imap_summary_free): Use camel_folder_info_free. (imap_sync): Use imap_create_flag_list. Clear CAMEL_MESSAGE_FOLDER_FLAGGED after syncing so we don't keep re-syncing. (imap_append_message): Use imap_create_flag_list. Don't leak the memstream if the append fails. (imap_move_message_to): Use camel_folder_delete_message rather than doing it by hand. (imap_get_summary_internal, imap_get_message_info_internal): Use imap_parse_flag_list and header_raw_clear. (camel_imap_folder_changed): Use camel_message_info_free. 2000-09-21 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-store.c (get_folder_name): INBOX is case-insensitive. (get_root_folder_name): Make the root folder "" rather than "/". (get_folder): Update for root folder name change. (camel_imap_store_get_toplevel_dir): Removed. (Unused, unneeded.) (camel_imap_store_folder_path): New function to turn a Camel folder name into the corresponding namespaced IMAP path. (imap_folder_exists): Make this take a store and a path rather than a folder. (imap_create): Likewise (get_folder): Update for camel_imap_store_folder_path and other changes. (check_current_folder): Likewise. * providers/imap/camel-imap-folder.c: Change a bunch of CamelStore variables to CamelImapStore (and add a few more) to prevent excess gratuitous casting. Use camel_imap_store_folder_path where appropriate. (camel_imap_folder_new): Update for root folder name change. 2000-09-19 Jeffrey Stedfast <fejj@helixcode.com> * providers/smtp/camel-smtp-transport.c (smtp_data): Use the linewrap filter to achieve full RFC0821 compliance. * camel-mime-filter-linewrap.[c,h]: New mime-filter to word-wrap. 2000-09-19 Jeffrey Stedfast <fejj@helixcode.com> * camel-internet-address.c (internet_encode): When encoding the internet address, quote the name as the name may have commas or any other token which may later confuse our address parser. 2000-09-19 Dan Winship <danw@helixcode.com> * providers/imap/camel-imap-folder.c (imap_get_subfolder_info_internal): Fix the case where INBOX isn't returned in the folder listing. 2000-09-19 Dan Winship <danw@helixcode.com> * camel-folder.


context:
space:
mode: