/* -*- Mode: C; tab-width: 8; indent-tabs-mode: t; c-basic-offset: 8 -*- * * Copyright (C) 2000 Helix Code, Inc. * * Authors: Michael Zucchi * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU Library General Public License * as published by the Free Software Foundation; either version 2 of * the License, or (at your option) any later version. * * This library is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * Library General Public License for more details. * * You should have received a copy of the GNU Library General Public * License along with the Gnome Library; see the file COPYING.LIB. If not, * write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, * Boston, MA 02111-1307, USA. */ /* code to manage a word index */ /* includes a cache for word index writes, but not for name index writes (currently), or any reads. Note the word cache is only needed during indexing of lots of words, and could then be discarded (:flush()). */ #include #include #include #include #include #include #include #include #include "block.h" #include "index.h" #include "wordindex.h" #define d(x) /*#define WORDCACHE_SIZE (256)*/ #define WORDCACHE_SIZE (4096) extern struct _IBEXStoreClass ibex_diskarray_class; extern struct _IBEXIndexClass ibex_hash_class; /* need 2 types of hash key? one that just stores the wordid / wordblock and one that stores the filecount/files? */ #define CACHE_FILE_COUNT (62) struct _wordcache { struct _wordcache *next; struct _wordcache *prev; nameid_t wordid; /* disk wordid */ blockid_t wordblock; /* head of disk list */ blockid_t wordtail; /* and the tail data */ short filecount; /* how many valid items in files[] */ short filealloc; /* how much allocated space in files[] */ union { nameid_t *files; /* memory cache of files */ nameid_t file0; /* if filecount == 1 && filealloc == 0, store directly */ } file; char word[1]; /* actual word follows */ }; static void unindex_name(struct _IBEXWord *, const char *name); /* unindex all entries for name */ static gboolean contains_name(struct _IBEXWord *, const char *name); /* index contains data for name */ static GPtrArray *find(struct _IBEXWord *, const char *word); /* returns all matches for word */ static gboolean find_name(struct _IBEXWord *, const char *name, const char *word); /* find if name contains word */ static void add(struct _IBEXWord *, const char *name, const char *word); /* adds a single word to name (slow) */ static void add_list(struct _IBEXWord *, const char *name, GPtrArray *words);/* adds a bunch of words to a given name */ static int word_sync(struct _IBEXWord *idx); static int word_flush(struct _IBEXWord *idx); static int word_close(struct _IBEXWord *idx); static void word_index_pre(struct _IBEXWord *idx); static void word_index_post(struct _IBEXWord *idx); struct _IBEXWordClass ibex_word_index_class = { word_sync, word_flush, word_close, word_index_pre, word_index_post, unindex_name, contains_name, find, find_name, add, add_list }; /* this interface isn't the best, but it'll do for now */ struct _IBEXWord * ibex_create_word_index(struct _memcache *bc, blockid_t *wordroot, blockid_t *nameroot) { struct _IBEXWord *idx; idx = g_malloc0(sizeof(*idx)); idx->wordcache = g_hash_table_new(g_str_hash, g_str_equal); ibex_list_new(&idx->wordnodes); idx->wordcount = 0; idx->klass = &ibex_word_index_class; /* we use the same block array storage for both indexes at the moment */ idx->wordstore = ibex_diskarray_class.create(bc); idx->namestore = idx->wordstore; /* but not the same indexes! */ if (*wordroot) { printf("opening wordindex root = %d\n", *wordroot); idx->wordindex = ibex_hash_class.open(bc, *wordroot); } else { idx->wordindex = ibex_hash_class.create(bc, 2048); *wordroot = idx->wordindex->root; printf("creating wordindex root = %d\n", *wordroot); } if (*nameroot) { printf("opening nameindex root = %d\n", *nameroot); idx->nameindex = ibex_hash_class.open(bc, *nameroot); } else { idx->nameindex = ibex_hash_class.create(bc, 2048); *nameroot = idx->nameindex->root; printf("creating nameindex root = %d\n", *nameroot); } return idx; } #if d(!)0 static void cache_sanity(struct _wordcache *head) { while (head->next) { g_assert(head->filecount <= head->filealloc); g_assert(strlen(head->word) != 0); head = head->next; } } #endif static void word_index_pre(struct _IBEXWord *idx) { } static void word_index_post(struct _IBEXWord *idx) { } /* unindex all entries for name */ static void unindex_name(struct _IBEXWord *idx, const char *name) { GArray *words; int i; nameid_t nameid, wordid; blockid_t nameblock, wordblock, newblock, nametail, wordtail, newtail; char *word; struct _wordcache *cache; d(printf("unindexing %s\n", name)); /* lookup the hash key */ nameid = idx->nameindex->klass->find(idx->nameindex, name, strlen(name)); /* get the block for this key */ nameblock = idx->nameindex->klass->get_data(idx->nameindex, nameid, &nametail); /* and the data for this block */ words = idx->namestore->klass->get(idx->namestore, nameblock, nametail); /* walk it ... */ for (i=0;ilen;i++) { /* get the word */ wordid = g_array_index(words, nameid_t, i); d(printf(" word %d\n", wordid)); /* get the data block */ wordblock = idx->wordindex->klass->get_data(idx->wordindex, wordid, &wordtail); /* clear this name from it */ newblock = wordblock; newtail = wordtail; idx->wordstore->klass->remove(idx->wordstore, &newblock, &newtail, nameid); if (newblock != wordblock || newtail != wordtail) idx->wordindex->klass->set_data(idx->wordindex, wordid, newblock, newtail); /* now check the cache as well */ word = idx->nameindex->klass->get_key(idx->wordindex, wordid, NULL); if (word) { cache = g_hash_table_lookup(idx->wordcache, word); if (cache) { /* its there, update our head/tail pointers */ cache->wordblock = newblock; cache->wordtail = newtail; /* now check that we have a data entry in it */ if (cache->filealloc == 0 && cache->filecount == 1) { if (cache->file.file0 == nameid) { cache->filecount = 0; } } else { int j; for (j=0;jfilecount;j++) { if (cache->file.files[j] == nameid) { cache->file.files[j] = cache->file.files[cache->filecount-1]; cache->filecount--; break; } } } } g_free(word); } } g_array_free(words, TRUE); /* and remove name data and itself */ idx->namestore->klass->free(idx->namestore, nameblock, nametail); idx->nameindex->klass->remove(idx->nameindex, name, strlen(name)); } /* index contains (any) data for name */ static gboolean contains_name(struct _IBEXWord *idx, const char *name) { return idx->nameindex->klass->find(idx->nameindex, name, strlen(name)) != 0; } /* returns all matches for word */ static GPtrArray *find(struct _IBEXWord *idx, const char *word) { nameid_t wordid, nameid; GPtrArray *res; GArray *names; int i; char *new; struct _wordcache *cache; blockid_t wordblock, wordtail; res = g_ptr_array_new(); cache = g_hash_table_lookup(idx->wordcache, word); if (cache) { /* freshen cache entry if we touch it */ ibex_list_remove((struct _listnode *)cache); ibex_list_addtail(&idx->wordnodes, (struct _listnode *)cache); wordid = cache->wordid; wordblock = cache->wordblock; wordtail = cache->wordtail; } else { /* lookup the hash key */ wordid = idx->wordindex->klass->find(idx->wordindex, word, strlen(word)); /* get the block for this key */ wordblock = idx->wordindex->klass->get_data(idx->wordindex, wordid, &wordtail); } /* and the data for this block */ names = idx->wordstore->klass->get(idx->wordstore, wordblock, wordtail); /* .. including any memory-only data */ if (cache) { if (cache->filealloc == 0 && cache->filecount == 1) g_array_append_val(names, cache->file.file0); else g_array_append_vals(names, cache->file.files, cache->filecount); } /* walk it ... converting id's back to strings */ g_ptr_array_set_size(res, names->len); for (i=0;ilen;i++) { nameid = g_array_index(names, nameid_t, i); new = idx->nameindex->klass->get_key(idx->nameindex, nameid, NULL); res->pdata[i] = new; } g_array_free(names, TRUE); return res; } /* find if name contains word */ static gboolean find_name(struct _IBEXWord *idx, const char *name, const char *word) { nameid_t wordid, nameid; blockid_t nameblock, nametail; struct _wordcache *cache; int i; /* lookup the hash key for the name */ nameid = idx->nameindex->klass->find(idx->nameindex, name, strlen(name)); /* get the block for this name */ nameblock = idx->nameindex->klass->get_data(idx->nameindex, nameid, &nametail); /* check if there is an in-memory cache for this word, check its file there first */ cache = g_hash_table_lookup(idx->wordcache, word); if (cache) { /* freshen cache entry if we touch it */ ibex_list_remove((struct _listnode *)cache); ibex_list_addtail(&idx->wordnodes, (struct _listnode *)cache); if (cache->filecount == 1 && cache->filealloc == 0) { if (cache->file.file0 == nameid) return TRUE; } else { for (i=0;ifilecount;i++) { if (cache->file.files[i] == nameid) return TRUE; } } /* not there? well we can use the wordid anyway */ wordid = cache->wordid; } else { /* lookup the hash key for word */ wordid = idx->wordindex->klass->find(idx->wordindex, word, strlen(word)); } /* see if wordid is in nameblock */ return idx->namestore->klass->find(idx->namestore, nameblock, nametail, wordid); } /* cache helper functions */ /* flush a cache entry to disk, and empty it out */ static void sync_cache_entry(struct _IBEXWord *idx, struct _wordcache *cache) { GArray array; /* just use this as a header */ blockid_t oldblock, oldtail; d(printf("syncing cache entry '%s' used %d\n", cache->word, cache->filecount)); if (cache->filecount == 1 && cache->filealloc == 0) array.data = (char *)&cache->file.file0; else array.data = (char *)cache->file.files; array.len = cache->filecount; oldblock = cache->wordblock; oldtail = cache->wordtail; idx->wordstore->klass->add_list(idx->wordstore, &cache->wordblock, &cache->wordtail, &array); if (oldblock != cache->wordblock || oldtail != cache->wordtail) { idx->wordindex->klass->set_data(idx->wordindex, cache->wordid, cache->wordblock, cache->wordtail); } cache->filecount = 0; } /* create a new key in an index, returning its id and head block */ static void add_index_key(struct _IBEXIndex *wordindex, const char *word, nameid_t *wordid, blockid_t *wordblock, blockid_t *wordtail) { /* initialise cache entry - id of word entry and head block */ *wordid = wordindex->klass->find(wordindex, word, strlen(word)); if (*wordid == 0) { *wordid = wordindex->klass->insert(wordindex, word, strlen(word)); *wordblock = 0; *wordtail = 0; } else { *wordblock = wordindex->klass->get_data(wordindex, *wordid, wordtail); } } /* create a new key in a cached index (only word cache so far), flushing old keys if too much space is being used */ static struct _wordcache * add_index_cache(struct _IBEXWord *idx, const char *word) { struct _wordcache *cache; d(printf("adding %s to cache\n", word)); cache = g_hash_table_lookup(idx->wordcache, word); if (cache == 0) { /* see if we have to flush off the last entry */ if (idx->wordcount >= WORDCACHE_SIZE) { struct _wordcache *mincache; int min, count=0; /* remove last entry, and flush it */ cache = (struct _wordcache *)idx->wordnodes.tailpred; mincache = cache; min = mincache->filecount; d(printf("flushing word from cache %s\n", cache->word)); /* instead of just using the last entry, we try and find an entry with with only 1 item (failing that, the smallest in the range we look at) */ /* this could probably benefit greatly from a more sophisticated aging algorithm */ while (cache->next && count < 100) { if (cache->filecount == 1) { mincache = cache; break; } if (cache->filecount > 0 && cache->filecount < min) { mincache = cache; min = cache->filecount; } cache = cache->next; count++; } ibex_list_remove((struct _listnode *)mincache); g_hash_table_remove(idx->wordcache, mincache->word); sync_cache_entry(idx, mincache); if (mincache->filealloc) g_free(mincache->file.files); g_free(mincache); idx->wordcount--; } cache = g_malloc0(sizeof(*cache)+strlen(word)); /* initialise cache entry - id of word entry and head block */ add_index_key(idx->wordindex, word, &cache->wordid, &cache->wordblock, &cache->wordtail); /* other fields */ strcpy(cache->word, word); cache->filecount = 0; g_hash_table_insert(idx->wordcache, cache->word, cache); ibex_list_addhead(&idx->wordnodes, (struct _listnode *)cache); idx->wordcount++; } else { /* move cache bucket ot the head of the cache list */ ibex_list_remove((struct _listnode *)cache); ibex_list_addhead(&idx->wordnodes, (struct _listnode *)cache); } return cache; } /* adds a single word to name (slow) */ static void add(struct _IBEXWord *idx, const char *name, const char *word) { nameid_t nameid; blockid_t nameblock, newblock, nametail, newtail; struct _wordcache *cache; g_error("Dont use wordindex::add()"); abort(); cache = add_index_cache(idx, word); /* get the nameid and block start for this name */ add_index_key(idx->nameindex, name, &nameid, &nameblock, &nametail); /* check for repeats of the last name - dont add anything */ if (cache->filecount == 1 && cache->filealloc == 0) { if (cache->file.file0 == nameid) return; } else { if (cache->file.files[cache->filecount] == nameid) return; } /* see if we are setting the first, drop it in the union */ if (cache->filecount == 0 && cache->filealloc == 0) { cache->file.file0 = nameid; } else if (cache->filecount == 1 && cache->filealloc == 0) { nameid_t saveid; /* we need to allocate space for words */ saveid = cache->file.file0; cache->file.files = g_malloc(sizeof(cache->file.files[0]) * CACHE_FILE_COUNT); /* this could possibly grow as needed, but i wont for now */ cache->filealloc = CACHE_FILE_COUNT; cache->file.files[0] = saveid; cache->file.files[1] = nameid; } else { cache->file.files[cache->filecount] = nameid; } cache->filecount++; /* if we are full, force a flush now */ if (cache->filealloc && cache->filecount >= cache->filealloc) { sync_cache_entry(idx, cache); } newtail = nametail; newblock = nameblock; idx->namestore->klass->add(idx->namestore, &newblock, &newtail, cache->wordid); if (newblock != nameblock || newtail != nametail) { idx->nameindex->klass->set_data(idx->nameindex, nameid, newblock, newtail); } } /* adds a bunch of words to a given name */ static void add_list(struct _IBEXWord *idx, const char *name, GPtrArray *words) { int i; GArray *data = g_array_new(0, 0, sizeof(nameid_t)); blockid_t nameblock, newblock, nametail, newtail; nameid_t nameid; struct _wordcache *cache; d(printf("Adding words to name %s\n", name)); d(cache_sanity((struct _wordcache *)idx->wordnodes.head)); /* get the nameid and block start for this name */ add_index_key(idx->nameindex, name, &nameid, &nameblock, &nametail); d(cache_sanity((struct _wordcache *)idx->wordnodes.head)); for (i=0;ilen;i++) { char *word = words->pdata[i]; cache = add_index_cache(idx, word); /*d(cache_sanity((struct _wordcache *)idx->wordnodes.head));*/ /* check for duplicates; doesn't catch duplicates over an overflow boundary. Watch me care. */ if (cache->filecount == 0 /* the 1 item case */ || (cache->filecount == 1 && cache->filealloc == 0 && cache->file.file0 != nameid) /* the normal case */ || (cache->filealloc > 0 && cache->file.files[cache->filecount-1] != nameid)) { /* see if we are setting the first, drop it in the union */ if (cache->filecount == 0 && cache->filealloc == 0) { cache->file.file0 = nameid; } else if (cache->filecount == 1 && cache->filealloc == 0) { nameid_t saveid; /* we need to allocate space for words */ saveid = cache->file.file0; cache->file.files = g_malloc(sizeof(cache->file.files[0]) * CACHE_FILE_COUNT); /* this could possibly grow as needed, but i wont for now */ cache->filealloc = CACHE_FILE_COUNT; cache->file.files[0] = saveid; cache->file.files[1] = nameid; } else { cache->file.files[cache->filecount] = nameid; } cache->filecount++; /* if we are full, force a flush now */ if (cache->filealloc && cache->filecount >= cache->filealloc) { sync_cache_entry(idx, cache); } /*d(cache_sanity((struct _wordcache *)idx->wordnodes.head));*/ /* and append this wordid for this name in memory */ g_array_append_val(data, cache->wordid); } /*d(cache_sanity((struct _wordcache *)idx->wordnodes.head));*/ } d(cache_sanity((struct _wordcache *)idx->wordnodes.head)); /* and append these word id's in one go */ newblock = nameblock; newtail = nametail; idx->namestore->klass->add_list(idx->namestore, &newblock, &newtail, data); if (newblock != nameblock || newtail != nametail) { idx->nameindex->klass->set_data(idx->nameindex, nameid, newblock, newtail); } d(cache_sanity((struct _wordcache *)idx->wordnodes.head)); g_array_free(data, TRUE); } /* sync any in-memory data to disk */ static int word_sync(struct _IBEXWord *idx) { /* we just flush also, save memory */ word_flush(idx); #if 0 struct _wordcache *cache = (struct _wordcache *)idx->wordnodes.head; while (cache->next) { sync_cache_entry(idx, cache); cache = cache->next; } /*ibex_hash_dump(idx->wordindex);*/ /*ibex_hash_dump(idx->nameindex);*/ #endif return 0; } /* sync and flush any in-memory data to disk and free it */ static int word_flush(struct _IBEXWord *idx) { struct _wordcache *cache = (struct _wordcache *)idx->wordnodes.head, *next; extern int block_log; int count= 0; int used=0, wasted=0; block_log = 0; next = cache->next; while (next) { count++; used += sizeof(struct _wordcache) + (cache->filealloc * sizeof(nameid_t)); if (cache->filealloc) wasted += (cache->filealloc-cache->filecount)*sizeof(nameid_t); else wasted += (1-cache->filecount) * sizeof(nameid_t); /*printf("syncing word %s\n", cache->word);*/ sync_cache_entry(idx, cache); g_hash_table_remove(idx->wordcache, cache->word); if (cache->filealloc) g_free(cache->file.files); g_free(cache); cache = next; next = cache->next; } printf("sync cache entries = %d\n used memory = %d\n wasted memory = %d\n", count, used, wasted); block_log = 0; ibex_list_new(&idx->wordnodes); idx->wordcount = 0; return 0; } static int word_close(struct _IBEXWord *idx) { struct _wordcache *cache = (struct _wordcache *)idx->wordnodes.head, *next; extern int block_log; int count= 0; int used=0, wasted=0; block_log = 0; next = cache->next; while (next) { count++; used += sizeof(struct _wordcache) + (cache->filealloc * sizeof(nameid_t)); if (cache->filealloc) wasted += (cache->filealloc-cache->filecount)*sizeof(nameid_t); else wasted += (1-cache->filecount) * sizeof(nameid_t); /*printf("closing word %s\n", cache->word);*/ sync_cache_entry(idx, cache); if (cache->filealloc) g_free(cache->file.files); g_free(cache); cache = next; next = cache->next; } block_log = 0; printf("cache entries = %d\n used memory = %d\n wasted memory = %d\n", count, used, wasted); idx->namestore->klass->close(idx->namestore); idx->nameindex->klass->close(idx->nameindex); /*same as namestore: idx->wordstore->klass->close(idx->wordstore);*/ idx->wordindex->klass->close(idx->wordindex); g_hash_table_destroy(idx->wordcache); g_free(idx); return 0; } href='/~lantw44/cgit/gsoc2013-evolution/commit/?h=GNUMERIC_0_60&id=6e486616edda3b7ef1ed273aa2aaecfbdea237e0'>Moved files to e-util - FedericoFederico Mena Quintero2000-09-2610-1746/+0 * CamelException is not for compile-time errors. Replace lots of argumentDan Winship2000-09-262-485/+95 * Update the shortcut bar in the shell view to match the changes in theEttore Perazzoli2000-09-267-158/+431 * Make the shortcut bar not alter the model by itself on a drag and dropEttore Perazzoli2000-09-263-6/+72 * add save_state stuff to ETreeModel, and prototypes for the public methods.Chris Toshok2000-09-262-7/+213 * *** empty log message ***Nat Friedman2000-09-261-0/+5 * Added a ton of new prefixes and suffixes.Nat Friedman2000-09-262-20/+60 * fix typo in tooltip. fix accelerator for "Save as" to be different fromDan Winship2000-09-263-2/+9 * Reverted somes changes in POTFILES.in and micro-updated French translation.Christophe Merlet2000-09-253-40/+33 * Updated Finnish translationjjranta2000-09-252-143/+149 * Updated French translation and POTFILES.in files listChristophe Merlet2000-09-253-2232/+2408 * Fix include typo.Ariel Rios2000-09-251-1/+1 * Fix a refcounting problem with the local storage. ("Somebody" added aEttore Perazzoli2000-09-252-2/+4 * fix sorting - need to update the row_array as well as the actual structureChris Toshok2000-09-251-4/+20 * Updated to handle FIFO streams.Jeffrey Stedfast2000-09-252-31/+23 * set the height of the scrolled window for the description field, since theDamon Chaplin2000-09-259-169/+488 * use pixmaps instead of GtkArrows to look better. Also set the canvasDamon Chaplin2000-09-258-247/+290 * Updates for the Bonobo changes from Michael who is having someEttore Perazzoli2000-09-2311-96/+55 * Fix a crash that happened on my system when sorting by the "From"Ettore Perazzoli2000-09-232-2/+16 * Ugh. Me hack icons. Me commit cvs. Me learns. duh.Tuomas Kuosmanen2000-09-231-0/+0 * oops, my goof with replacing a wrong icon :P Should work now. Ettore, can you...Tuomas Kuosmanen2000-09-231-0/+9 * Again...wrap the folder names in quotesJeffrey Stedfast2000-09-233-8/+9 * lots of changes. flesh out the remove/modify/create functions. add anotherChris Toshok2000-09-232-111/+472 * New comparison function for email addresses. (subject_compare): NewJeffrey Stedfast2000-09-231-6/+23 * add E_CARD_SIMPLE_FIELD_FAMILY_NAME to the enum.Chris Toshok2000-09-233-0/+15 * The root folder's name is "", not the namespace. (camel_imap_folder_new):Dan Winship2000-09-234-11/+18 * added yahoo, whowhere, and infospace servers.Chris Toshok2000-09-232-0/+28 * Fixed some warnings. Fixed a bug where quoted printable fields wereChristopher James Lahey2000-09-224-25/+40 * Added a function to set the status message associated with a given view.Christopher James Lahey2000-09-228-22/+118 * Fixed operation of the keypad keys in EText and ECellText.Christopher James Lahey2000-09-222-12/+46 * Added evolution-addressbook-ldap.xml.Christopher James Lahey2000-09-2223-142/+442 * New comparison function for email addresses. (subject_compare): NewJeffrey Stedfast2000-09-222-3/+169 * Removed the "about calendar" command, since we don't want to have bothFederico Mena Quintero2000-09-224-30/+8 * New function to convert Camel flags to an IMAP flag_list.Dan Winship2000-09-225-224/+161 * Dear native speakers,Federico Mena Quintero2000-09-225-4/+12 * Fixed some memory leakage. Call free_recipients() so we don't leak memory.Jeffrey Stedfast2000-09-223-10/+26 * INBOX is case-insensitive. (get_root_folder_name): Make the root folder ""Dan Winship2000-09-224-263/+132 * Switched to the "official" FSF markup. I will have to make changes to theAaron Weber2000-09-2212-1590/+1298 * s/Bonobo_UIHandler/Bonobo_UIContainer/Michael Meeks2000-09-2116-16/+56 * Updated Norwegian translation.Kjartan Maraas2000-09-212-408/+295 * Include errno.hMiguel de Icaza2000-09-211-0/+1 * Typo caused readability problems with ETable.Christopher James Lahey2000-09-211-1/+1 * Changed the #ifndef used so that it doesn't collide with one inside ofChristopher James Lahey2000-09-211-3/+3 * Oh my god! Chris Lahey killed Kenny! You bastard!Christopher James Lahey2000-09-215-38/+96 * Made e_list a bit more reentrant. If a iterator gets its data pulled outChristopher James Lahey2000-09-214-8/+31 * Don't fail if there is no map file.JP Rosevear2000-09-212-7/+20 * Make this work when a field is spread across multiple lines.Christopher James Lahey2000-09-212-20/+55 * Added a wants_html field to cards. Uses "x-mozilla-html".Christopher James Lahey2000-09-2113-502/+206 * Added a cvsignore file here.Christopher James Lahey2000-09-212-0/+11 * Added evolution-contact-editor.xml.Christopher James Lahey2000-09-213-0/+252 * Add since field to contextJP Rosevear2000-09-214-5/+43 * Make the log file name relevant to the actual calendar file, rather thanJP Rosevear2000-09-212-4/+10 * OopsJP Rosevear2000-09-212-1/+5 * Use a local sax handler.JP Rosevear2000-09-213-61/+15 * Properly assign the CalObjChange type. (cal_backend_log_sax_parse): DeleteJP Rosevear2000-09-212-57/+54 * Updated russian translation.Valek Frob2000-09-202-1087/+825 * Added 2 new icons for the compose -dialog.. /tigertTuomas Kuosmanen2000-09-203-17/+86 * Updated the Swedish translation.Christian Rose2000-09-201-2/+2 * Fixed the Swedish translation.Christian Rose2000-09-201-1/+1 * Fixed sig stuff here and in setupassist.Aaron Weber2000-09-2010-70/+92 * Fixed display of the minicards when the addressbook was first loading. (ItChristopher James Lahey2000-09-202-3/+8 * New utility function (cal_backend_load): use above (cal_backend_create):JP Rosevear2000-09-203-52/+89 * alter the krb4 check a bit to deal with configure cache suckage. (If youDan Winship2000-09-202-1/+9 * Fixed to install the stylesheet-images as well.Federico Mena Quintero2000-09-208-14/+44 * Moved files, fixed header files and paths - FedericoFederico Mena Quintero2000-09-2084-174/+180 * Add some other cases where a slow sync is in order (pre_sync): Pre loadJP Rosevear2000-09-2017-1996/+1170 * ack, remove the ~ and object filesJeffrey Stedfast2000-09-204-197/+0 * Use the linewrap filter to achieve full RFC0821 compliance.Jeffrey Stedfast2000-09-2010-129/+531 * Updated Norwegian translation.Kjartan Maraas2000-09-203-1147/+891 * Change "$(prefix)/share" to "$(datadir)" in a few places. Problem noticedDan Winship2000-09-206-4/+18 * Use the folder's full_name so recursive directory structures displayJeffrey Stedfast2000-09-202-2/+7 * When encoding the internet address, quote the name as the name may haveJeffrey Stedfast2000-09-202-31/+42 * Fix the case where INBOX isn't returned in the folder listing.Dan Winship2000-09-202-1/+12 * Update for CamelFolder changes (subfolder_names -> subfolder_info).Dan Winship2000-09-202-5/+13 * Removed (camel_folder_init, camel_folder_construct): New object initDan Winship2000-09-2021-646/+502 * Deal with NULL composer.Dan Winship2000-09-195-16/+36 * Don't g_error out if the html-editor-control fails. (create_composer): NewDan Winship2000-09-193-64/+70 * ** Merged from IBEX_DISK branch to head.Not Zed2000-09-1916-924/+3681 * add gal as a dependencyDan Winship2000-09-192-1/+7 * Use gnome_vfs_get_file_info.Dan Winship2000-09-193-12/+36 * Call gnome_vfs_init() since the composer now does file operations (to getDan Winship2000-09-192-0/+8 * retarded looking icons for use with displaying scores in the message-listJeffrey Stedfast2000-09-199-0/+192 * Removed COL_ONLINE_STATUS because we don't want that. Renamed COL_PRIORITYJeffrey Stedfast2000-09-193-123/+169 * Modified to use more intuitive namingJeffrey Stedfast2000-09-198-605/+760 * set this to 4 to specify how much X/Open we want with that.Dan Winship2000-09-192-1/+6 * Added check for gnome-app-lib. Removed directories that have been moved toChristopher James Lahey2000-09-18