index : gsoc2013-evolution	archive-integration archive-integration-review master
	Google Summer of Code 2013 - GNOME - Archive Integration workspace

aboutsummaryrefslogtreecommitdiffstats

log msg author committer range

path: root/po/en_GB.po

	Commit message (Expand)	Author	Age	Files	Lines
*	Sync for 0.99.2.	Ettore Perazzoli	2001-11-15	1	-339/+351
*	0.99.0.	Ettore Perazzoli	2001-11-06	1	-261/+262
*	Syncing up for the 0.16.100 internal release.	Ettore Perazzoli	2001-11-01	1	-2338/+2635
*	0.16.	Ettore Perazzoli	2001-10-10	1	-709/+784
*	0.15.	Ettore Perazzoli	2001-10-02	1	-929/+1087
*	Sync for 0.14.	Ettore Perazzoli	2001-09-22	1	-1837/+2320
*	Sync for 0.13.	Ettore Perazzoli	2001-08-23	1	-1992/+3537
*	changed all %m/%d/%Y date formats to %d/%m/%Y.	Damon Chaplin	2001-08-20	1	-7/+7
*	Sync for the release.	Ettore Perazzoli	2001-08-01	1	-3099/+5953
*	Do this everywhere, also fixup CLEANFILES	Michael Meeks	2001-04-14	1	-1857/+5372
*	Removed addressbook/gui/widgets/test-minicard-view.c,	Dave Camp	2001-01-18	1	-2957/+1065
*	Updated Norwegian (bokmål) translation.	Kjartan Maraas	2001-01-09	1	-1570/+1939
*	I like monkeys - Federico	Federico Mena Quintero	2000-12-14	1	-1577/+3367
*	Added British translation.	Robert Brady	2000-10-22	1	-0/+4659

generated by cgit (git 2.34.1) at 2024-11-18 22:19:54 +0800

#include <glib.h>

#include <stdio.h>
#include <gal/unicode/gunicode.h>
#include <ctype.h>
#include <string.h>
#include <errno.h>

#include "ibex_internal.h"

#define d(x)

static signed char utf8_trans[] = {
    'A', 'A', 'A', 'A', 'A', 'A', -1, 'C', 'E', 'E', 'E', 'E', 'I', 'I',
    'I', 'I', -2, 'N', 'O', 'O', 'O', 'O', 'O', '*', 'O', 'U', 'U', 'U',
    'U', 'Y', -3, -4, 'a', 'a', 'a', 'a', 'a', 'a', -5, 'c', 'e', 'e',
    'e', 'e', 'i', 'i', 'i', 'i', -6, 'n', 'o', 'o', 'o', 'o', 'o', '/',
    'o', 'u', 'u', 'u', 'u', 'y', -7, 'y', 'A', 'a', 'A', 'a', 'A', 'a',
    'C', 'c', 'C', 'c', 'C', 'c', 'C', 'c', 'D', 'd', 'D', 'd', 'E', 'e',
    'E', 'e', 'E', 'e', 'E', 'e', 'E', 'e', 'G', 'g', 'G', 'g', 'G', 'g',
    'G', 'g', 'H', 'h', 'H', 'h', 'I', 'i', 'I', 'i', 'I', 'i', 'I', 'i',
    'I', 'i', -8, -9, 'J', 'j', 'K', 'k', 'k', 'L', 'l', 'L', 'l', 'L',
    'l', 'L', 'l', 'L', 'l', 'N', 'n', 'N', 'n', 'N', 'n', 'n', -10, -11,
    'O', 'o', 'O', 'o', 'O', 'o', -12, -13, 'R', 'r', 'R', 'r', 'R', 'r',
    'S', 'r', 'S', 's', 'S', 's', 'S', 's', 'T', 't', 'T', 't', 'T', 't',
    'U', 'u', 'U', 'u', 'U', 'u', 'U', 'u', 'U', 'u', 'U', 'u', 'W', 'w',
    'Y', 'y', 'Y', 'Z', 'z', 'Z', 'z', 'Z', 'z', 's'
};

static char *utf8_long_trans[] = {
    "AE", "TH", "TH", "ss", "ae", "th", "th", "IJ", "ij",
    "NG", "ng", "OE", "oe"
};

/* This is a bit weird. It takes pointers to the start and end (actually
 * just past the end) of a UTF-8-encoded word, and a buffer at least 1
 * byte longer than the length of the word. It copies the word into the
 * buffer in all lowercase without accents, and splits up ligatures.
 * (Since any ligature would be a multi-byte character in UTF-8, splitting
 * them into two US-ASCII characters won't overrun the buffer.)
 *
 * It is not safe to call this routine with bad UTF-8.
 */
static void
ibex_normalise_word(char *start, char *end, char *buf)
{
    unsigned char *s, *d;
    gunichar uc;

    s = (unsigned char *)start;
    d = (unsigned char *)buf;
    while (s < (unsigned char *)end) {
        if (*s < 0x80) {
            /* US-ASCII character: copy unless it's
             * an apostrophe.
             */
            if (*s != '\'')
                *d++ = tolower (*s);
            s++;
        } else {
            char *next = g_utf8_next_char (s);
            uc = g_utf8_get_char (s);
            if (uc >= 0xc0 && uc < 0xc0 + sizeof (utf8_trans)) {
                signed char ch = utf8_trans[uc - 0xc0];
                if (ch > 0)
                    *d++ = tolower (ch);
                else {
                    *d++ = tolower (utf8_long_trans[-ch - 1][0]);
                    *d++ = tolower (utf8_long_trans[-ch - 1][1]);
                }
                s = next;
            } else {
                while (s < (unsigned char *)next)
                    *d++ = *s++;
            }
        }
    }
    *d = '\0';
}

enum { IBEX_ALPHA, IBEX_NONALPHA, IBEX_INVALID, IBEX_INCOMPLETE };

static int
utf8_category (char *p, char **np, char *end)
{
    if (isascii ((unsigned char)*p)) {
        *np = p + 1;
        if (isalpha ((unsigned char)*p) || *p == '\'')
            return IBEX_ALPHA;
        return IBEX_NONALPHA;
    } else {
        gunichar uc;

        *np = g_utf8_find_next_char (p, end);
        if (!*np)
            return IBEX_INCOMPLETE;
        uc = g_utf8_get_char (p);
        if (uc == (gunichar) -1)
            return IBEX_INVALID;
        else if (g_unichar_isalpha (uc))
            return IBEX_ALPHA;
        else
            return IBEX_NONALPHA;
    }
}

/**
 * ibex_index_buffer: the lowest-level ibex indexing interface
 * @ib: an ibex
 * @name: the name of the file being indexed
 * @buffer: a buffer containing data from the file
 * @len: the length of @buffer
 * @unread: an output argument containing the number of unread bytes
 *
 * This routine indexes up to @len bytes from @buffer into @ib.
 * If @unread is NULL, the indexer assumes that the buffer ends on a
 * word boundary, and will index all the way to the end of the
 * buffer. If @unread is not NULL, and the buffer ends with an
 * alphabetic character, the indexer will assume that the buffer has
 * been cut off in the middle of a word, and return the number of
 * un-indexed bytes at the end of the buffer in *@unread. The caller
 * should then read in more data through whatever means it has
 * and pass in the unread bytes from the original buffer, followed
 * by the new data, on its next call.
 *
 * Return value: 0 on success, -1 on failure.
 **/
int
ibex_index_buffer (ibex *ib, char *name, char *buffer, size_t len, size_t *unread)
{
    char *p, *q, *nq, *end, *word;
    int wordsiz, cat = 0;
    GHashTable *words = g_hash_table_new(g_str_hash, g_str_equal);
    GPtrArray *wordlist = g_ptr_array_new();
    int i, ret=-1;

    if (unread)
        *unread = 0;

    end = buffer + len;
    wordsiz = 20;
    word = g_malloc (wordsiz);

    p = buffer;
    while (p < end) {
        while (p < end) {
            cat = utf8_category (p, &q, end);
            if (cat != IBEX_NONALPHA)
                break;
            p = q;
        }
        if (p == end) {
            goto done;
        } else if (cat == IBEX_INVALID) {
            goto error;
        } else if (cat == IBEX_INCOMPLETE)
            q = end;

        while (q < end) {
            cat = utf8_category (q, &nq, end);
            if (cat != IBEX_ALPHA)
                break;
            q = nq;
        }
        if (cat == IBEX_INVALID ||
            (cat == IBEX_INCOMPLETE && !unread)) {
            goto error;
        } else if (cat == IBEX_INCOMPLETE || (q == end && unread)) {
            *unread = end - p;
            goto done;
        }

        if (wordsiz < q - p + 1) {
            wordsiz = q - p + 1;
            word = g_realloc (word, wordsiz);
        }
        ibex_normalise_word (p, q, word);
        if (word[0]) {
            if (g_hash_table_lookup(words, word) == 0) {
                char *newword = g_strdup(word);
                g_ptr_array_add(wordlist, newword);
                g_hash_table_insert(words, newword, name);
            }
        }
        p = q;
    }
done:
    IBEX_LOCK(ib);

    d(printf("name %s count %d size %d\n", name, wordlist->len, len));
    if (!ib->predone) {
        ib->words->klass->index_pre(ib->words);
        ib->predone = TRUE;
    }
    ib->words->klass->add_list(ib->words, name, wordlist);

    IBEX_UNLOCK(ib);

    ret = 0;
error:
    for (i=0;i<wordlist->len;i++)
        g_free(wordlist->pdata[i]);
    g_ptr_array_free(wordlist, TRUE);
    g_hash_table_destroy(words);
    g_free (word);
    return ret;
}


ibex *ibex_open (char *file, int flags, int mode)
{
    ibex *ib;

    ib = g_malloc0(sizeof(*ib));
    ib->blocks = ibex_block_cache_open(file, flags, mode);
    if (ib->blocks == 0) {
        g_warning("create: Error occured?: %s\n", strerror(errno));
        g_free(ib);
        return NULL;
    }
    /* FIXME: the blockcache or the wordindex needs to manage the other one */
    ib->words = ib->blocks->words;

#ifdef ENABLE_THREADS
    ib->lock = g_mutex_new();
#endif
    return ib;
}

int ibex_save (ibex *ib)
{
    d(printf("syncing database\n"));

    IBEX_LOCK(ib);

    if (ib->predone) {
        ib->words->klass->index_post(ib->words);
        ib->predone = FALSE;
    }
    ib->words->klass->sync(ib->words);
    /* FIXME: some return */
    ibex_block_cache_sync(ib->blocks);

    IBEX_UNLOCK(ib);

    return 0;
}

int ibex_close (ibex *ib)
{
    int ret = 0;

    d(printf("closing database\n"));

    if (ib->predone) {
        ib->words->klass->index_post(ib->words);
        ib->predone = FALSE;
    }

    ib->words->klass->close(ib->words);
    ibex_block_cache_close(ib->blocks);
#ifdef ENABLE_THREADS
    g_mutex_free(ib->lock);
#endif
    g_free(ib);
    return ret;
}

void ibex_unindex (ibex *ib, char *name)
{
    d(printf("trying to unindex '%s'\n", name));
    IBEX_LOCK(ib);
    ib->words->klass->unindex_name(ib->words, name);
    IBEX_UNLOCK(ib);
}

GPtrArray *ibex_find (ibex *ib, char *word)
{
    char *normal;
    int len;
    GPtrArray *ret;

    len = strlen(word);
    normal = alloca(len+1);
    ibex_normalise_word(word, word+len, normal);
    IBEX_LOCK(ib);
    ret = ib->words->klass->find(ib->words, normal);
    IBEX_UNLOCK(ib);
    return ret;
}

gboolean ibex_find_name (ibex *ib, char *name, char *word)
{
    char *normal;
    int len;
    gboolean ret;

    len = strlen(word);
    normal = alloca(len+1);
    ibex_normalise_word(word, word+len, normal);
    IBEX_LOCK(ib);
    ret = ib->words->klass->find_name(ib->words, name, normal);
    IBEX_UNLOCK(ib);
    return ret;
}

gboolean ibex_contains_name(ibex *ib, char *name)
{
    gboolean ret;

    IBEX_LOCK(ib);
    ret = ib->words->klass->contains_name(ib->words, name);
    IBEX_UNLOCK(ib);
    return ret;
}