path: root/libempathy-gtk/empathy-spell.c

/* -*- Mode: C; tab-width: 8; indent-tabs-mode: t; c-basic-offset: 8 -*- */
/*
 * Copyright (C) 2004-2007 Imendio AB
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License as
 * published by the Free Software Foundation; either version 2 of the
 * License, or (at your option) any later version.
 *
 * This program 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
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public
 * License along with this program; if not, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 02111-1307, USA.
 *
 * Authors: Martyn Russell <martyn@imendio.com>
 *          Richard Hult <richard@imendio.com>
 */

#include "config.h"

#include <string.h>
#include <stdlib.h>

#include <glib/gi18n.h>

#ifdef HAVE_ASPELL
#include <aspell.h>
#endif

#include <libempathy/empathy-debug.h>

#include "empathy-spell.h"
#include "empathy-conf.h"
#include "empathy-preferences.h"

#define DEBUG_DOMAIN "Spell"

#ifdef HAVE_ASPELL

/* Note: We could use aspell_reset_cache (NULL); periodically if we wanted
 * to...
 */

typedef struct {
    AspellConfig       *spell_config;
    AspellCanHaveError *spell_possible_err;
    AspellSpeller      *spell_checker;
} SpellLanguage;

#define ISO_CODES_DATADIR    ISO_CODES_PREFIX "/share/xml/iso-codes"
#define ISO_CODES_LOCALESDIR ISO_CODES_PREFIX "/share/locale"

static GHashTable  *iso_code_names = NULL;
static GList       *languages = NULL;
static gboolean     empathy_conf_notify_inited = FALSE;

static void
spell_iso_codes_parse_start_tag (GMarkupParseContext  *ctx,
                 const gchar          *element_name,
                 const gchar         **attr_names,
                 const gchar         **attr_values,
                 gpointer              data,
                 GError              **error)
{
    const gchar *ccode_longB, *ccode_longT, *ccode;
    const gchar *lang_name;

    if (!g_str_equal (element_name, "iso_639_entry") ||
        attr_names == NULL || attr_values == NULL) {
        return;
    }

    ccode = NULL;
    ccode_longB = NULL;
    ccode_longT = NULL;
    lang_name = NULL;

    while (*attr_names && *attr_values) {
        if (g_str_equal (*attr_names, "iso_639_1_code")) {
            if (**attr_values) {
                ccode = *attr_values;
            }
        }
        else if (g_str_equal (*attr_names, "iso_639_2B_code")) {
            if (**attr_values) {
                ccode_longB = *attr_values;
            }
        }
        else if (g_str_equal (*attr_names, "iso_639_2T_code")) {
            if (**attr_values) {
                ccode_longT = *attr_values;
            }
        }
        else if (g_str_equal (*attr_names, "name")) {
            lang_name = *attr_values;
        }

        attr_names++;
        attr_values++;
    }

    if (!lang_name) {
        return;
    }

    if (ccode) {
        g_hash_table_insert (iso_code_names,
                     g_strdup (ccode),
                     g_strdup (lang_name));
    }

    if (ccode_longB) {
        g_hash_table_insert (iso_code_names,
                     g_strdup (ccode_longB),
                     g_strdup (lang_name));
    }

    if (ccode_longT) {
        g_hash_table_insert (iso_code_names,
                     g_strdup (ccode_longT),
                     g_strdup (lang_name));
    }
}

static void
spell_iso_code_names_init (void)
{
    GError *err = NULL;
    gchar  *buf;
    gsize   buf_len;

    iso_code_names = g_hash_table_new_full (g_str_hash, g_str_equal,
                        g_free, g_free);

    bindtextdomain ("iso_639", ISO_CODES_LOCALESDIR);
    bind_textdomain_codeset ("iso_639", "UTF-8");

    /* FIXME: We should read this in chunks and pass to the parser. */
    if (g_file_get_contents (ISO_CODES_DATADIR "/iso_639.xml", &buf, &buf_len, &err)) {
        GMarkupParseContext *ctx;
        GMarkupParser        parser = {
            spell_iso_codes_parse_start_tag,
            NULL, NULL, NULL, NULL
        };

        ctx = g_markup_parse_context_new (&parser, 0, NULL, NULL);
        if (!g_markup_parse_context_parse (ctx, buf, buf_len, &err)) {
            g_warning ("Failed to parse '%s': %s",
                   ISO_CODES_DATADIR"/iso_639.xml",
                   err->message);
            g_error_free (err);
        }

        g_markup_parse_context_free (ctx);
        g_free (buf);
    } else {
        g_warning ("Failed to load '%s': %s",
                ISO_CODES_DATADIR"/iso_639.xml", err->message);
        g_error_free (err);
    }
}

static void
spell_notify_languages_cb (EmpathyConf  *conf,
               const gchar *key,
               gpointer     user_data)
{
    GList *l;

    empathy_debug (DEBUG_DOMAIN, "Resetting languages due to config change");

    /* We just reset the languages list. */
    for (l = languages; l; l = l->next) {
        SpellLanguage *lang;

        lang = l->data;

        delete_aspell_config (lang->spell_config);
        delete_aspell_speller (lang->spell_checker);

        g_slice_free (SpellLanguage, lang);
    }

    g_list_free (languages);
    languages = NULL;
}

static void
spell_setup_languages (void)
{
    gchar  *str;

    if (!empathy_conf_notify_inited) {
        empathy_conf_notify_add (empathy_conf_get (),
                     EMPATHY_PREFS_CHAT_SPELL_CHECKER_LANGUAGES,
                     spell_notify_languages_cb, NULL);

        empathy_conf_notify_inited = TRUE;
    }

    if (languages) {
        return;
    }

    if (empathy_conf_get_string (empathy_conf_get (),
                     EMPATHY_PREFS_CHAT_SPELL_CHECKER_LANGUAGES,
                     &str) && str) {
        gchar **strv;
        gint    i;

        strv = g_strsplit (str, ",", -1);

        i = 0;
        while (strv && strv[i]) {
            SpellLanguage *lang;

            empathy_debug (DEBUG_DOMAIN, "Setting up language:'%s'", strv[i]);

            lang = g_slice_new0 (SpellLanguage);

            lang->spell_config = new_aspell_config();

            aspell_config_replace (lang->spell_config, "encoding", "utf-8");
            aspell_config_replace (lang->spell_config, "lang", strv[i++]);

            lang->spell_possible_err = new_aspell_speller (lang->spell_config);

            if (aspell_error_number (lang->spell_possible_err) == 0) {
                lang->spell_checker = to_aspell_speller (lang->spell_possible_err);
                languages = g_list_append (languages, lang);
            } else {
                delete_aspell_config (lang->spell_config);
                g_slice_free (SpellLanguage, lang);
            }
        }

        if (strv) {
            g_strfreev (strv);
        }

        g_free (str);
    }
}

const char *
empathy_spell_get_language_name (const char *code)
{
    const gchar *name;

    g_return_val_if_fail (code != NULL, NULL);

    if (!iso_code_names) {
        spell_iso_code_names_init ();
    }

    name = g_hash_table_lookup (iso_code_names, code);
    if (!name) {
        return NULL;
    }

    return dgettext ("iso_639", name);
}

GList *
empathy_spell_get_language_codes (void)
{
    AspellConfig              *config;
    AspellDictInfoList        *dlist;
    AspellDictInfoEnumeration *dels;
    const AspellDictInfo      *entry;
    GList                     *codes = NULL;

    config = new_aspell_config ();
    dlist = get_aspell_dict_info_list (config);
    dels = aspell_dict_info_list_elements (dlist);

    while ((entry = aspell_dict_info_enumeration_next (dels)) != 0) {
        if (g_list_find_custom (codes, entry->code, (GCompareFunc) strcmp)) {
            continue;
        }

        codes = g_list_append (codes, g_strdup (entry->code));
    }

    delete_aspell_dict_info_enumeration (dels);
    delete_aspell_config (config);

    return codes;
}

void
empathy_spell_free_language_codes (GList *codes)
{
    g_list_foreach (codes, (GFunc) g_free, NULL);
    g_list_free (codes);
}

gboolean
empathy_spell_check (const gchar *word)
{
    GList       *l;
    gint         n_langs;
    gboolean     correct = FALSE;
    gint         len;
    const gchar *p;
    gunichar     c;
    gboolean     digit;

    g_return_val_if_fail (word != NULL, FALSE);

    spell_setup_languages ();

    if (!languages) {
        empathy_debug (DEBUG_DOMAIN, "No languages to check against");
        return TRUE;
    }

    /* Ignore certain cases like numbers, etc. */
    for (p = word, digit = TRUE; *p && digit; p = g_utf8_next_char (p)) {
        c = g_utf8_get_char (p);
        digit = g_unichar_isdigit (c);
    }

    if (digit) {
        /* We don't spell check digits. */
        empathy_debug (DEBUG_DOMAIN, "Not spell checking word:'%s', it is all digits", word);
        return TRUE;
    }

    len = strlen (word);
    n_langs = g_list_length (languages);
    for (l = languages; l; l = l->next) {
        SpellLanguage *lang;

        lang = l->data;

        correct = aspell_speller_check (lang->spell_checker, word, len);
        if (n_langs > 1 && correct) {
            break;
        }
    }

    return correct;
}

GList *
empathy_spell_get_suggestions (const gchar *word)
{
    GList                   *l1;
    GList                   *l2 = NULL;
    const AspellWordList    *suggestions;
    AspellStringEnumeration *elements;
    const char              *next;
    gint                     len;

    g_return_val_if_fail (word != NULL, NULL);

    spell_setup_languages ();

    len = strlen (word);

    for (l1 = languages; l1; l1 = l1->next) {
        SpellLanguage *lang;

        lang = l1->data;

        suggestions = aspell_speller_suggest (lang->spell_checker,
                              word, len);

        elements = aspell_word_list_elements (suggestions);

        while ((next = aspell_string_enumeration_next (elements))) {
            l2 = g_list_append (l2, g_strdup (next));
        }

        delete_aspell_string_enumeration (elements);
    }

    return l2;
}

gboolean
empathy_spell_supported (void)
{
    if (g_getenv ("EMPATHY_SPELL_DISABLED")) {
        empathy_debug (DEBUG_DOMAIN, "EMPATHY_SPELL_DISABLE env variable defined");
        return FALSE;
    }

    return TRUE;
}

#else /* not HAVE_ASPELL */

gboolean
empathy_spell_supported (void)
{
    return FALSE;
}

GList *
empathy_spell_get_suggestions (const gchar *word)
{
    empathy_debug (DEBUG_DOMAIN, "Support disabled, could not get suggestions");

    return NULL;
}

gboolean
empathy_spell_check (const gchar *word)
{
    empathy_debug (DEBUG_DOMAIN, "Support disabled, could not check spelling");

    return TRUE;
}

const char *
empathy_spell_get_language_name (const char *lang)
{
    empathy_debug (DEBUG_DOMAIN, "Support disabled, could not get language name");

    return NULL;
}

GList *
empathy_spell_get_language_codes (void)
{
    empathy_debug (DEBUG_DOMAIN, "Support disabled, could not get language codes");

    return NULL;
}

void
empathy_spell_free_language_codes (GList *codes)
{
}

#endif /* HAVE_ASPELL */


void
empathy_spell_free_suggestions (GList *suggestions)
{
    g_list_foreach (suggestions, (GFunc) g_free, NULL);
    g_list_free (suggestions);
}