1 /* -*- Mode: C; tab-width: 8; indent-tabs-mode: t; c-basic-offset: 8 -*- */
3 * Copyright (C) 2000-2012 Jeffrey Stedfast
5 * This library is free software; you can redistribute it and/or
6 * modify it under the terms of the GNU Lesser General Public License
7 * as published by the Free Software Foundation; either version 2.1
8 * of the License, or (at your option) any later version.
10 * This library is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 * Lesser General Public License for more details.
15 * You should have received a copy of the GNU Lesser General Public
16 * License along with this library; if not, write to the Free
17 * Software Foundation, 51 Franklin Street, Fifth Floor, Boston, MA
26 #include "gmime-filter-windows.h"
27 #include "gmime-charset.h"
33 * SECTION: gmime-filter-windows
34 * @title: GMimeFilterWindows
35 * @short_description: Determine if text is in a Microsoft Windows codepage
36 * @see_also: #GMimeFilter
38 * A #GMimeFilter used for determining if text marked as iso-8859-##
39 * is actually encoded in one of the Windows-CP125# charsets.
43 static void g_mime_filter_windows_class_init (GMimeFilterWindowsClass *klass);
44 static void g_mime_filter_windows_init (GMimeFilterWindows *filter, GMimeFilterWindowsClass *klass);
45 static void g_mime_filter_windows_finalize (GObject *object);
47 static GMimeFilter *filter_copy (GMimeFilter *filter);
48 static void filter_filter (GMimeFilter *filter, char *in, size_t len, size_t prespace,
49 char **out, size_t *outlen, size_t *outprespace);
50 static void filter_complete (GMimeFilter *filter, char *in, size_t len, size_t prespace,
51 char **out, size_t *outlen, size_t *outprespace);
52 static void filter_reset (GMimeFilter *filter);
55 static GMimeFilterClass *parent_class = NULL;
59 g_mime_filter_windows_get_type (void)
61 static GType type = 0;
64 static const GTypeInfo info = {
65 sizeof (GMimeFilterWindowsClass),
66 NULL, /* base_class_init */
67 NULL, /* base_class_finalize */
68 (GClassInitFunc) g_mime_filter_windows_class_init,
69 NULL, /* class_finalize */
70 NULL, /* class_data */
71 sizeof (GMimeFilterWindows),
73 (GInstanceInitFunc) g_mime_filter_windows_init,
76 type = g_type_register_static (GMIME_TYPE_FILTER, "GMimeFilterWindows", &info, 0);
84 g_mime_filter_windows_class_init (GMimeFilterWindowsClass *klass)
86 GObjectClass *object_class = G_OBJECT_CLASS (klass);
87 GMimeFilterClass *filter_class = GMIME_FILTER_CLASS (klass);
89 parent_class = g_type_class_ref (GMIME_TYPE_FILTER);
91 object_class->finalize = g_mime_filter_windows_finalize;
93 filter_class->copy = filter_copy;
94 filter_class->filter = filter_filter;
95 filter_class->complete = filter_complete;
96 filter_class->reset = filter_reset;
100 g_mime_filter_windows_init (GMimeFilterWindows *filter, GMimeFilterWindowsClass *klass)
102 filter->claimed_charset = NULL;
103 filter->is_windows = FALSE;
107 g_mime_filter_windows_finalize (GObject *object)
109 GMimeFilterWindows *filter = (GMimeFilterWindows *) object;
111 g_free (filter->claimed_charset);
113 G_OBJECT_CLASS (parent_class)->finalize (object);
118 filter_copy (GMimeFilter *filter)
120 GMimeFilterWindows *windows = (GMimeFilterWindows *) filter;
122 return g_mime_filter_windows_new (windows->claimed_charset);
126 filter_filter (GMimeFilter *filter, char *in, size_t len, size_t prespace,
127 char **out, size_t *outlen, size_t *outprespace)
129 GMimeFilterWindows *windows = (GMimeFilterWindows *) filter;
130 register unsigned char *inptr;
131 unsigned char *inend;
133 if (!windows->is_windows) {
134 inptr = (unsigned char *) in;
137 while (inptr < inend) {
138 register unsigned char c = *inptr++;
140 if (c >= 128 && c <= 159) {
141 d(g_warning ("Encountered text encoded in a Windows charset trying "
142 "to pass itself off as being encoded in %s",
143 windows->claimed_charset));
144 windows->is_windows = TRUE;
152 *outprespace = prespace;
156 filter_complete (GMimeFilter *filter, char *in, size_t len, size_t prespace,
157 char **out, size_t *outlen, size_t *outprespace)
159 filter_filter (filter, in, len, prespace, out, outlen, outprespace);
163 filter_reset (GMimeFilter *filter)
165 GMimeFilterWindows *windows = (GMimeFilterWindows *) filter;
167 windows->is_windows = FALSE;
172 * g_mime_filter_windows_new:
173 * @claimed_charset: charset that a text stream claims to be
175 * Creates a new GMimeFilterWindows filter. When a stream of text has
176 * been filtered, it can be determined whether or not said text stream
177 * was in @claimed_charset or the equivalent Windows-CP125# charset.
179 * Returns: a new windows filter.
182 g_mime_filter_windows_new (const char *claimed_charset)
184 GMimeFilterWindows *new;
186 g_return_val_if_fail (claimed_charset != NULL, NULL);
188 new = g_object_newv (GMIME_TYPE_FILTER_WINDOWS, 0, NULL);
189 new->claimed_charset = g_strdup (claimed_charset);
191 return (GMimeFilter *) new;
196 * g_mime_filter_windows_is_windows_charset:
197 * @filter: windows filter object
199 * Determines whether or not a Windows-CP125# charset has been
202 * Returns: %TRUE if the filtered stream has been detected to contain
203 * Windows-CP125# characters or %FALSE otherwise.
206 g_mime_filter_windows_is_windows_charset (GMimeFilterWindows *filter)
208 g_return_val_if_fail (GMIME_IS_FILTER_WINDOWS (filter), FALSE);
210 return filter->is_windows;
215 * g_mime_filter_windows_real_charset:
216 * @filter: windows filter object
218 * Figures out the real charset that the text is encoded in based on whether or not Windows-CP125# characters were found.
220 * Returns: a const string pointer to the claimed charset if filtered
221 * text stream was found not to contain any Windows-CP125# characters
222 * or the proper Windows-CP125# charset.
225 g_mime_filter_windows_real_charset (GMimeFilterWindows *filter)
227 g_return_val_if_fail (GMIME_IS_FILTER_WINDOWS (filter), NULL);
229 if (filter->is_windows)
230 return g_mime_charset_iso_to_windows (filter->claimed_charset);
232 return filter->claimed_charset;