1 /* -*- Mode: C; indent-tabs-mode: nil; c-basic-offset: 8 -*- */
2
3 /* libcroco - Library for parsing and applying CSS
4 * Copyright (C) 2006-2019 Free Software Foundation, Inc.
5 *
6 * This file is not part of the GNU gettext program, but is used with
7 * GNU gettext.
8 *
9 * The original copyright notice is as follows:
10 */
11
12 /*
13 * This file is part of The Croco Library
14 *
15 * Copyright (C) 2002-2003 Dodji Seketeli <dodji@seketeli.org>
16 *
17 * This program is free software; you can redistribute it and/or
18 * modify it under the terms of version 2.1 of the GNU Lesser General Public
19 * License as published by the Free Software Foundation.
20 *
21 * This program is distributed in the hope that it will be useful,
22 * but WITHOUT ANY WARRANTY; without even the implied warranty of
23 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
24 * GNU General Public License for more details.
25 *
26 * You should have received a copy of the GNU Lesser General Public License
27 * along with this program; if not, write to the Free Software
28 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
29 * USA
30 */
31
32 /*
33 *$Id$
34 */
35
36 /**
37 *@file
38 *The definition of the #CREncHandler class.
39 */
40
41 #include <config.h>
42 #include "cr-enc-handler.h"
43 #include "cr-utils.h"
44
45 #include <string.h>
46
47 struct CREncAlias {
48 const gchar *name;
49 enum CREncoding encoding;
50 };
51
52 static struct CREncAlias gv_default_aliases[] = {
53 {"UTF-8", CR_UTF_8},
54 {"UTF_8", CR_UTF_8},
55 {"UTF8", CR_UTF_8},
56 {"UTF-16", CR_UTF_16},
57 {"UTF_16", CR_UTF_16},
58 {"UTF16", CR_UTF_16},
59 {"UCS1", CR_UCS_1},
60 {"UCS-1", CR_UCS_1},
61 {"UCS_1", CR_UCS_1},
62 {"ISO-8859-1", CR_UCS_1},
63 {"ISO_8859-1", CR_UCS_1},
64 {"UCS-1", CR_UCS_1},
65 {"UCS_1", CR_UCS_1},
66 {"UCS4", CR_UCS_4},
67 {"UCS-4", CR_UCS_4},
68 {"UCS_4", CR_UCS_4},
69 {"ASCII", CR_ASCII},
70 {0, 0}
71 };
72
73 static CREncHandler gv_default_enc_handlers[] = {
74 {CR_UCS_1, cr_utils_ucs1_to_utf8, cr_utils_utf8_to_ucs1,
75 cr_utils_ucs1_str_len_as_utf8, cr_utils_utf8_str_len_as_ucs1},
76
77 {CR_ISO_8859_1, cr_utils_ucs1_to_utf8, cr_utils_utf8_to_ucs1,
78 cr_utils_ucs1_str_len_as_utf8, cr_utils_utf8_str_len_as_ucs1},
79
80 {CR_ASCII, cr_utils_ucs1_to_utf8, cr_utils_utf8_to_ucs1,
81 cr_utils_ucs1_str_len_as_utf8, cr_utils_utf8_str_len_as_ucs1},
82
83 {0, NULL, NULL, NULL, NULL}
84 };
85
86 /**
87 * cr_enc_handler_get_instance:
88 *@a_enc: the encoding of the Handler.
89 *
90 *Gets the instance of encoding handler.
91 *This function implements a singleton pattern.
92 *
93 *Returns the instance of #CREncHandler.
94 */
95 CREncHandler *
96 cr_enc_handler_get_instance (enum CREncoding a_enc)
97 {
98 gulong i = 0;
99
100 for (i = 0; gv_default_enc_handlers[i].encoding; i++) {
101 if (gv_default_enc_handlers[i].encoding == a_enc) {
102 return (CREncHandler *) & gv_default_enc_handlers[i];
103 }
104 }
105
106 return NULL;
107 }
108
109 /**
110 * cr_enc_handler_resolve_enc_alias:
111 *@a_alias_name: the encoding name.
112 *@a_enc: output param. The returned encoding type
113 *or 0 if the alias is not supported.
114 *
115 *Given an encoding name (called an alias name)
116 *the function returns the matching encoding type.
117 *
118 *Returns CR_OK upon successfull completion, an error code otherwise.
119 */
120 enum CRStatus
121 cr_enc_handler_resolve_enc_alias (const guchar * a_alias_name,
122 enum CREncoding *a_enc)
123 {
124 gulong i = 0;
125 guchar *alias_name_up = NULL;
126 enum CRStatus status = CR_ENCODING_NOT_FOUND_ERROR;
127
128 g_return_val_if_fail (a_alias_name != NULL, CR_BAD_PARAM_ERROR);
129
130 alias_name_up = (guchar *) g_ascii_strup ((const gchar *) a_alias_name, -1);
131
132 for (i = 0; gv_default_aliases[i].name; i++) {
133 if (!strcmp (gv_default_aliases[i].name, (const gchar *) alias_name_up)) {
134 *a_enc = gv_default_aliases[i].encoding;
135 status = CR_OK;
136 break;
137 }
138 }
139
140 return status;
141 }
142
143 /**
144 * cr_enc_handler_convert_input:
145 *@a_this: the current instance of #CREncHandler.
146 *@a_in: the input buffer to convert.
147 *@a_in_len: in/out parameter. The len of the input
148 *buffer to convert. After return, contains the number of
149 *bytes actually consumed.
150 *@a_out: output parameter. The converted output buffer.
151 *Must be freed by the buffer.
152 *@a_out_len: output parameter. The length of the output buffer.
153 *
154 *Converts a raw input buffer into an utf8 buffer.
155 *
156 *Returns CR_OK upon successfull completion, an error code otherwise.
157 */
158 enum CRStatus
159 cr_enc_handler_convert_input (CREncHandler * a_this,
160 const guchar * a_in,
161 gulong * a_in_len,
162 guchar ** a_out, gulong * a_out_len)
163 {
164 enum CRStatus status = CR_OK;
165
166 g_return_val_if_fail (a_this && a_in && a_in_len && a_out,
167 CR_BAD_PARAM_ERROR);
168
169 if (a_this->decode_input == NULL)
170 return CR_OK;
171
172 if (a_this->enc_str_len_as_utf8) {
173 status = a_this->enc_str_len_as_utf8 (a_in,
174 &a_in[*a_in_len - 1],
175 a_out_len);
176
177 g_return_val_if_fail (status == CR_OK, status);
178 } else {
179 *a_out_len = *a_in_len;
180 }
181
182 *a_out = g_malloc0 (*a_out_len);
183
184 status = a_this->decode_input (a_in, a_in_len, *a_out, a_out_len);
185
186 if (status != CR_OK) {
187 g_free (*a_out);
188 *a_out = NULL;
189 }
190
191 g_return_val_if_fail (status == CR_OK, status);
192
193 return CR_OK;
194 }