?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Removed calls to strerror() function from own libsamba implementation.
[midnight-commander.git] / vfs / samba / lib / charset.c
blob58be25308214199f509f14ece5ac78939b8b2e37
1 /*
2 Unix SMB/Netbios implementation.
3 Version 1.9.
4 Character set handling
5 Copyright (C) Andrew Tridgell 1992-1998
6
7 This program is free software; you can redistribute it and/or modify
8 it under the terms of the GNU General Public License as published by
9 the Free Software Foundation; either version 2 of the License, or
10 (at your option) any later version.
11
12 This program is distributed in the hope that it will be useful,
13 but WITHOUT ANY WARRANTY; without even the implied warranty of
14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 GNU General Public License for more details.
16
17 You should have received a copy of the GNU General Public License
18 along with this program; if not, write to the Free Software
19 Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
20 */
21
22 #define CHARSET_C
23 #include "includes.h"
24
25 const char *unix_error_string (int error_num);
26 extern int DEBUGLEVEL;
27
28 /*
29 * Codepage definitions.
30 */
31
32 #if !defined(KANJI)
33 /* lower->upper mapping for IBM Code Page 850 - MS-DOS Latin 1 */
34 unsigned char const cp_850[][4] = {
35 /* dec col/row oct hex description */
36 /* 133 08/05 205 85 a grave */
37 /* 183 11/07 267 B7 A grave */ {0x85,0xB7,1,1},
38 /* 160 10/00 240 A0 a acute */
39 /* 181 11/05 265 B5 A acute */ {0xA0,0xB5,1,1},
40 /* 131 08/03 203 83 a circumflex */
41 /* 182 11/06 266 B6 A circumflex */ {0x83,0xB6,1,1},
42 /* 198 12/06 306 C6 a tilde */
43 /* 199 12/07 307 C7 A tilde */ {0xC6,0xC7,1,1},
44 /* 132 08/04 204 84 a diaeresis */
45 /* 142 08/14 216 8E A diaeresis */ {0x84,0x8E,1,1},
46 /* 134 08/06 206 86 a ring */
47 /* 143 08/15 217 8F A ring */ {0x86,0x8F,1,1},
48 /* 145 09/01 221 91 ae diphthong */
49 /* 146 09/02 222 92 AE diphthong */ {0x91,0x92,1,1},
50 /* 135 08/07 207 87 c cedilla */
51 /* 128 08/00 200 80 C cedilla */ {0x87,0x80,1,1},
52 /* 138 08/10 212 8A e grave */
53 /* 212 13/04 324 D4 E grave */ {0x8A,0xD4,1,1},
54 /* 130 08/02 202 82 e acute */
55 /* 144 09/00 220 90 E acute */ {0x82,0x90,1,1},
56 /* 136 08/08 210 88 e circumflex */
57 /* 210 13/02 322 D2 E circumflex */ {0x88,0xD2,1,1},
58 /* 137 08/09 211 89 e diaeresis */
59 /* 211 13/03 323 D3 E diaeresis */ {0x89,0xD3,1,1},
60 /* 141 08/13 215 8D i grave */
61 /* 222 13/14 336 DE I grave */ {0x8D,0xDE,1,1},
62 /* 161 10/01 241 A1 i acute */
63 /* 214 13/06 326 D6 I acute */ {0xA1,0xD6,1,1},
64 /* 140 08/12 214 8C i circumflex */
65 /* 215 13/07 327 D7 I circumflex */ {0x8C,0xD7,1,1},
66 /* 139 08/11 213 8B i diaeresis */
67 /* 216 13/08 330 D8 I diaeresis */ {0x8B,0xD8,1,1},
68 /* 208 13/00 320 D0 Icelandic eth */
69 /* 209 13/01 321 D1 Icelandic Eth */ {0xD0,0xD1,1,1},
70 /* 164 10/04 244 A4 n tilde */
71 /* 165 10/05 245 A5 N tilde */ {0xA4,0xA5,1,1},
72 /* 149 09/05 225 95 o grave */
73 /* 227 14/03 343 E3 O grave */ {0x95,0xE3,1,1},
74 /* 162 10/02 242 A2 o acute */
75 /* 224 14/00 340 E0 O acute */ {0xA2,0xE0,1,1},
76 /* 147 09/03 223 93 o circumflex */
77 /* 226 14/02 342 E2 O circumflex */ {0x93,0xE2,1,1},
78 /* 228 14/04 344 E4 o tilde */
79 /* 229 14/05 345 E5 O tilde */ {0xE4,0xE5,1,1},
80 /* 148 09/04 224 94 o diaeresis */
81 /* 153 09/09 231 99 O diaeresis */ {0x94,0x99,1,1},
82 /* 155 09/11 233 9B o slash */
83 /* 157 09/13 235 9D O slash */ {0x9B,0x9D,1,1},
84 /* 151 09/07 227 97 u grave */
85 /* 235 14/11 353 EB U grave */ {0x97,0xEB,1,1},
86 /* 163 10/03 243 A3 u acute */
87 /* 233 14/09 351 E9 U acute */ {0xA3,0xE9,1,1},
88 /* 150 09/06 226 96 u circumflex */
89 /* 234 14/10 352 EA U circumflex */ {0x96,0xEA,1,1},
90 /* 129 08/01 201 81 u diaeresis */
91 /* 154 09/10 232 9A U diaeresis */ {0x81,0x9A,1,1},
92 /* 236 14/12 354 EC y acute */
93 /* 237 14/13 355 ED Y acute */ {0xEC,0xED,1,1},
94 /* 231 14/07 347 E7 Icelandic thorn */
95 /* 232 14/08 350 E8 Icelandic Thorn */ {0xE7,0xE8,1,1},
96
97 {0x9C,0,0,0}, /* Pound */
98 {0,0,0,0}
99 };
100 #else /* KANJI */
101 /* lower->upper mapping for IBM Code Page 932 - MS-DOS Japanese SJIS */
102 unsigned char const cp_932[][4] = {
103 {0,0,0,0}
104 };
105 #endif /* KANJI */
106
107 char xx_dos_char_map[256];
108 char xx_upper_char_map[256];
109 char xx_lower_char_map[256];
110
111 char *dos_char_map = xx_dos_char_map;
112 char *upper_char_map = xx_upper_char_map;
113 char *lower_char_map = xx_lower_char_map;
114
115 /*
116 * This code has been extended to deal with ascynchronous mappings
117 * like MS-DOS Latin US (Code page 437) where things like :
118 * a acute are capitalized to 'A', but the reverse mapping
119 * must not hold true. This allows the filename case insensitive
120 * matching in do_match() to work, as the DOS/Win95/NT client
121 * uses 'A' as a mask to match against characters like a acute.
122 * This is the meaning behind the parameters that allow a
123 * mapping from lower to upper, but not upper to lower.
124 */
125
126 static void add_dos_char(int lower, BOOL map_lower_to_upper,
127 int upper, BOOL map_upper_to_lower)
128 {
129 lower &= 0xff;
130 upper &= 0xff;
131 DEBUGADD( 6, ( "Adding chars 0x%x 0x%x (l->u = %s) (u->l = %s)\n",
132 lower, upper,
133 map_lower_to_upper ? "True" : "False",
134 map_upper_to_lower ? "True" : "False" ) );
135 if (lower) dos_char_map[lower] = 1;
136 if (upper) dos_char_map[upper] = 1;
137 lower_char_map[lower] = (char)lower; /* Define tolower(lower) */
138 upper_char_map[upper] = (char)upper; /* Define toupper(upper) */
139 if (lower && upper) {
140 if(map_upper_to_lower)
141 lower_char_map[upper] = (char)lower;
142 if(map_lower_to_upper)
143 upper_char_map[lower] = (char)upper;
144 }
145 }
146
147 /****************************************************************************
148 initialise the charset arrays
149 ****************************************************************************/
150 void charset_initialise(void)
151 {
152 int i;
153
154 #ifdef LC_ALL
155 /* include <locale.h> in includes.h if available for OS */
156 /* we take only standard 7-bit ASCII definitions from ctype */
157 setlocale(LC_ALL,"C");
158 #endif
159
160 for (i= 0;i<=255;i++) {
161 dos_char_map[i] = 0;
162 }
163
164 for (i=0;i<=127;i++) {
165 if (isalnum(i) || strchr("._^$~!#%&-{}()@'`",(char)i))
166 add_dos_char(i,False,0,False);
167 }
168
169 for (i=0; i<=255; i++) {
170 char c = (char)i;
171 upper_char_map[i] = lower_char_map[i] = c;
172
173 /* Some systems have buggy isupper/islower for characters
174 above 127. Best not to rely on them. */
175 if(i < 128) {
176 if (isupper((int)c)) lower_char_map[i] = tolower(c);
177 if (islower((int)c)) upper_char_map[i] = toupper(c);
178 }
179 }
180 }
181
182 /****************************************************************************
183 load the client codepage.
184 ****************************************************************************/
185
186 typedef const unsigned char (*codepage_p)[4];
187
188 static codepage_p load_client_codepage( int client_codepage )
189 {
190 pstring codepage_file_name;
191 unsigned char buf[8];
192 FILE *fp = NULL;
193 SMB_OFF_T size;
194 codepage_p cp_p = NULL;
195 SMB_STRUCT_STAT st;
196
197 DEBUG(5, ("load_client_codepage: loading codepage %d.\n", client_codepage));
198
199 if(strlen(CODEPAGEDIR) + 14 > sizeof(codepage_file_name))
200 {
201 DEBUG(0,("load_client_codepage: filename too long to load\n"));
202 return NULL;
203 }
204
205 pstrcpy(codepage_file_name, CODEPAGEDIR);
206 pstrcat(codepage_file_name, "/");
207 pstrcat(codepage_file_name, "codepage.");
208 slprintf(&codepage_file_name[strlen(codepage_file_name)],
209 sizeof(pstring)-(strlen(codepage_file_name)+1),
210 "%03d",
211 client_codepage);
212
213 if(sys_stat(codepage_file_name,&st)!=0)
214 {
215 DEBUG(0,("load_client_codepage: filename %s does not exist.\n",
216 codepage_file_name));
217 return NULL;
218 }
219
220 /* Check if it is at least big enough to hold the required
221 data. Should be 2 byte version, 2 byte codepage, 4 byte length,
222 plus zero or more bytes of data. Note that the data cannot be more
223 than 4 * MAXCODEPAGELINES bytes.
224 */
225 size = st.st_size;
226
227 if( size < CODEPAGE_HEADER_SIZE || size > (CODEPAGE_HEADER_SIZE + 4 * MAXCODEPAGELINES))
228 {
229 DEBUG(0,("load_client_codepage: file %s is an incorrect size for a \
230 code page file (size=%d).\n", codepage_file_name, (int)size));
231 return NULL;
232 }
233
234 /* Read the first 8 bytes of the codepage file - check
235 the version number and code page number. All the data
236 is held in little endian format.
237 */
238
239 if((fp = sys_fopen( codepage_file_name, "r")) == NULL)
240 {
241 DEBUG(0,("load_client_codepage: cannot open file %s. Error was %s\n",
242 codepage_file_name, unix_error_string (errno)));
243 return NULL;
244 }
245
246 if(fread( buf, 1, CODEPAGE_HEADER_SIZE, fp)!=CODEPAGE_HEADER_SIZE)
247 {
248 DEBUG(0,("load_client_codepage: cannot read header from file %s. Error was %s\n",
249 codepage_file_name, unix_error_string (errno)));
250 goto clean_and_exit;
251 }
252
253 /* Check the version value */
254 if(SVAL(buf,CODEPAGE_VERSION_OFFSET) != CODEPAGE_FILE_VERSION_ID)
255 {
256 DEBUG(0,("load_client_codepage: filename %s has incorrect version id. \
257 Needed %hu, got %hu.\n",
258 codepage_file_name, (uint16)CODEPAGE_FILE_VERSION_ID,
259 SVAL(buf,CODEPAGE_VERSION_OFFSET)));
260 goto clean_and_exit;
261 }
262
263 /* Check the codepage matches */
264 if(SVAL(buf,CODEPAGE_CLIENT_CODEPAGE_OFFSET) != (uint16)client_codepage)
265 {
266 DEBUG(0,("load_client_codepage: filename %s has incorrect codepage. \
267 Needed %hu, got %hu.\n",
268 codepage_file_name, (uint16)client_codepage,
269 SVAL(buf,CODEPAGE_CLIENT_CODEPAGE_OFFSET)));
270 goto clean_and_exit;
271 }
272
273 /* Check the length is correct. */
274 if(IVAL(buf,CODEPAGE_LENGTH_OFFSET) != (size - CODEPAGE_HEADER_SIZE))
275 {
276 DEBUG(0,("load_client_codepage: filename %s has incorrect size headers. \
277 Needed %u, got %u.\n", codepage_file_name, (uint32)(size - CODEPAGE_HEADER_SIZE),
278 IVAL(buf,CODEPAGE_LENGTH_OFFSET)));
279 goto clean_and_exit;
280 }
281
282 size -= CODEPAGE_HEADER_SIZE; /* Remove header */
283
284 /* Make sure the size is a multiple of 4. */
285 if((size % 4 ) != 0)
286 {
287 DEBUG(0,("load_client_codepage: filename %s has a codepage size not a \
288 multiple of 4.\n", codepage_file_name));
289 goto clean_and_exit;
290 }
291
292 /* Allocate space for the code page file and read it all in. */
293 if((cp_p = (codepage_p)malloc( size + 4 )) == NULL)
294 {
295 DEBUG(0,("load_client_codepage: malloc fail.\n"));
296 goto clean_and_exit;
297 }
298
299 if(fread( (char *)cp_p, 1, size, fp)!=size)
300 {
301 DEBUG(0,("load_client_codepage: read fail on file %s. Error was %s.\n",
302 codepage_file_name, unix_error_string (errno)));
303 goto clean_and_exit;
304 }
305
306 /* Ensure array is correctly terminated. */
307 memset(((char *)cp_p) + size, '\0', 4);
308
309 fclose(fp);
310 return cp_p;
311
312 clean_and_exit:
313
314 /* pseudo destructor :-) */
315
316 if(fp != NULL)
317 fclose(fp);
318 if(cp_p)
319 free((char *)cp_p);
320 return NULL;
321 }
322
323 /****************************************************************************
324 initialise the client codepage.
325 ****************************************************************************/
326 void codepage_initialise(int client_codepage)
327 {
328 int i;
329 static codepage_p cp = NULL;
330
331 if(cp != NULL)
332 {
333 DEBUG(6,
334 ("codepage_initialise: called twice - ignoring second client code page = %d\n",
335 client_codepage));
336 return;
337 }
338
339 DEBUG(6,("codepage_initialise: client code page = %d\n", client_codepage));
340
341 /*
342 * Known client codepages - these can be added to.
343 */
344 cp = load_client_codepage( client_codepage );
345
346 if(cp == NULL)
347 {
348 #ifdef KANJI
349 DEBUG(6,("codepage_initialise: loading dynamic codepage file %s/codepage.%d \
350 for code page %d failed. Using default client codepage 932\n",
351 CODEPAGEDIR, client_codepage, client_codepage));
352 cp = cp_932;
353 client_codepage = KANJI_CODEPAGE;
354 #else /* KANJI */
355 DEBUG(6,("codepage_initialise: loading dynamic codepage file %s/codepage.%d \
356 for code page %d failed. Using default client codepage 850\n",
357 CODEPAGEDIR, client_codepage, client_codepage));
358 cp = cp_850;
359 client_codepage = MSDOS_LATIN_1_CODEPAGE;
360 #endif /* KANJI */
361 }
362
363 /*
364 * Setup the function pointers for the loaded codepage.
365 */
366 initialize_multibyte_vectors( client_codepage );
367
368 if(cp)
369 {
370 for(i = 0; !((cp[i][0] == '\0') && (cp[i][1] == '\0')); i++)
371 add_dos_char(cp[i][0], (BOOL)cp[i][2], cp[i][1], (BOOL)cp[i][3]);
372 }
373 }
374
375 /*******************************************************************
376 add characters depending on a string passed by the user
377 ********************************************************************/
378 void add_char_string(const char *s)
379 {
380 char *extra_chars = (char *)strdup(s);
381 char *t;
382 if (!extra_chars) return;
383
384 for (t=strtok(extra_chars," \t\r\n"); t; t=strtok(NULL," \t\r\n")) {
385 char c1=0,c2=0;
386 int i1=0,i2=0;
387 if (isdigit((unsigned char)*t) || (*t)=='-') {
388 sscanf(t,"%i:%i",&i1,&i2);
389 add_dos_char(i1,True,i2,True);
390 } else {
391 sscanf(t,"%c:%c",&c1,&c2);
392 add_dos_char((unsigned char)c1,True,(unsigned char)c2, True);
393 }
394 }
395
396 free(extra_chars);
397 }
Midnight Commander