[b7d3cc34] | 1 | /********************************************************************\ |
---|
| 2 | * BitlBee -- An IRC to other IM-networks gateway * |
---|
| 3 | * * |
---|
[1719464] | 4 | * Copyright 2002-2006 Wilmer van der Gaast and others * |
---|
[b7d3cc34] | 5 | \********************************************************************/ |
---|
| 6 | |
---|
| 7 | /* |
---|
[c88999c] | 8 | * Various utility functions. Some are copied from Gaim to support the |
---|
| 9 | * IM-modules, most are from BitlBee. |
---|
[b7d3cc34] | 10 | * |
---|
| 11 | * Copyright (C) 1998-1999, Mark Spencer <markster@marko.net> |
---|
| 12 | * (and possibly other members of the Gaim team) |
---|
[1719464] | 13 | * Copyright 2002-2006 Wilmer van der Gaast <wilmer@gaast.net> |
---|
[b7d3cc34] | 14 | */ |
---|
| 15 | |
---|
| 16 | /* |
---|
| 17 | This program is free software; you can redistribute it and/or modify |
---|
| 18 | it under the terms of the GNU General Public License as published by |
---|
| 19 | the Free Software Foundation; either version 2 of the License, or |
---|
| 20 | (at your option) any later version. |
---|
| 21 | |
---|
| 22 | This program is distributed in the hope that it will be useful, |
---|
| 23 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
| 24 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
---|
| 25 | GNU General Public License for more details. |
---|
| 26 | |
---|
| 27 | You should have received a copy of the GNU General Public License with |
---|
| 28 | the Debian GNU/Linux distribution in /usr/share/common-licenses/GPL; |
---|
| 29 | if not, write to the Free Software Foundation, Inc., 59 Temple Place, |
---|
| 30 | Suite 330, Boston, MA 02111-1307 USA |
---|
| 31 | */ |
---|
| 32 | |
---|
| 33 | #define BITLBEE_CORE |
---|
| 34 | #include "nogaim.h" |
---|
| 35 | #include <stdio.h> |
---|
| 36 | #include <stdlib.h> |
---|
| 37 | #include <string.h> |
---|
[dd8d4c5] | 38 | #include <ctype.h> |
---|
[b7d3cc34] | 39 | #include <glib.h> |
---|
| 40 | #include <time.h> |
---|
| 41 | |
---|
| 42 | void strip_linefeed(gchar *text) |
---|
| 43 | { |
---|
| 44 | int i, j; |
---|
| 45 | gchar *text2 = g_malloc(strlen(text) + 1); |
---|
| 46 | |
---|
| 47 | for (i = 0, j = 0; text[i]; i++) |
---|
| 48 | if (text[i] != '\r') |
---|
| 49 | text2[j++] = text[i]; |
---|
| 50 | text2[j] = '\0'; |
---|
| 51 | |
---|
| 52 | strcpy(text, text2); |
---|
| 53 | g_free(text2); |
---|
| 54 | } |
---|
| 55 | |
---|
| 56 | char *add_cr(char *text) |
---|
| 57 | { |
---|
| 58 | char *ret = NULL; |
---|
| 59 | int count = 0, j; |
---|
| 60 | unsigned int i; |
---|
| 61 | |
---|
| 62 | if (text[0] == '\n') |
---|
| 63 | count++; |
---|
| 64 | for (i = 1; i < strlen(text); i++) |
---|
| 65 | if (text[i] == '\n' && text[i - 1] != '\r') |
---|
| 66 | count++; |
---|
| 67 | |
---|
| 68 | if (count == 0) |
---|
| 69 | return g_strdup(text); |
---|
| 70 | |
---|
| 71 | ret = g_malloc0(strlen(text) + count + 1); |
---|
| 72 | |
---|
| 73 | i = 0; j = 0; |
---|
| 74 | if (text[i] == '\n') |
---|
| 75 | ret[j++] = '\r'; |
---|
| 76 | ret[j++] = text[i++]; |
---|
| 77 | for (; i < strlen(text); i++) { |
---|
| 78 | if (text[i] == '\n' && text[i - 1] != '\r') |
---|
| 79 | ret[j++] = '\r'; |
---|
| 80 | ret[j++] = text[i]; |
---|
| 81 | } |
---|
| 82 | |
---|
| 83 | return ret; |
---|
| 84 | } |
---|
| 85 | |
---|
| 86 | char *normalize(const char *s) |
---|
| 87 | { |
---|
| 88 | static char buf[BUF_LEN]; |
---|
| 89 | char *t, *u; |
---|
| 90 | int x = 0; |
---|
| 91 | |
---|
| 92 | g_return_val_if_fail((s != NULL), NULL); |
---|
| 93 | |
---|
| 94 | u = t = g_strdup(s); |
---|
| 95 | |
---|
| 96 | strcpy(t, s); |
---|
| 97 | g_strdown(t); |
---|
| 98 | |
---|
| 99 | while (*t && (x < BUF_LEN - 1)) { |
---|
| 100 | if (*t != ' ') { |
---|
| 101 | buf[x] = *t; |
---|
| 102 | x++; |
---|
| 103 | } |
---|
| 104 | t++; |
---|
| 105 | } |
---|
| 106 | buf[x] = '\0'; |
---|
| 107 | g_free(u); |
---|
| 108 | return buf; |
---|
| 109 | } |
---|
| 110 | |
---|
| 111 | time_t get_time(int year, int month, int day, int hour, int min, int sec) |
---|
| 112 | { |
---|
| 113 | struct tm tm; |
---|
| 114 | |
---|
| 115 | tm.tm_year = year - 1900; |
---|
| 116 | tm.tm_mon = month - 1; |
---|
| 117 | tm.tm_mday = day; |
---|
| 118 | tm.tm_hour = hour; |
---|
| 119 | tm.tm_min = min; |
---|
| 120 | tm.tm_sec = sec >= 0 ? sec : time(NULL) % 60; |
---|
| 121 | return mktime(&tm); |
---|
| 122 | } |
---|
| 123 | |
---|
| 124 | typedef struct htmlentity |
---|
| 125 | { |
---|
| 126 | char code[8]; |
---|
[39cc341] | 127 | char is[4]; |
---|
[b7d3cc34] | 128 | } htmlentity_t; |
---|
| 129 | |
---|
| 130 | /* FIXME: This is ISO8859-1(5) centric, so might cause problems with other charsets. */ |
---|
| 131 | |
---|
[39cc341] | 132 | static const htmlentity_t ent[] = |
---|
[b7d3cc34] | 133 | { |
---|
[39cc341] | 134 | { "lt", "<" }, |
---|
| 135 | { "gt", ">" }, |
---|
| 136 | { "amp", "&" }, |
---|
| 137 | { "quot", "\"" }, |
---|
| 138 | { "aacute", "á" }, |
---|
| 139 | { "eacute", "é" }, |
---|
| 140 | { "iacute", "é" }, |
---|
| 141 | { "oacute", "ó" }, |
---|
| 142 | { "uacute", "ú" }, |
---|
| 143 | { "agrave", "à" }, |
---|
| 144 | { "egrave", "è" }, |
---|
| 145 | { "igrave", "ì" }, |
---|
| 146 | { "ograve", "ò" }, |
---|
| 147 | { "ugrave", "ù" }, |
---|
| 148 | { "acirc", "â" }, |
---|
| 149 | { "ecirc", "ê" }, |
---|
| 150 | { "icirc", "î" }, |
---|
| 151 | { "ocirc", "ô" }, |
---|
| 152 | { "ucirc", "û" }, |
---|
| 153 | { "auml", "ä" }, |
---|
| 154 | { "euml", "ë" }, |
---|
| 155 | { "iuml", "ï" }, |
---|
| 156 | { "ouml", "ö" }, |
---|
| 157 | { "uuml", "ü" }, |
---|
| 158 | { "nbsp", " " }, |
---|
| 159 | { "", "" } |
---|
[b7d3cc34] | 160 | }; |
---|
| 161 | |
---|
| 162 | void strip_html( char *in ) |
---|
| 163 | { |
---|
| 164 | char *start = in; |
---|
| 165 | char *out = g_malloc( strlen( in ) + 1 ); |
---|
| 166 | char *s = out, *cs; |
---|
| 167 | int i, matched; |
---|
| 168 | |
---|
| 169 | memset( out, 0, strlen( in ) + 1 ); |
---|
| 170 | |
---|
| 171 | while( *in ) |
---|
| 172 | { |
---|
| 173 | if( *in == '<' && ( isalpha( *(in+1) ) || *(in+1) == '/' ) ) |
---|
| 174 | { |
---|
| 175 | /* If in points at a < and in+1 points at a letter or a slash, this is probably |
---|
| 176 | a HTML-tag. Try to find a closing > and continue there. If the > can't be |
---|
| 177 | found, assume that it wasn't a HTML-tag after all. */ |
---|
| 178 | |
---|
| 179 | cs = in; |
---|
| 180 | |
---|
| 181 | while( *in && *in != '>' ) |
---|
| 182 | in ++; |
---|
| 183 | |
---|
| 184 | if( *in ) |
---|
| 185 | { |
---|
| 186 | if( g_strncasecmp( cs+1, "br", 2) == 0 ) |
---|
| 187 | *(s++) = '\n'; |
---|
| 188 | in ++; |
---|
| 189 | } |
---|
| 190 | else |
---|
| 191 | { |
---|
| 192 | in = cs; |
---|
| 193 | *(s++) = *(in++); |
---|
| 194 | } |
---|
| 195 | } |
---|
| 196 | else if( *in == '&' ) |
---|
| 197 | { |
---|
| 198 | cs = ++in; |
---|
| 199 | while( *in && isalpha( *in ) ) |
---|
| 200 | in ++; |
---|
| 201 | |
---|
| 202 | if( *in == ';' ) in ++; |
---|
| 203 | matched = 0; |
---|
| 204 | |
---|
| 205 | for( i = 0; *ent[i].code; i ++ ) |
---|
| 206 | if( g_strncasecmp( ent[i].code, cs, strlen( ent[i].code ) ) == 0 ) |
---|
| 207 | { |
---|
[39cc341] | 208 | int j; |
---|
| 209 | |
---|
| 210 | for( j = 0; ent[i].is[j]; j ++ ) |
---|
| 211 | *(s++) = ent[i].is[j]; |
---|
| 212 | |
---|
[b7d3cc34] | 213 | matched = 1; |
---|
| 214 | break; |
---|
| 215 | } |
---|
| 216 | |
---|
| 217 | /* None of the entities were matched, so return the string */ |
---|
| 218 | if( !matched ) |
---|
| 219 | { |
---|
| 220 | in = cs - 1; |
---|
| 221 | *(s++) = *(in++); |
---|
| 222 | } |
---|
| 223 | } |
---|
| 224 | else |
---|
| 225 | { |
---|
| 226 | *(s++) = *(in++); |
---|
| 227 | } |
---|
| 228 | } |
---|
| 229 | |
---|
| 230 | strcpy( start, out ); |
---|
| 231 | g_free( out ); |
---|
| 232 | } |
---|
| 233 | |
---|
| 234 | char *escape_html( const char *html ) |
---|
| 235 | { |
---|
| 236 | const char *c = html; |
---|
| 237 | GString *ret; |
---|
| 238 | char *str; |
---|
| 239 | |
---|
| 240 | if( html == NULL ) |
---|
| 241 | return( NULL ); |
---|
| 242 | |
---|
| 243 | ret = g_string_new( "" ); |
---|
| 244 | |
---|
| 245 | while( *c ) |
---|
| 246 | { |
---|
| 247 | switch( *c ) |
---|
| 248 | { |
---|
| 249 | case '&': |
---|
| 250 | ret = g_string_append( ret, "&" ); |
---|
| 251 | break; |
---|
| 252 | case '<': |
---|
| 253 | ret = g_string_append( ret, "<" ); |
---|
| 254 | break; |
---|
| 255 | case '>': |
---|
| 256 | ret = g_string_append( ret, ">" ); |
---|
| 257 | break; |
---|
| 258 | case '"': |
---|
| 259 | ret = g_string_append( ret, """ ); |
---|
| 260 | break; |
---|
| 261 | default: |
---|
| 262 | ret = g_string_append_c( ret, *c ); |
---|
| 263 | } |
---|
| 264 | c ++; |
---|
| 265 | } |
---|
| 266 | |
---|
| 267 | str = ret->str; |
---|
| 268 | g_string_free( ret, FALSE ); |
---|
| 269 | return( str ); |
---|
| 270 | } |
---|
| 271 | |
---|
| 272 | void info_string_append(GString *str, char *newline, char *name, char *value) |
---|
| 273 | { |
---|
| 274 | if( value && value[0] ) |
---|
| 275 | g_string_sprintfa( str, "%s%s: %s", newline, name, value ); |
---|
| 276 | } |
---|
[c88999c] | 277 | |
---|
| 278 | /* Decode%20a%20file%20name */ |
---|
| 279 | void http_decode( char *s ) |
---|
| 280 | { |
---|
| 281 | char *t; |
---|
| 282 | int i, j, k; |
---|
| 283 | |
---|
| 284 | t = g_new( char, strlen( s ) + 1 ); |
---|
| 285 | |
---|
| 286 | for( i = j = 0; s[i]; i ++, j ++ ) |
---|
| 287 | { |
---|
| 288 | if( s[i] == '%' ) |
---|
| 289 | { |
---|
| 290 | if( sscanf( s + i + 1, "%2x", &k ) ) |
---|
| 291 | { |
---|
| 292 | t[j] = k; |
---|
| 293 | i += 2; |
---|
| 294 | } |
---|
| 295 | else |
---|
| 296 | { |
---|
| 297 | *t = 0; |
---|
| 298 | break; |
---|
| 299 | } |
---|
| 300 | } |
---|
| 301 | else |
---|
| 302 | { |
---|
| 303 | t[j] = s[i]; |
---|
| 304 | } |
---|
| 305 | } |
---|
| 306 | t[j] = 0; |
---|
| 307 | |
---|
| 308 | strcpy( s, t ); |
---|
| 309 | g_free( t ); |
---|
| 310 | } |
---|
| 311 | |
---|
| 312 | /* Warning: This one explodes the string. Worst-cases can make the string 3x its original size! */ |
---|
| 313 | /* This fuction is safe, but make sure you call it safely as well! */ |
---|
| 314 | void http_encode( char *s ) |
---|
| 315 | { |
---|
| 316 | char *t; |
---|
| 317 | int i, j; |
---|
| 318 | |
---|
| 319 | t = g_strdup( s ); |
---|
| 320 | |
---|
| 321 | for( i = j = 0; t[i]; i ++, j ++ ) |
---|
| 322 | { |
---|
[dd8d4c5] | 323 | /* if( t[i] <= ' ' || ((unsigned char *)t)[i] >= 128 || t[i] == '%' ) */ |
---|
| 324 | if( !isalnum( t[i] ) ) |
---|
[c88999c] | 325 | { |
---|
| 326 | sprintf( s + j, "%%%02X", ((unsigned char*)t)[i] ); |
---|
| 327 | j += 2; |
---|
| 328 | } |
---|
| 329 | else |
---|
| 330 | { |
---|
| 331 | s[j] = t[i]; |
---|
| 332 | } |
---|
| 333 | } |
---|
| 334 | s[j] = 0; |
---|
| 335 | |
---|
| 336 | g_free( t ); |
---|
| 337 | } |
---|
| 338 | |
---|
| 339 | /* Strip newlines from a string. Modifies the string passed to it. */ |
---|
| 340 | char *strip_newlines( char *source ) |
---|
| 341 | { |
---|
| 342 | int i; |
---|
| 343 | |
---|
| 344 | for( i = 0; source[i] != '\0'; i ++ ) |
---|
| 345 | if( source[i] == '\n' || source[i] == '\r' ) |
---|
| 346 | source[i] = ' '; |
---|
| 347 | |
---|
| 348 | return source; |
---|
| 349 | } |
---|
[2a6ca4f] | 350 | |
---|
| 351 | #ifdef IPV6 |
---|
| 352 | /* Wrap an IPv4 address into IPv6 space. Not thread-safe... */ |
---|
| 353 | char *ipv6_wrap( char *src ) |
---|
| 354 | { |
---|
| 355 | static char dst[64]; |
---|
| 356 | int i; |
---|
| 357 | |
---|
| 358 | for( i = 0; src[i]; i ++ ) |
---|
| 359 | if( ( src[i] < '0' || src[i] > '9' ) && src[i] != '.' ) |
---|
| 360 | break; |
---|
| 361 | |
---|
| 362 | /* Hmm, it's not even an IP... */ |
---|
| 363 | if( src[i] ) |
---|
| 364 | return src; |
---|
| 365 | |
---|
| 366 | g_snprintf( dst, sizeof( dst ), "::ffff:%s", src ); |
---|
| 367 | |
---|
| 368 | return dst; |
---|
| 369 | } |
---|
| 370 | |
---|
| 371 | /* Unwrap an IPv4 address into IPv6 space. Thread-safe, because it's very simple. :-) */ |
---|
| 372 | char *ipv6_unwrap( char *src ) |
---|
| 373 | { |
---|
| 374 | int i; |
---|
| 375 | |
---|
| 376 | if( g_strncasecmp( src, "::ffff:", 7 ) != 0 ) |
---|
| 377 | return src; |
---|
| 378 | |
---|
| 379 | for( i = 7; src[i]; i ++ ) |
---|
| 380 | if( ( src[i] < '0' || src[i] > '9' ) && src[i] != '.' ) |
---|
| 381 | break; |
---|
| 382 | |
---|
| 383 | /* Hmm, it's not even an IP... */ |
---|
| 384 | if( src[i] ) |
---|
| 385 | return src; |
---|
| 386 | |
---|
| 387 | return ( src + 7 ); |
---|
| 388 | } |
---|
| 389 | #endif |
---|
[e27661d] | 390 | |
---|
| 391 | /* Convert from one charset to another. |
---|
| 392 | |
---|
| 393 | from_cs, to_cs: Source and destination charsets |
---|
| 394 | src, dst: Source and destination strings |
---|
| 395 | size: Size if src. 0 == use strlen(). strlen() is not reliable for UNICODE/UTF16 strings though. |
---|
| 396 | maxbuf: Maximum number of bytes to write to dst |
---|
| 397 | |
---|
| 398 | Returns the number of bytes written to maxbuf or -1 on an error. |
---|
| 399 | */ |
---|
| 400 | signed int do_iconv( char *from_cs, char *to_cs, char *src, char *dst, size_t size, size_t maxbuf ) |
---|
| 401 | { |
---|
[574af7e] | 402 | GIConv cd; |
---|
[e27661d] | 403 | size_t res; |
---|
| 404 | size_t inbytesleft, outbytesleft; |
---|
| 405 | char *inbuf = src; |
---|
| 406 | char *outbuf = dst; |
---|
| 407 | |
---|
[574af7e] | 408 | cd = g_iconv_open( to_cs, from_cs ); |
---|
| 409 | if( cd == (GIConv) -1 ) |
---|
[e27661d] | 410 | return( -1 ); |
---|
| 411 | |
---|
| 412 | inbytesleft = size ? size : strlen( src ); |
---|
| 413 | outbytesleft = maxbuf - 1; |
---|
[574af7e] | 414 | res = g_iconv( cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft ); |
---|
[e27661d] | 415 | *outbuf = '\0'; |
---|
[574af7e] | 416 | g_iconv_close( cd ); |
---|
[e27661d] | 417 | |
---|
| 418 | if( res == (size_t) -1 ) |
---|
| 419 | return( -1 ); |
---|
| 420 | else |
---|
| 421 | return( outbuf - dst ); |
---|
| 422 | } |
---|
| 423 | |
---|
[1719464] | 424 | /* A pretty reliable random number generator. Tries to use the /dev/random |
---|
| 425 | devices first, and falls back to the random number generator from libc |
---|
| 426 | when it fails. Opens randomizer devices with O_NONBLOCK to make sure a |
---|
| 427 | lack of entropy won't halt BitlBee. */ |
---|
| 428 | void random_bytes( unsigned char *buf, int count ) |
---|
[e27661d] | 429 | { |
---|
[1719464] | 430 | static int use_dev = -1; |
---|
| 431 | |
---|
| 432 | /* Actually this probing code isn't really necessary, is it? */ |
---|
| 433 | if( use_dev == -1 ) |
---|
| 434 | { |
---|
| 435 | if( access( "/dev/random", R_OK ) == 0 || access( "/dev/urandom", R_OK ) == 0 ) |
---|
| 436 | use_dev = 1; |
---|
| 437 | else |
---|
| 438 | { |
---|
| 439 | use_dev = 0; |
---|
| 440 | srand( ( getpid() << 16 ) ^ time( NULL ) ); |
---|
| 441 | } |
---|
| 442 | } |
---|
| 443 | |
---|
| 444 | if( use_dev ) |
---|
| 445 | { |
---|
| 446 | int fd; |
---|
| 447 | |
---|
| 448 | /* At least on Linux, /dev/random can block if there's not |
---|
| 449 | enough entropy. We really don't want that, so if it can't |
---|
| 450 | give anything, use /dev/urandom instead. */ |
---|
| 451 | if( ( fd = open( "/dev/random", O_RDONLY | O_NONBLOCK ) ) >= 0 ) |
---|
| 452 | if( read( fd, buf, count ) == count ) |
---|
| 453 | { |
---|
| 454 | close( fd ); |
---|
| 455 | return; |
---|
| 456 | } |
---|
| 457 | close( fd ); |
---|
| 458 | |
---|
| 459 | /* urandom isn't supposed to block at all, but just to be |
---|
| 460 | sure. If it blocks, we'll disable use_dev and use the libc |
---|
| 461 | randomizer instead. */ |
---|
| 462 | if( ( fd = open( "/dev/urandom", O_RDONLY | O_NONBLOCK ) ) >= 0 ) |
---|
| 463 | if( read( fd, buf, count ) == count ) |
---|
| 464 | { |
---|
| 465 | close( fd ); |
---|
| 466 | return; |
---|
| 467 | } |
---|
| 468 | close( fd ); |
---|
| 469 | |
---|
| 470 | /* If /dev/random blocks once, we'll still try to use it |
---|
| 471 | again next time. If /dev/urandom also fails for some |
---|
| 472 | reason, stick with libc during this session. */ |
---|
| 473 | |
---|
| 474 | use_dev = 0; |
---|
| 475 | srand( ( getpid() << 16 ) ^ time( NULL ) ); |
---|
| 476 | } |
---|
| 477 | |
---|
| 478 | if( !use_dev ) |
---|
| 479 | { |
---|
| 480 | int i; |
---|
| 481 | |
---|
| 482 | /* Possibly the LSB of rand() isn't very random on some |
---|
| 483 | platforms. Seems okay on at least Linux and OSX though. */ |
---|
| 484 | for( i = 0; i < count; i ++ ) |
---|
| 485 | buf[i] = rand() & 0xff; |
---|
| 486 | } |
---|
[e27661d] | 487 | } |
---|