| OLD | NEW |
|---|
| (Empty) |
| 1 /* | |
| 2 * Copyright 2004 The WebRTC Project Authors. All rights reserved. | |
| 3 * | |
| 4 * Use of this source code is governed by a BSD-style license | |
| 5 * that can be found in the LICENSE file in the root of the source | |
| 6 * tree. An additional intellectual property rights grant can be found | |
| 7 * in the file PATENTS. All contributing project authors may | |
| 8 * be found in the AUTHORS file in the root of the source tree. | |
| 9 */ | |
| 10 | |
| 11 #include "webrtc/base/stringencode.h" | |
| 12 | |
| 13 #include <stdio.h> | |
| 14 #include <stdlib.h> | |
| 15 | |
| 16 #include "webrtc/base/checks.h" | |
| 17 #include "webrtc/base/stringutils.h" | |
| 18 | |
| 19 namespace rtc { | |
| 20 | |
| 21 ///////////////////////////////////////////////////////////////////////////// | |
| 22 // String Encoding Utilities | |
| 23 ///////////////////////////////////////////////////////////////////////////// | |
| 24 | |
| 25 size_t escape(char * buffer, size_t buflen, | |
| 26 const char * source, size_t srclen, | |
| 27 const char * illegal, char escape) { | |
| 28 RTC_DCHECK(buffer); // TODO(grunell): estimate output size | |
| 29 if (buflen <= 0) | |
| 30 return 0; | |
| 31 | |
| 32 size_t srcpos = 0, bufpos = 0; | |
| 33 while ((srcpos < srclen) && (bufpos + 1 < buflen)) { | |
| 34 char ch = source[srcpos++]; | |
| 35 if ((ch == escape) || ::strchr(illegal, ch)) { | |
| 36 if (bufpos + 2 >= buflen) | |
| 37 break; | |
| 38 buffer[bufpos++] = escape; | |
| 39 } | |
| 40 buffer[bufpos++] = ch; | |
| 41 } | |
| 42 | |
| 43 buffer[bufpos] = '\0'; | |
| 44 return bufpos; | |
| 45 } | |
| 46 | |
| 47 size_t unescape(char * buffer, size_t buflen, | |
| 48 const char * source, size_t srclen, | |
| 49 char escape) { | |
| 50 RTC_DCHECK(buffer); // TODO(grunell): estimate output size | |
| 51 if (buflen <= 0) | |
| 52 return 0; | |
| 53 | |
| 54 size_t srcpos = 0, bufpos = 0; | |
| 55 while ((srcpos < srclen) && (bufpos + 1 < buflen)) { | |
| 56 char ch = source[srcpos++]; | |
| 57 if ((ch == escape) && (srcpos < srclen)) { | |
| 58 ch = source[srcpos++]; | |
| 59 } | |
| 60 buffer[bufpos++] = ch; | |
| 61 } | |
| 62 buffer[bufpos] = '\0'; | |
| 63 return bufpos; | |
| 64 } | |
| 65 | |
| 66 size_t encode(char * buffer, size_t buflen, | |
| 67 const char * source, size_t srclen, | |
| 68 const char * illegal, char escape) { | |
| 69 RTC_DCHECK(buffer); // TODO(grunell): estimate output size | |
| 70 if (buflen <= 0) | |
| 71 return 0; | |
| 72 | |
| 73 size_t srcpos = 0, bufpos = 0; | |
| 74 while ((srcpos < srclen) && (bufpos + 1 < buflen)) { | |
| 75 char ch = source[srcpos++]; | |
| 76 if ((ch != escape) && !::strchr(illegal, ch)) { | |
| 77 buffer[bufpos++] = ch; | |
| 78 } else if (bufpos + 3 >= buflen) { | |
| 79 break; | |
| 80 } else { | |
| 81 buffer[bufpos+0] = escape; | |
| 82 buffer[bufpos+1] = hex_encode((static_cast<unsigned char>(ch) >> 4) & 0xF)
; | |
| 83 buffer[bufpos+2] = hex_encode((static_cast<unsigned char>(ch) ) & 0xF)
; | |
| 84 bufpos += 3; | |
| 85 } | |
| 86 } | |
| 87 buffer[bufpos] = '\0'; | |
| 88 return bufpos; | |
| 89 } | |
| 90 | |
| 91 size_t decode(char * buffer, size_t buflen, | |
| 92 const char * source, size_t srclen, | |
| 93 char escape) { | |
| 94 if (buflen <= 0) | |
| 95 return 0; | |
| 96 | |
| 97 unsigned char h1, h2; | |
| 98 size_t srcpos = 0, bufpos = 0; | |
| 99 while ((srcpos < srclen) && (bufpos + 1 < buflen)) { | |
| 100 char ch = source[srcpos++]; | |
| 101 if ((ch == escape) | |
| 102 && (srcpos + 1 < srclen) | |
| 103 && hex_decode(source[srcpos], &h1) | |
| 104 && hex_decode(source[srcpos+1], &h2)) { | |
| 105 buffer[bufpos++] = (h1 << 4) | h2; | |
| 106 srcpos += 2; | |
| 107 } else { | |
| 108 buffer[bufpos++] = ch; | |
| 109 } | |
| 110 } | |
| 111 buffer[bufpos] = '\0'; | |
| 112 return bufpos; | |
| 113 } | |
| 114 | |
| 115 const char* unsafe_filename_characters() { | |
| 116 // It might be better to have a single specification which is the union of | |
| 117 // all operating systems, unless one system is overly restrictive. | |
| 118 #if defined(WEBRTC_WIN) | |
| 119 return "\\/:*?\"<>|"; | |
| 120 #else // !WEBRTC_WIN | |
| 121 // TODO(grunell): Should this never be reached? | |
| 122 RTC_NOTREACHED(); | |
| 123 return ""; | |
| 124 #endif // !WEBRTC_WIN | |
| 125 } | |
| 126 | |
| 127 const unsigned char URL_UNSAFE = 0x1; // 0-33 "#$%&+,/:;<=>?@[\]^`{|} 127 | |
| 128 const unsigned char XML_UNSAFE = 0x2; // "&'<> | |
| 129 const unsigned char HTML_UNSAFE = 0x2; // "&'<> | |
| 130 | |
| 131 // ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 6 5 7 8 9 : ; < = > ? | |
| 132 //@ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z [ \ ] ^ _ | |
| 133 //` a b c d e f g h i j k l m n o p q r s t u v w x y z { | } ~ | |
| 134 | |
| 135 const unsigned char ASCII_CLASS[128] = { | |
| 136 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, | |
| 137 1,0,3,1,1,1,3,2,0,0,0,1,1,0,0,1,0,0,0,0,0,0,0,0,0,0,1,1,3,1,3,1, | |
| 138 1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,0, | |
| 139 1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,0,1, | |
| 140 }; | |
| 141 | |
| 142 size_t url_encode(char * buffer, size_t buflen, | |
| 143 const char * source, size_t srclen) { | |
| 144 if (nullptr == buffer) | |
| 145 return srclen * 3 + 1; | |
| 146 if (buflen <= 0) | |
| 147 return 0; | |
| 148 | |
| 149 size_t srcpos = 0, bufpos = 0; | |
| 150 while ((srcpos < srclen) && (bufpos + 1 < buflen)) { | |
| 151 unsigned char ch = source[srcpos++]; | |
| 152 if ((ch < 128) && (ASCII_CLASS[ch] & URL_UNSAFE)) { | |
| 153 if (bufpos + 3 >= buflen) { | |
| 154 break; | |
| 155 } | |
| 156 buffer[bufpos+0] = '%'; | |
| 157 buffer[bufpos+1] = hex_encode((ch >> 4) & 0xF); | |
| 158 buffer[bufpos+2] = hex_encode((ch ) & 0xF); | |
| 159 bufpos += 3; | |
| 160 } else { | |
| 161 buffer[bufpos++] = ch; | |
| 162 } | |
| 163 } | |
| 164 buffer[bufpos] = '\0'; | |
| 165 return bufpos; | |
| 166 } | |
| 167 | |
| 168 size_t url_decode(char * buffer, size_t buflen, | |
| 169 const char * source, size_t srclen) { | |
| 170 if (nullptr == buffer) | |
| 171 return srclen + 1; | |
| 172 if (buflen <= 0) | |
| 173 return 0; | |
| 174 | |
| 175 unsigned char h1, h2; | |
| 176 size_t srcpos = 0, bufpos = 0; | |
| 177 while ((srcpos < srclen) && (bufpos + 1 < buflen)) { | |
| 178 unsigned char ch = source[srcpos++]; | |
| 179 if (ch == '+') { | |
| 180 buffer[bufpos++] = ' '; | |
| 181 } else if ((ch == '%') | |
| 182 && (srcpos + 1 < srclen) | |
| 183 && hex_decode(source[srcpos], &h1) | |
| 184 && hex_decode(source[srcpos+1], &h2)) | |
| 185 { | |
| 186 buffer[bufpos++] = (h1 << 4) | h2; | |
| 187 srcpos += 2; | |
| 188 } else { | |
| 189 buffer[bufpos++] = ch; | |
| 190 } | |
| 191 } | |
| 192 buffer[bufpos] = '\0'; | |
| 193 return bufpos; | |
| 194 } | |
| 195 | |
| 196 size_t utf8_decode(const char* source, size_t srclen, unsigned long* value) { | |
| 197 const unsigned char* s = reinterpret_cast<const unsigned char*>(source); | |
| 198 if ((s[0] & 0x80) == 0x00) { // Check s[0] == 0xxxxxxx | |
| 199 *value = s[0]; | |
| 200 return 1; | |
| 201 } | |
| 202 if ((srclen < 2) || ((s[1] & 0xC0) != 0x80)) { // Check s[1] != 10xxxxxx | |
| 203 return 0; | |
| 204 } | |
| 205 // Accumulate the trailer byte values in value16, and combine it with the | |
| 206 // relevant bits from s[0], once we've determined the sequence length. | |
| 207 unsigned long value16 = (s[1] & 0x3F); | |
| 208 if ((s[0] & 0xE0) == 0xC0) { // Check s[0] == 110xxxxx | |
| 209 *value = ((s[0] & 0x1F) << 6) | value16; | |
| 210 return 2; | |
| 211 } | |
| 212 if ((srclen < 3) || ((s[2] & 0xC0) != 0x80)) { // Check s[2] != 10xxxxxx | |
| 213 return 0; | |
| 214 } | |
| 215 value16 = (value16 << 6) | (s[2] & 0x3F); | |
| 216 if ((s[0] & 0xF0) == 0xE0) { // Check s[0] == 1110xxxx | |
| 217 *value = ((s[0] & 0x0F) << 12) | value16; | |
| 218 return 3; | |
| 219 } | |
| 220 if ((srclen < 4) || ((s[3] & 0xC0) != 0x80)) { // Check s[3] != 10xxxxxx | |
| 221 return 0; | |
| 222 } | |
| 223 value16 = (value16 << 6) | (s[3] & 0x3F); | |
| 224 if ((s[0] & 0xF8) == 0xF0) { // Check s[0] == 11110xxx | |
| 225 *value = ((s[0] & 0x07) << 18) | value16; | |
| 226 return 4; | |
| 227 } | |
| 228 return 0; | |
| 229 } | |
| 230 | |
| 231 size_t utf8_encode(char* buffer, size_t buflen, unsigned long value) { | |
| 232 if ((value <= 0x7F) && (buflen >= 1)) { | |
| 233 buffer[0] = static_cast<unsigned char>(value); | |
| 234 return 1; | |
| 235 } | |
| 236 if ((value <= 0x7FF) && (buflen >= 2)) { | |
| 237 buffer[0] = 0xC0 | static_cast<unsigned char>(value >> 6); | |
| 238 buffer[1] = 0x80 | static_cast<unsigned char>(value & 0x3F); | |
| 239 return 2; | |
| 240 } | |
| 241 if ((value <= 0xFFFF) && (buflen >= 3)) { | |
| 242 buffer[0] = 0xE0 | static_cast<unsigned char>(value >> 12); | |
| 243 buffer[1] = 0x80 | static_cast<unsigned char>((value >> 6) & 0x3F); | |
| 244 buffer[2] = 0x80 | static_cast<unsigned char>(value & 0x3F); | |
| 245 return 3; | |
| 246 } | |
| 247 if ((value <= 0x1FFFFF) && (buflen >= 4)) { | |
| 248 buffer[0] = 0xF0 | static_cast<unsigned char>(value >> 18); | |
| 249 buffer[1] = 0x80 | static_cast<unsigned char>((value >> 12) & 0x3F); | |
| 250 buffer[2] = 0x80 | static_cast<unsigned char>((value >> 6) & 0x3F); | |
| 251 buffer[3] = 0x80 | static_cast<unsigned char>(value & 0x3F); | |
| 252 return 4; | |
| 253 } | |
| 254 return 0; | |
| 255 } | |
| 256 | |
| 257 size_t html_encode(char * buffer, size_t buflen, | |
| 258 const char * source, size_t srclen) { | |
| 259 RTC_DCHECK(buffer); // TODO(grunell): estimate output size | |
| 260 if (buflen <= 0) | |
| 261 return 0; | |
| 262 | |
| 263 size_t srcpos = 0, bufpos = 0; | |
| 264 while ((srcpos < srclen) && (bufpos + 1 < buflen)) { | |
| 265 unsigned char ch = source[srcpos]; | |
| 266 if (ch < 128) { | |
| 267 srcpos += 1; | |
| 268 if (ASCII_CLASS[ch] & HTML_UNSAFE) { | |
| 269 const char * escseq = 0; | |
| 270 size_t esclen = 0; | |
| 271 switch (ch) { | |
| 272 case '<': escseq = "<"; esclen = 4; break; | |
| 273 case '>': escseq = ">"; esclen = 4; break; | |
| 274 case '\'': escseq = "'"; esclen = 5; break; | |
| 275 case '\"': escseq = """; esclen = 6; break; | |
| 276 case '&': escseq = "&"; esclen = 5; break; | |
| 277 default: RTC_NOTREACHED(); | |
| 278 } | |
| 279 if (bufpos + esclen >= buflen) { | |
| 280 break; | |
| 281 } | |
| 282 memcpy(buffer + bufpos, escseq, esclen); | |
| 283 bufpos += esclen; | |
| 284 } else { | |
| 285 buffer[bufpos++] = ch; | |
| 286 } | |
| 287 } else { | |
| 288 // Largest value is 0x1FFFFF => � (10 characters) | |
| 289 const size_t kEscseqSize = 11; | |
| 290 char escseq[kEscseqSize]; | |
| 291 unsigned long val; | |
| 292 if (size_t vallen = utf8_decode(&source[srcpos], srclen - srcpos, &val)) { | |
| 293 srcpos += vallen; | |
| 294 } else { | |
| 295 // Not a valid utf8 sequence, just use the raw character. | |
| 296 val = static_cast<unsigned char>(source[srcpos++]); | |
| 297 } | |
| 298 size_t esclen = sprintfn(escseq, kEscseqSize, "&#%lu;", val); | |
| 299 if (bufpos + esclen >= buflen) { | |
| 300 break; | |
| 301 } | |
| 302 memcpy(buffer + bufpos, escseq, esclen); | |
| 303 bufpos += esclen; | |
| 304 } | |
| 305 } | |
| 306 buffer[bufpos] = '\0'; | |
| 307 return bufpos; | |
| 308 } | |
| 309 | |
| 310 size_t html_decode(char * buffer, size_t buflen, | |
| 311 const char * source, size_t srclen) { | |
| 312 RTC_DCHECK(buffer); // TODO(grunell): estimate output size | |
| 313 return xml_decode(buffer, buflen, source, srclen); | |
| 314 } | |
| 315 | |
| 316 size_t xml_encode(char * buffer, size_t buflen, | |
| 317 const char * source, size_t srclen) { | |
| 318 RTC_DCHECK(buffer); // TODO(grunell): estimate output size | |
| 319 if (buflen <= 0) | |
| 320 return 0; | |
| 321 | |
| 322 size_t srcpos = 0, bufpos = 0; | |
| 323 while ((srcpos < srclen) && (bufpos + 1 < buflen)) { | |
| 324 unsigned char ch = source[srcpos++]; | |
| 325 if ((ch < 128) && (ASCII_CLASS[ch] & XML_UNSAFE)) { | |
| 326 const char * escseq = 0; | |
| 327 size_t esclen = 0; | |
| 328 switch (ch) { | |
| 329 case '<': escseq = "<"; esclen = 4; break; | |
| 330 case '>': escseq = ">"; esclen = 4; break; | |
| 331 case '\'': escseq = "'"; esclen = 6; break; | |
| 332 case '\"': escseq = """; esclen = 6; break; | |
| 333 case '&': escseq = "&"; esclen = 5; break; | |
| 334 default: RTC_NOTREACHED(); | |
| 335 } | |
| 336 if (bufpos + esclen >= buflen) { | |
| 337 break; | |
| 338 } | |
| 339 memcpy(buffer + bufpos, escseq, esclen); | |
| 340 bufpos += esclen; | |
| 341 } else { | |
| 342 buffer[bufpos++] = ch; | |
| 343 } | |
| 344 } | |
| 345 buffer[bufpos] = '\0'; | |
| 346 return bufpos; | |
| 347 } | |
| 348 | |
| 349 size_t xml_decode(char * buffer, size_t buflen, | |
| 350 const char * source, size_t srclen) { | |
| 351 RTC_DCHECK(buffer); // TODO(grunell): estimate output size | |
| 352 if (buflen <= 0) | |
| 353 return 0; | |
| 354 | |
| 355 size_t srcpos = 0, bufpos = 0; | |
| 356 while ((srcpos < srclen) && (bufpos + 1 < buflen)) { | |
| 357 unsigned char ch = source[srcpos++]; | |
| 358 if (ch != '&') { | |
| 359 buffer[bufpos++] = ch; | |
| 360 } else if ((srcpos + 2 < srclen) | |
| 361 && (memcmp(source + srcpos, "lt;", 3) == 0)) { | |
| 362 buffer[bufpos++] = '<'; | |
| 363 srcpos += 3; | |
| 364 } else if ((srcpos + 2 < srclen) | |
| 365 && (memcmp(source + srcpos, "gt;", 3) == 0)) { | |
| 366 buffer[bufpos++] = '>'; | |
| 367 srcpos += 3; | |
| 368 } else if ((srcpos + 4 < srclen) | |
| 369 && (memcmp(source + srcpos, "apos;", 5) == 0)) { | |
| 370 buffer[bufpos++] = '\''; | |
| 371 srcpos += 5; | |
| 372 } else if ((srcpos + 4 < srclen) | |
| 373 && (memcmp(source + srcpos, "quot;", 5) == 0)) { | |
| 374 buffer[bufpos++] = '\"'; | |
| 375 srcpos += 5; | |
| 376 } else if ((srcpos + 3 < srclen) | |
| 377 && (memcmp(source + srcpos, "amp;", 4) == 0)) { | |
| 378 buffer[bufpos++] = '&'; | |
| 379 srcpos += 4; | |
| 380 } else if ((srcpos < srclen) && (source[srcpos] == '#')) { | |
| 381 int int_base = 10; | |
| 382 if ((srcpos + 1 < srclen) && (source[srcpos+1] == 'x')) { | |
| 383 int_base = 16; | |
| 384 srcpos += 1; | |
| 385 } | |
| 386 char * ptr; | |
| 387 // TODO(grunell): Fix hack (ptr may go past end of data) | |
| 388 unsigned long val = strtoul(source + srcpos + 1, &ptr, int_base); | |
| 389 if ((static_cast<size_t>(ptr - source) < srclen) && (*ptr == ';')) { | |
| 390 srcpos = ptr - source + 1; | |
| 391 } else { | |
| 392 // Not a valid escape sequence. | |
| 393 break; | |
| 394 } | |
| 395 if (size_t esclen = utf8_encode(buffer + bufpos, buflen - bufpos, val)) { | |
| 396 bufpos += esclen; | |
| 397 } else { | |
| 398 // Not enough room to encode the character, or illegal character | |
| 399 break; | |
| 400 } | |
| 401 } else { | |
| 402 // Unrecognized escape sequence. | |
| 403 break; | |
| 404 } | |
| 405 } | |
| 406 buffer[bufpos] = '\0'; | |
| 407 return bufpos; | |
| 408 } | |
| 409 | |
| 410 static const char HEX[] = "0123456789abcdef"; | |
| 411 | |
| 412 char hex_encode(unsigned char val) { | |
| 413 RTC_DCHECK_LT(val, 16); | |
| 414 return (val < 16) ? HEX[val] : '!'; | |
| 415 } | |
| 416 | |
| 417 bool hex_decode(char ch, unsigned char* val) { | |
| 418 if ((ch >= '0') && (ch <= '9')) { | |
| 419 *val = ch - '0'; | |
| 420 } else if ((ch >= 'A') && (ch <= 'Z')) { | |
| 421 *val = (ch - 'A') + 10; | |
| 422 } else if ((ch >= 'a') && (ch <= 'z')) { | |
| 423 *val = (ch - 'a') + 10; | |
| 424 } else { | |
| 425 return false; | |
| 426 } | |
| 427 return true; | |
| 428 } | |
| 429 | |
| 430 size_t hex_encode(char* buffer, size_t buflen, | |
| 431 const char* csource, size_t srclen) { | |
| 432 return hex_encode_with_delimiter(buffer, buflen, csource, srclen, 0); | |
| 433 } | |
| 434 | |
| 435 size_t hex_encode_with_delimiter(char* buffer, size_t buflen, | |
| 436 const char* csource, size_t srclen, | |
| 437 char delimiter) { | |
| 438 RTC_DCHECK(buffer); // TODO(grunell): estimate output size | |
| 439 if (buflen == 0) | |
| 440 return 0; | |
| 441 | |
| 442 // Init and check bounds. | |
| 443 const unsigned char* bsource = | |
| 444 reinterpret_cast<const unsigned char*>(csource); | |
| 445 size_t srcpos = 0, bufpos = 0; | |
| 446 size_t needed = delimiter ? (srclen * 3) : (srclen * 2 + 1); | |
| 447 if (buflen < needed) | |
| 448 return 0; | |
| 449 | |
| 450 while (srcpos < srclen) { | |
| 451 unsigned char ch = bsource[srcpos++]; | |
| 452 buffer[bufpos ] = hex_encode((ch >> 4) & 0xF); | |
| 453 buffer[bufpos+1] = hex_encode((ch ) & 0xF); | |
| 454 bufpos += 2; | |
| 455 | |
| 456 // Don't write a delimiter after the last byte. | |
| 457 if (delimiter && (srcpos < srclen)) { | |
| 458 buffer[bufpos] = delimiter; | |
| 459 ++bufpos; | |
| 460 } | |
| 461 } | |
| 462 | |
| 463 // Null terminate. | |
| 464 buffer[bufpos] = '\0'; | |
| 465 return bufpos; | |
| 466 } | |
| 467 | |
| 468 std::string hex_encode(const std::string& str) { | |
| 469 return hex_encode(str.c_str(), str.size()); | |
| 470 } | |
| 471 | |
| 472 std::string hex_encode(const char* source, size_t srclen) { | |
| 473 return hex_encode_with_delimiter(source, srclen, 0); | |
| 474 } | |
| 475 | |
| 476 std::string hex_encode_with_delimiter(const char* source, size_t srclen, | |
| 477 char delimiter) { | |
| 478 const size_t kBufferSize = srclen * 3; | |
| 479 char* buffer = STACK_ARRAY(char, kBufferSize); | |
| 480 size_t length = hex_encode_with_delimiter(buffer, kBufferSize, | |
| 481 source, srclen, delimiter); | |
| 482 RTC_DCHECK(srclen == 0 || length > 0); | |
| 483 return std::string(buffer, length); | |
| 484 } | |
| 485 | |
| 486 size_t hex_decode(char * cbuffer, size_t buflen, | |
| 487 const char * source, size_t srclen) { | |
| 488 return hex_decode_with_delimiter(cbuffer, buflen, source, srclen, 0); | |
| 489 } | |
| 490 | |
| 491 size_t hex_decode_with_delimiter(char* cbuffer, size_t buflen, | |
| 492 const char* source, size_t srclen, | |
| 493 char delimiter) { | |
| 494 RTC_DCHECK(cbuffer); // TODO(grunell): estimate output size | |
| 495 if (buflen == 0) | |
| 496 return 0; | |
| 497 | |
| 498 // Init and bounds check. | |
| 499 unsigned char* bbuffer = reinterpret_cast<unsigned char*>(cbuffer); | |
| 500 size_t srcpos = 0, bufpos = 0; | |
| 501 size_t needed = (delimiter) ? (srclen + 1) / 3 : srclen / 2; | |
| 502 if (buflen < needed) | |
| 503 return 0; | |
| 504 | |
| 505 while (srcpos < srclen) { | |
| 506 if ((srclen - srcpos) < 2) { | |
| 507 // This means we have an odd number of bytes. | |
| 508 return 0; | |
| 509 } | |
| 510 | |
| 511 unsigned char h1, h2; | |
| 512 if (!hex_decode(source[srcpos], &h1) || | |
| 513 !hex_decode(source[srcpos + 1], &h2)) | |
| 514 return 0; | |
| 515 | |
| 516 bbuffer[bufpos++] = (h1 << 4) | h2; | |
| 517 srcpos += 2; | |
| 518 | |
| 519 // Remove the delimiter if needed. | |
| 520 if (delimiter && (srclen - srcpos) > 1) { | |
| 521 if (source[srcpos] != delimiter) | |
| 522 return 0; | |
| 523 ++srcpos; | |
| 524 } | |
| 525 } | |
| 526 | |
| 527 return bufpos; | |
| 528 } | |
| 529 | |
| 530 size_t hex_decode(char* buffer, size_t buflen, const std::string& source) { | |
| 531 return hex_decode_with_delimiter(buffer, buflen, source, 0); | |
| 532 } | |
| 533 size_t hex_decode_with_delimiter(char* buffer, size_t buflen, | |
| 534 const std::string& source, char delimiter) { | |
| 535 return hex_decode_with_delimiter(buffer, buflen, | |
| 536 source.c_str(), source.length(), delimiter); | |
| 537 } | |
| 538 | |
| 539 size_t transform(std::string& value, size_t maxlen, const std::string& source, | |
| 540 Transform t) { | |
| 541 char* buffer = STACK_ARRAY(char, maxlen + 1); | |
| 542 size_t length = t(buffer, maxlen + 1, source.data(), source.length()); | |
| 543 value.assign(buffer, length); | |
| 544 return length; | |
| 545 } | |
| 546 | |
| 547 std::string s_transform(const std::string& source, Transform t) { | |
| 548 // Ask transformation function to approximate the destination size (returns up
per bound) | |
| 549 size_t maxlen = t(nullptr, 0, source.data(), source.length()); | |
| 550 char * buffer = STACK_ARRAY(char, maxlen); | |
| 551 size_t len = t(buffer, maxlen, source.data(), source.length()); | |
| 552 std::string result(buffer, len); | |
| 553 return result; | |
| 554 } | |
| 555 | |
| 556 size_t tokenize(const std::string& source, char delimiter, | |
| 557 std::vector<std::string>* fields) { | |
| 558 fields->clear(); | |
| 559 size_t last = 0; | |
| 560 for (size_t i = 0; i < source.length(); ++i) { | |
| 561 if (source[i] == delimiter) { | |
| 562 if (i != last) { | |
| 563 fields->push_back(source.substr(last, i - last)); | |
| 564 } | |
| 565 last = i + 1; | |
| 566 } | |
| 567 } | |
| 568 if (last != source.length()) { | |
| 569 fields->push_back(source.substr(last, source.length() - last)); | |
| 570 } | |
| 571 return fields->size(); | |
| 572 } | |
| 573 | |
| 574 size_t tokenize_with_empty_tokens(const std::string& source, | |
| 575 char delimiter, | |
| 576 std::vector<std::string>* fields) { | |
| 577 fields->clear(); | |
| 578 size_t last = 0; | |
| 579 for (size_t i = 0; i < source.length(); ++i) { | |
| 580 if (source[i] == delimiter) { | |
| 581 fields->push_back(source.substr(last, i - last)); | |
| 582 last = i + 1; | |
| 583 } | |
| 584 } | |
| 585 fields->push_back(source.substr(last, source.length() - last)); | |
| 586 return fields->size(); | |
| 587 } | |
| 588 | |
| 589 size_t tokenize_append(const std::string& source, char delimiter, | |
| 590 std::vector<std::string>* fields) { | |
| 591 if (!fields) return 0; | |
| 592 | |
| 593 std::vector<std::string> new_fields; | |
| 594 tokenize(source, delimiter, &new_fields); | |
| 595 fields->insert(fields->end(), new_fields.begin(), new_fields.end()); | |
| 596 return fields->size(); | |
| 597 } | |
| 598 | |
| 599 size_t tokenize(const std::string& source, char delimiter, char start_mark, | |
| 600 char end_mark, std::vector<std::string>* fields) { | |
| 601 if (!fields) return 0; | |
| 602 fields->clear(); | |
| 603 | |
| 604 std::string remain_source = source; | |
| 605 while (!remain_source.empty()) { | |
| 606 size_t start_pos = remain_source.find(start_mark); | |
| 607 if (std::string::npos == start_pos) break; | |
| 608 std::string pre_mark; | |
| 609 if (start_pos > 0) { | |
| 610 pre_mark = remain_source.substr(0, start_pos - 1); | |
| 611 } | |
| 612 | |
| 613 ++start_pos; | |
| 614 size_t end_pos = remain_source.find(end_mark, start_pos); | |
| 615 if (std::string::npos == end_pos) break; | |
| 616 | |
| 617 // We have found the matching marks. First tokenize the pre-mask. Then add | |
| 618 // the marked part as a single field. Finally, loop back for the post-mark. | |
| 619 tokenize_append(pre_mark, delimiter, fields); | |
| 620 fields->push_back(remain_source.substr(start_pos, end_pos - start_pos)); | |
| 621 remain_source = remain_source.substr(end_pos + 1); | |
| 622 } | |
| 623 | |
| 624 return tokenize_append(remain_source, delimiter, fields); | |
| 625 } | |
| 626 | |
| 627 bool tokenize_first(const std::string& source, | |
| 628 const char delimiter, | |
| 629 std::string* token, | |
| 630 std::string* rest) { | |
| 631 // Find the first delimiter | |
| 632 size_t left_pos = source.find(delimiter); | |
| 633 if (left_pos == std::string::npos) { | |
| 634 return false; | |
| 635 } | |
| 636 | |
| 637 // Look for additional occurrances of delimiter. | |
| 638 size_t right_pos = left_pos + 1; | |
| 639 while (source[right_pos] == delimiter) { | |
| 640 right_pos++; | |
| 641 } | |
| 642 | |
| 643 *token = source.substr(0, left_pos); | |
| 644 *rest = source.substr(right_pos); | |
| 645 return true; | |
| 646 } | |
| 647 | |
| 648 size_t split(const std::string& source, char delimiter, | |
| 649 std::vector<std::string>* fields) { | |
| 650 RTC_DCHECK(fields); | |
| 651 fields->clear(); | |
| 652 size_t last = 0; | |
| 653 for (size_t i = 0; i < source.length(); ++i) { | |
| 654 if (source[i] == delimiter) { | |
| 655 fields->push_back(source.substr(last, i - last)); | |
| 656 last = i + 1; | |
| 657 } | |
| 658 } | |
| 659 fields->push_back(source.substr(last, source.length() - last)); | |
| 660 return fields->size(); | |
| 661 } | |
| 662 | |
| 663 char make_char_safe_for_filename(char c) { | |
| 664 if (c < 32) | |
| 665 return '_'; | |
| 666 | |
| 667 switch (c) { | |
| 668 case '<': | |
| 669 case '>': | |
| 670 case ':': | |
| 671 case '"': | |
| 672 case '/': | |
| 673 case '\\': | |
| 674 case '|': | |
| 675 case '*': | |
| 676 case '?': | |
| 677 return '_'; | |
| 678 | |
| 679 default: | |
| 680 return c; | |
| 681 } | |
| 682 } | |
| 683 | |
| 684 /* | |
| 685 void sprintf(std::string& value, size_t maxlen, const char * format, ...) { | |
| 686 char * buffer = STACK_ARRAY(char, maxlen + 1); | |
| 687 va_list args; | |
| 688 va_start(args, format); | |
| 689 value.assign(buffer, vsprintfn(buffer, maxlen + 1, format, args)); | |
| 690 va_end(args); | |
| 691 } | |
| 692 */ | |
| 693 | |
| 694 ///////////////////////////////////////////////////////////////////////////// | |
| 695 | |
| 696 } // namespace rtc | |
| OLD | NEW |
|---|
Chromium Code Reviews
Issue 2877023002:
Move webrtc/{base => rtc_base} (Closed)
