comparison: src/json.c
src/json.c
- changeset 1149
- df5665de7344
- parent 1139
- 7dfa5bcf39ee
- child 1150
- 7b0bd5e76b5d
equal
deleted
inserted
replaced
| 1139:7dfa5bcf39ee | 1149:df5665de7344 |
|---|---|
| 351 } | 351 } |
| 352 | 352 |
| 353 return CX_JSON_INCOMPLETE_DATA; | 353 return CX_JSON_INCOMPLETE_DATA; |
| 354 } | 354 } |
| 355 | 355 |
| 356 static int codepoint_to_utf8(uint32_t codepoint, char *output_buf) { | |
| 357 if (codepoint <= 0x7F) { | |
| 358 *output_buf = (char)codepoint; | |
| 359 return 1; | |
| 360 } else if (codepoint <= 0x7FF) { | |
| 361 output_buf[0] = (char)(0xC0 | ((codepoint >> 6) & 0x1F)); | |
| 362 output_buf[1] = (char)(0x80 | (codepoint & 0x3F)); | |
| 363 return 2; | |
| 364 } else if (codepoint <= 0xFFFF) { | |
| 365 output_buf[0] = (char)(0xE0 | ((codepoint >> 12) & 0x0F)); | |
| 366 output_buf[1] = (char)(0x80 | ((codepoint >> 6) & 0x3F)); | |
| 367 output_buf[2] = (char)(0x80 | (codepoint & 0x3F)); | |
| 368 return 3; | |
| 369 } | |
| 370 | |
| 371 return 0; | |
| 372 } | |
| 373 | |
| 356 static cxmutstr unescape_string(const CxAllocator *a, cxmutstr str) { | 374 static cxmutstr unescape_string(const CxAllocator *a, cxmutstr str) { |
| 357 // note: this function expects that str contains the enclosing quotes! | 375 // note: this function expects that str contains the enclosing quotes! |
| 358 | 376 |
| 359 cxmutstr result; | 377 cxmutstr result; |
| 360 result.length = 0; | 378 result.length = 0; |
| 378 c = '/'; // always unescape, we don't need settings here | 396 c = '/'; // always unescape, we don't need settings here |
| 379 } else if (c == 'f') { | 397 } else if (c == 'f') { |
| 380 c = '\f'; | 398 c = '\f'; |
| 381 } else if (c == 'b') { | 399 } else if (c == 'b') { |
| 382 c = '\b'; | 400 c = '\b'; |
| 383 } | 401 } else if (c == 'u') { |
| 384 // TODO: support \uXXXX escape sequences | 402 if (i+4 < str.length) { |
| 403 cxstring codepoint_str = { str.ptr + i + 1, 4}; | |
| 404 uint32_t codepoint; | |
| 405 if(!cx_strtou32_lc_(codepoint_str, &codepoint, 16, "")) { | |
| 406 char utf8buf[4]; | |
| 407 int utf8len = codepoint_to_utf8(codepoint, utf8buf); | |
| 408 if(utf8len > 0) { | |
| 409 // add all bytes from utf8buf expect the last char | |
| 410 // to the result | |
| 411 utf8len--; | |
| 412 c = utf8buf[utf8len]; | |
| 413 for(int i=0;i<utf8len;i++) { | |
| 414 result.ptr[result.length++] = utf8buf[i]; | |
| 415 } | |
| 416 } | |
| 417 i += 4; | |
| 418 } | |
| 419 } | |
| 420 } | |
| 421 | |
| 385 // TODO: discuss the behavior for unrecognized escape sequences | 422 // TODO: discuss the behavior for unrecognized escape sequences |
| 386 // most parsers throw an error here | 423 // most parsers throw an error here |
| 387 result.ptr[result.length++] = c; | 424 result.ptr[result.length++] = c; |
| 388 } else { | 425 } else { |
| 389 if (c == '\\') { | 426 if (c == '\\') { |
