MCPcopy Create free account
hub / github.com/antmachineintelligence/mtgbmcode / parse_string

Method parse_string

src/io/json11.cpp:477–564  ·  view source on GitHub ↗

parse_string() * * Parse a string, starting at the current position. */

Source from the content-addressed store, hash-verified

475 * Parse a string, starting at the current position.
476 */
477 string parse_string() {
478 string out;
479 long last_escaped_codepoint = -1;
480 while (true) {
481 if (i == str.size())
482 return fail("Unexpected end of input in string", "");
483
484 char ch = str[i++];
485
486 if (ch == '"') {
487 encode_utf8(last_escaped_codepoint, out);
488 return out;
489 }
490
491 if (in_range(ch, 0, 0x1f))
492 return fail("Unescaped " + esc(ch) + " in string", "");
493
494 // The usual case: non-escaped characters
495 if (ch != '\\') {
496 encode_utf8(last_escaped_codepoint, out);
497 last_escaped_codepoint = -1;
498 out += ch;
499 continue;
500 }
501
502 // Handle escapes
503 if (i == str.size())
504 return fail("Unexpected end of input in string", "");
505
506 ch = str[i++];
507
508 if (ch == 'u') {
509 // Extract 4-byte escape sequence
510 string esc = str.substr(i, 4);
511 // Explicitly check length of the substring. The following loop
512 // relies on std::string returning the terminating NUL when
513 // accessing str[length]. Checking here reduces brittleness.
514 if (esc.length() < 4) {
515 return fail("Bad \\u escape: " + esc, "");
516 }
517 for (size_t j = 0; j < 4; j++) {
518 if (!in_range(esc[j], 'a', 'f') && !in_range(esc[j], 'A', 'F')
519 && !in_range(esc[j], '0', '9'))
520 return fail("Bad \\u escape: " + esc, "");
521 }
522
523 long codepoint = strtol(esc.data(), nullptr, 16);
524
525 // JSON specifies that characters outside the BMP shall be encoded as a pair
526 // of 4-hex-digit \u escapes encoding their surrogate pair components. Check
527 // whether we're in the middle of such a beast: the previous codepoint was an
528 // escaped lead (high) surrogate, and this is a trail (low) surrogate.
529 if (in_range(last_escaped_codepoint, 0xD800, 0xDBFF)
530 && in_range(codepoint, 0xDC00, 0xDFFF)) {
531 // Reassemble the two surrogate pairs into one astral-plane character, per
532 // the UTF-16 algorithm.
533 encode_utf8((((last_escaped_codepoint - 0xD800) << 10)
534 | (codepoint - 0xDC00)) + 0x10000, out);

Callers

nothing calls this directly

Calls 6

substrMethod · 0.80
lengthMethod · 0.80
dataMethod · 0.80
in_rangeFunction · 0.70
escFunction · 0.70
sizeMethod · 0.45

Tested by

no test coverage detected