OLD | NEW |
---|---|
1 // Copyright 2016 The Chromium Authors. All rights reserved. | 1 // Copyright 2016 The Chromium Authors. All rights reserved. |
2 // Use of this source code is governed by a BSD-style license that can be | 2 // Use of this source code is governed by a BSD-style license that can be |
3 // found in the LICENSE file. | 3 // found in the LICENSE file. |
4 | 4 |
5 #include "platform/inspector_protocol/InspectorProtocol.h" | 5 #include "platform/v8_inspector/String16.h" |
6 | 6 |
7 #include <algorithm> | 7 #include <algorithm> |
8 #include <cctype> | 8 #include <cctype> |
9 #include <cstdio> | 9 #include <cstdio> |
10 #include <cstdlib> | |
11 #include <cstring> | |
10 #include <locale> | 12 #include <locale> |
13 #include <string> | |
11 | 14 |
12 namespace blink { | 15 namespace v8_inspector { |
13 namespace protocol { | 16 |
17 namespace { | |
18 | |
19 bool isASCII(UChar c) | |
20 { | |
21 return !(c & ~0x7F); | |
22 } | |
23 | |
24 bool isSpaceOrNewLine(UChar c) | |
25 { | |
26 return isASCII(c) && c <= ' ' && (c == ' ' || (c <= 0xD && c >= 0x9)); | |
27 } | |
28 | |
29 double charactersToDouble(const UChar* characters, size_t length, bool* ok = nul lptr) | |
30 { | |
31 std::vector<char> buffer; | |
32 buffer.reserve(length + 1); | |
33 for (size_t i = 0; i < length; ++i) { | |
34 if (!isASCII(characters[i])) { | |
35 if (ok) | |
36 *ok = false; | |
37 return 0; | |
38 } | |
39 buffer.push_back(static_cast<char>(characters[i])); | |
40 } | |
41 buffer.push_back('\0'); | |
42 | |
43 char* endptr; | |
44 double result = std::strtod(buffer.data(), &endptr); | |
45 if (ok) | |
46 *ok = !(*endptr); | |
47 return result; | |
48 } | |
49 | |
50 int charactersToInteger(const UChar* characters, size_t length, bool* ok = nullp tr) | |
51 { | |
52 std::vector<char> buffer; | |
53 buffer.reserve(length + 1); | |
54 for (size_t i = 0; i < length; ++i) { | |
55 if (!isASCII(characters[i])) { | |
56 if (ok) | |
57 *ok = false; | |
58 return 0; | |
59 } | |
60 buffer.push_back(static_cast<char>(characters[i])); | |
61 } | |
62 buffer.push_back('\0'); | |
63 | |
64 char* endptr; | |
65 int result = std::strtol(buffer.data(), &endptr, 10); | |
66 if (ok) | |
67 *ok = !(*endptr); | |
68 return result; | |
69 } | |
14 | 70 |
15 const UChar replacementCharacter = 0xFFFD; | 71 const UChar replacementCharacter = 0xFFFD; |
16 using UChar32 = uint32_t; | 72 using UChar32 = uint32_t; |
17 | 73 |
18 inline int inlineUTF8SequenceLengthNonASCII(char b0) | 74 inline int inlineUTF8SequenceLengthNonASCII(char b0) |
19 { | 75 { |
20 if ((b0 & 0xC0) != 0xC0) | 76 if ((b0 & 0xC0) != 0xC0) |
21 return 0; | 77 return 0; |
22 if ((b0 & 0xE0) == 0xC0) | 78 if ((b0 & 0xE0) == 0xC0) |
23 return 2; | 79 return 2; |
24 if ((b0 & 0xF0) == 0xE0) | 80 if ((b0 & 0xF0) == 0xE0) |
25 return 3; | 81 return 3; |
26 if ((b0 & 0xF8) == 0xF0) | 82 if ((b0 & 0xF8) == 0xF0) |
27 return 4; | 83 return 4; |
28 return 0; | 84 return 0; |
29 } | 85 } |
30 | 86 |
31 inline int inlineUTF8SequenceLength(char b0) | 87 inline int inlineUTF8SequenceLength(char b0) |
32 { | 88 { |
33 return String16::isASCII(b0) ? 1 : inlineUTF8SequenceLengthNonASCII(b0); | 89 return isASCII(b0) ? 1 : inlineUTF8SequenceLengthNonASCII(b0); |
34 } | 90 } |
35 | 91 |
36 // Once the bits are split out into bytes of UTF-8, this is a mask OR-ed | 92 // Once the bits are split out into bytes of UTF-8, this is a mask OR-ed |
37 // into the first byte, depending on how many bytes follow. There are | 93 // into the first byte, depending on how many bytes follow. There are |
38 // as many entries in this table as there are UTF-8 sequence types. | 94 // as many entries in this table as there are UTF-8 sequence types. |
39 // (I.e., one byte sequence, two byte... etc.). Remember that sequences | 95 // (I.e., one byte sequence, two byte... etc.). Remember that sequences |
40 // for *legal* UTF-8 will be 4 or fewer bytes total. | 96 // for *legal* UTF-8 will be 4 or fewer bytes total. |
41 static const unsigned char firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0x F8, 0xFC }; | 97 static const unsigned char firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0x F8, 0xFC }; |
42 | 98 |
43 typedef enum { | 99 typedef enum { |
(...skipping 281 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
325 | 381 |
326 if (sourceAllASCII) | 382 if (sourceAllASCII) |
327 *sourceAllASCII = !(orAllData & ~0x7f); | 383 *sourceAllASCII = !(orAllData & ~0x7f); |
328 | 384 |
329 return result; | 385 return result; |
330 } | 386 } |
331 | 387 |
332 // Helper to write a three-byte UTF-8 code point to the buffer, caller must chec k room is available. | 388 // Helper to write a three-byte UTF-8 code point to the buffer, caller must chec k room is available. |
333 static inline void putUTF8Triple(char*& buffer, UChar ch) | 389 static inline void putUTF8Triple(char*& buffer, UChar ch) |
334 { | 390 { |
335 DCHECK_GE(ch, 0x0800); | |
336 *buffer++ = static_cast<char>(((ch >> 12) & 0x0F) | 0xE0); | 391 *buffer++ = static_cast<char>(((ch >> 12) & 0x0F) | 0xE0); |
337 *buffer++ = static_cast<char>(((ch >> 6) & 0x3F) | 0x80); | 392 *buffer++ = static_cast<char>(((ch >> 6) & 0x3F) | 0x80); |
338 *buffer++ = static_cast<char>((ch & 0x3F) | 0x80); | 393 *buffer++ = static_cast<char>((ch & 0x3F) | 0x80); |
339 } | 394 } |
340 | 395 |
396 } // namespace | |
397 | |
398 // static | |
399 String16 String16::fromInteger(int number) | |
400 { | |
401 const size_t kBufferSize = 50; | |
402 char buffer[kBufferSize]; | |
403 std::snprintf(buffer, kBufferSize, "%d", number); | |
404 return String16(buffer); | |
405 } | |
406 | |
407 // static | |
408 String16 String16::fromDouble(double number) | |
409 { | |
410 const size_t kBufferSize = 100; | |
caseq
2016/08/19 02:50:36
100 looks like overkill :-)
| |
411 char buffer[kBufferSize]; | |
412 std::snprintf(buffer, kBufferSize, "%f", number); | |
413 return String16(buffer); | |
414 } | |
415 | |
416 // static | |
417 String16 String16::fromDoublePrecision3(double number) | |
418 { | |
419 const size_t kBufferSize = 100; | |
420 char buffer[kBufferSize]; | |
421 std::snprintf(buffer, kBufferSize, "%.3g", number); | |
422 return String16(buffer); | |
423 } | |
424 | |
425 // static | |
426 String16 String16::fromDoublePrecision6(double number) | |
427 { | |
428 const size_t kBufferSize = 100; | |
429 char buffer[kBufferSize]; | |
430 std::snprintf(buffer, kBufferSize, "%.6g", number); | |
431 return String16(buffer); | |
432 } | |
433 | |
434 double String16::toDouble(bool* ok) const | |
435 { | |
436 return charactersToDouble(characters16(), length(), ok); | |
437 } | |
438 | |
439 int String16::toInteger(bool* ok) const | |
440 { | |
441 return charactersToInteger(characters16(), length(), ok); | |
442 } | |
443 | |
444 String16 String16::stripWhiteSpace() const | |
445 { | |
446 if (!length()) | |
447 return String16(); | |
448 | |
449 unsigned start = 0; | |
450 unsigned end = length() - 1; | |
451 | |
452 // skip white space from start | |
453 while (start <= end && isSpaceOrNewLine(characters16()[start])) | |
454 ++start; | |
455 | |
456 // only white space | |
457 if (start > end) | |
458 return String16(); | |
459 | |
460 // skip white space from end | |
461 while (end && isSpaceOrNewLine(characters16()[end])) | |
462 --end; | |
463 | |
464 if (!start && end == length() - 1) | |
465 return *this; | |
466 return String16(characters16() + start, end + 1 - start); | |
467 } | |
468 | |
469 bool String16::startsWith(const char* prefix) const | |
470 { | |
471 for (size_t i = 0, j = 0; prefix[j] && i < length(); ++i, ++j) { | |
472 if (characters16()[i] != prefix[j]) | |
473 return false; | |
474 } | |
475 return true; | |
476 } | |
477 | |
478 String16Builder::String16Builder() | |
479 { | |
480 } | |
481 | |
482 void String16Builder::append(const String16& s) | |
483 { | |
484 m_buffer.insert(m_buffer.end(), s.characters16(), s.characters16() + s.lengt h()); | |
485 } | |
486 | |
487 void String16Builder::append(UChar c) | |
488 { | |
489 m_buffer.push_back(c); | |
490 } | |
491 | |
492 void String16Builder::append(char c) | |
493 { | |
494 UChar u = c; | |
495 m_buffer.push_back(u); | |
496 } | |
497 | |
498 void String16Builder::append(const UChar* characters, size_t length) | |
499 { | |
500 m_buffer.insert(m_buffer.end(), characters, characters + length); | |
501 } | |
502 | |
503 void String16Builder::append(const char* characters, size_t length) | |
504 { | |
505 m_buffer.reserve(m_buffer.size() + length); | |
506 for (size_t i = 0; i < length; ++i, ++characters) { | |
507 UChar u = *characters; | |
508 m_buffer.push_back(u); | |
509 } | |
510 } | |
511 | |
512 String16 String16Builder::toString() | |
513 { | |
514 return String16(m_buffer.data(), m_buffer.size()); | |
515 } | |
516 | |
517 void String16Builder::reserveCapacity(size_t capacity) | |
518 { | |
519 m_buffer.reserve(capacity); | |
520 } | |
521 | |
341 String16 String16::fromUTF8(const char* stringStart, size_t length) | 522 String16 String16::fromUTF8(const char* stringStart, size_t length) |
342 { | 523 { |
343 if (!stringStart || !length) | 524 if (!stringStart || !length) |
344 return String16(); | 525 return String16(); |
345 | 526 |
346 std::vector<UChar> buffer(length); | 527 std::vector<UChar> buffer(length); |
347 UChar* bufferStart = buffer.data(); | 528 UChar* bufferStart = buffer.data(); |
348 | 529 |
349 UChar* bufferCurrent = bufferStart; | 530 UChar* bufferCurrent = bufferStart; |
350 const char* stringCurrent = stringStart; | 531 const char* stringCurrent = stringStart; |
(...skipping 21 matching lines...) Expand all Loading... | |
372 // * We could allocate a CStringBuffer with an appropriate size to | 553 // * We could allocate a CStringBuffer with an appropriate size to |
373 // have a good chance of being able to write the string into the | 554 // have a good chance of being able to write the string into the |
374 // buffer without reallocing (say, 1.5 x length). | 555 // buffer without reallocing (say, 1.5 x length). |
375 if (length > std::numeric_limits<unsigned>::max() / 3) | 556 if (length > std::numeric_limits<unsigned>::max() / 3) |
376 return std::string(); | 557 return std::string(); |
377 std::vector<char> bufferVector(length * 3); | 558 std::vector<char> bufferVector(length * 3); |
378 char* buffer = bufferVector.data(); | 559 char* buffer = bufferVector.data(); |
379 const UChar* characters = m_impl.data(); | 560 const UChar* characters = m_impl.data(); |
380 | 561 |
381 ConversionResult result = convertUTF16ToUTF8(&characters, characters + lengt h, &buffer, buffer + bufferVector.size(), false); | 562 ConversionResult result = convertUTF16ToUTF8(&characters, characters + lengt h, &buffer, buffer + bufferVector.size(), false); |
382 DCHECK(result != targetExhausted); // (length * 3) should be sufficient for any conversion | |
383 | |
384 // Only produced from strict conversion. | |
385 DCHECK(result != sourceIllegal); | |
386 | 563 |
387 // Check for an unconverted high surrogate. | 564 // Check for an unconverted high surrogate. |
388 if (result == sourceExhausted) { | 565 if (result == sourceExhausted) |
389 // This should be one unpaired high surrogate. Treat it the same | |
390 // was as an unpaired high surrogate would have been handled in | |
391 // the middle of a string with non-strict conversion - which is | |
392 // to say, simply encode it to UTF-8. | |
393 DCHECK((characters + 1) == (m_impl.data() + length)); | |
394 DCHECK((*characters >= 0xD800) && (*characters <= 0xDBFF)); | |
395 // There should be room left, since one UChar hasn't been | |
396 // converted. | |
397 DCHECK((buffer + 3) <= (buffer + bufferVector.size())); | |
398 putUTF8Triple(buffer, *characters); | 566 putUTF8Triple(buffer, *characters); |
399 } | |
400 | 567 |
401 return std::string(bufferVector.data(), buffer - bufferVector.data()); | 568 return std::string(bufferVector.data(), buffer - bufferVector.data()); |
402 } | 569 } |
403 | 570 |
404 } // namespace protocol | 571 } // namespace v8_inspector |
405 } // namespace blink | |
OLD | NEW |