diff options
author | Steve Block <steveblock@google.com> | 2011-01-05 12:15:11 +0000 |
---|---|---|
committer | Steve Block <steveblock@google.com> | 2011-01-06 14:14:00 +0000 |
commit | d06194330da2bb8da887d2e1adeacb3a5c1504b2 (patch) | |
tree | e0af8413af65a8e30630563441af7bdb8478e513 /JavaScriptCore/wtf/unicode/UTF8.cpp | |
parent | 419a5cf2f8db6ca014df624865197ffb82caad37 (diff) | |
download | external_webkit-d06194330da2bb8da887d2e1adeacb3a5c1504b2.zip external_webkit-d06194330da2bb8da887d2e1adeacb3a5c1504b2.tar.gz external_webkit-d06194330da2bb8da887d2e1adeacb3a5c1504b2.tar.bz2 |
Merge WebKit at r72805: Initial merge by Git
Note that this is a backwards merge from Chromium release 9.0.600.0
to 9.0.597.0, to align with the Chromium 9 stable release branch.
Change-Id: I5d2bb4e8cee9d39ae8485abf48bdb55ecf8b3790
Diffstat (limited to 'JavaScriptCore/wtf/unicode/UTF8.cpp')
-rw-r--r-- | JavaScriptCore/wtf/unicode/UTF8.cpp | 90 |
1 files changed, 38 insertions, 52 deletions
diff --git a/JavaScriptCore/wtf/unicode/UTF8.cpp b/JavaScriptCore/wtf/unicode/UTF8.cpp index ca4fc1c..40c5609 100644 --- a/JavaScriptCore/wtf/unicode/UTF8.cpp +++ b/JavaScriptCore/wtf/unicode/UTF8.cpp @@ -26,14 +26,9 @@ #include "config.h" #include "UTF8.h" -#include "ASCIICType.h" - namespace WTF { namespace Unicode { -// FIXME: Use definition from CharacterNames.h. -const UChar replacementCharacter = 0xFFFD; - inline int inlineUTF8SequenceLengthNonASCII(char b0) { if ((b0 & 0xC0) != 0xC0) @@ -49,12 +44,12 @@ inline int inlineUTF8SequenceLengthNonASCII(char b0) inline int inlineUTF8SequenceLength(char b0) { - return isASCII(b0) ? 1 : inlineUTF8SequenceLengthNonASCII(b0); + return (b0 & 0x80) == 0 ? 1 : inlineUTF8SequenceLengthNonASCII(b0); } int UTF8SequenceLength(char b0) { - return isASCII(b0) ? 1 : inlineUTF8SequenceLengthNonASCII(b0); + return (b0 & 0x80) == 0 ? 1 : inlineUTF8SequenceLengthNonASCII(b0); } int decodeUTF8Sequence(const char* sequence) @@ -177,7 +172,7 @@ ConversionResult convertUTF16ToUTF8( bytesToWrite = 4; } else { bytesToWrite = 3; - ch = replacementCharacter; + ch = 0xFFFD; } target += bytesToWrite; @@ -236,23 +231,6 @@ static bool isLegalUTF8(const unsigned char* source, int length) static const UChar32 offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL, 0x03C82080UL, 0xFA082080UL, 0x82082080UL }; -static inline UChar32 readUTF8Sequence(const char*& sequence, unsigned length) -{ - UChar32 character = 0; - - // The cases all fall through. - switch (length) { - case 6: character += static_cast<unsigned char>(*sequence++); character <<= 6; - case 5: character += static_cast<unsigned char>(*sequence++); character <<= 6; - case 4: character += static_cast<unsigned char>(*sequence++); character <<= 6; - case 3: character += static_cast<unsigned char>(*sequence++); character <<= 6; - case 2: character += static_cast<unsigned char>(*sequence++); character <<= 6; - case 1: character += static_cast<unsigned char>(*sequence++); - } - - return character - offsetsFromUTF8[length - 1]; -} - ConversionResult convertUTF8ToUTF16( const char** sourceStart, const char* sourceEnd, UChar** targetStart, UChar* targetEnd, bool strict) @@ -261,52 +239,60 @@ ConversionResult convertUTF8ToUTF16( const char* source = *sourceStart; UChar* target = *targetStart; while (source < sourceEnd) { - int utf8SequenceLength = inlineUTF8SequenceLength(*source); - if (sourceEnd - source < utf8SequenceLength) { + UChar32 ch = 0; + int extraBytesToRead = inlineUTF8SequenceLength(*source) - 1; + if (source + extraBytesToRead >= sourceEnd) { result = sourceExhausted; break; } // Do this check whether lenient or strict - if (!isLegalUTF8(reinterpret_cast<const unsigned char*>(source), utf8SequenceLength)) { + if (!isLegalUTF8(reinterpret_cast<const unsigned char*>(source), extraBytesToRead + 1)) { result = sourceIllegal; break; } - - UChar32 character = readUTF8Sequence(source, utf8SequenceLength); + // The cases all fall through. + switch (extraBytesToRead) { + case 5: ch += static_cast<unsigned char>(*source++); ch <<= 6; // remember, illegal UTF-8 + case 4: ch += static_cast<unsigned char>(*source++); ch <<= 6; // remember, illegal UTF-8 + case 3: ch += static_cast<unsigned char>(*source++); ch <<= 6; + case 2: ch += static_cast<unsigned char>(*source++); ch <<= 6; + case 1: ch += static_cast<unsigned char>(*source++); ch <<= 6; + case 0: ch += static_cast<unsigned char>(*source++); + } + ch -= offsetsFromUTF8[extraBytesToRead]; if (target >= targetEnd) { - source -= utf8SequenceLength; // Back up source pointer! - result = targetExhausted; - break; + source -= (extraBytesToRead + 1); // Back up source pointer! + result = targetExhausted; break; } - - if (U_IS_BMP(character)) { + if (ch <= 0xFFFF) { // UTF-16 surrogate values are illegal in UTF-32 - if (U_IS_SURROGATE(character)) { + if (ch >= 0xD800 && ch <= 0xDFFF) { if (strict) { - source -= utf8SequenceLength; // return to the illegal value itself + source -= (extraBytesToRead + 1); // return to the illegal value itself result = sourceIllegal; break; } else - *target++ = replacementCharacter; + *target++ = 0xFFFD; } else - *target++ = character; // normal case - } else if (U_IS_SUPPLEMENTARY(character)) { + *target++ = (UChar)ch; // normal case + } else if (ch > 0x10FFFF) { + if (strict) { + result = sourceIllegal; + source -= (extraBytesToRead + 1); // return to the start + break; // Bail out; shouldn't continue + } else + *target++ = 0xFFFD; + } else { // target is a character in range 0xFFFF - 0x10FFFF if (target + 1 >= targetEnd) { - source -= utf8SequenceLength; // Back up source pointer! + source -= (extraBytesToRead + 1); // Back up source pointer! result = targetExhausted; break; } - *target++ = U16_LEAD(character); - *target++ = U16_TRAIL(character); - } else { - if (strict) { - source -= utf8SequenceLength; // return to the start - result = sourceIllegal; - break; // Bail out; shouldn't continue - } else - *target++ = replacementCharacter; + ch -= 0x0010000UL; + *target++ = (UChar)((ch >> 10) + 0xD800); + *target++ = (UChar)((ch & 0x03FF) + 0xDC00); } } *sourceStart = source; @@ -314,5 +300,5 @@ ConversionResult convertUTF8ToUTF16( return result; } -} // namespace Unicode -} // namespace WTF +} +} |