languages/dart-sdk
languages/dart-sdk
1
0
Fork 0
mirror of https://github.com/dart-lang/sdk synced 2024-09-21 10:46:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Improve utf8 decoding of single-char bytes, by isolating the loop.

Browse source 
BUG=
R=fschneider@google.com

Review URL: https://codereview.chromium.org//435553002

git-svn-id: https://dart.googlecode.com/svn/branches/bleeding_edge/dart@38776 260f80e4-7a28-3924-810f-c04153c831b5
This commit is contained in:
ajohnsen@google.com 2014-07-31 12:23:11 +00:00
parent 8118611087
commit ce13411e56
1 changed files with 17 additions and 17 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

34
sdk/lib/convert/utf.dart
View file

@ -420,13 +420,20 @@ class _Utf8Decoder {
int value = _value;
int expectedUnits = _expectedUnits;
int extraUnits = _extraUnits;
int singleBytesCount = 0;
_value = 0;
_expectedUnits = 0;
_extraUnits = 0;
int scanOneByteCharacters(units, int from) {
final to = endIndex;
final mask = ~_ONE_BYTE_LIMIT;
for (var i = from; i < to; i++) {
if ((units[i] & mask) != 0) return i - from;
}
return to - from;
}
void addSingleBytes(int from, int to) {
assert(singleBytesCount > 0);
assert(from >= startIndex && from <= endIndex);
assert(to >= startIndex && to <= endIndex);
if (from == 0 && to == codeUnits.length) {
@ -435,7 +442,6 @@ class _Utf8Decoder {
_stringSink.write(
new String.fromCharCodes(codeUnits.sublist(from, to)));
}
singleBytesCount = 0;
}
int i = startIndex;
@ -485,6 +491,13 @@ class _Utf8Decoder {
}
while (i < endIndex) {
int oneBytes = scanOneByteCharacters(codeUnits, i);
if (oneBytes > 0) {
_isFirstCharacter = false;
addSingleBytes(i, i + oneBytes);
i += oneBytes;
if (i == endIndex) break;
}
int unit = codeUnits[i++];
// TODO(floitsch): the way we test we could potentially allow
// units that are too large, if they happen to have the
@ -493,23 +506,13 @@ class _Utf8Decoder {
// https://codereview.chromium.org/22929022/diff/1/sdk/lib/convert/utf.dart?column_width=80
if (unit < 0) {
// TODO(floitsch): should this be unit <= 0 ?
if (singleBytesCount > 0) {
int to = i - 1;
addSingleBytes(to - singleBytesCount, to);
}
if (!_allowMalformed) {
throw new FormatException(
"Negative UTF-8 code unit: -0x${(-unit).toRadixString(16)}");
}
_stringSink.writeCharCode(UNICODE_REPLACEMENT_CHARACTER_RUNE);
} else if (unit <= _ONE_BYTE_LIMIT) {
_isFirstCharacter = false;
singleBytesCount++;
} else {
if (singleBytesCount > 0) {
int to = i - 1;
addSingleBytes(to - singleBytesCount, to);
}
assert(unit > _ONE_BYTE_LIMIT);
if ((unit & 0xE0) == 0xC0) {
value = unit & 0x1F;
expectedUnits = extraUnits = 1;
@ -538,9 +541,6 @@ class _Utf8Decoder {
}
break loop;
}
if (singleBytesCount > 0) {
addSingleBytes(i - singleBytesCount, endIndex);
}
if (expectedUnits > 0) {
_value = value;
_expectedUnits = expectedUnits;