diff options
| author | Marc Mutz <marc.mutz@qt.io> | 2025-10-21 18:10:31 +0200 |
|---|---|---|
| committer | Marc Mutz <marc.mutz@qt.io> | 2025-10-30 22:56:25 +0000 |
| commit | e08bcaaa87d7832c158e631cb6df2aecd71fe62d (patch) | |
| tree | b4e4e50b4af63384de464d03cc46cc9ad6294dbb /src/corelib/text/qunicodetools.cpp | |
| parent | 9e442a8e27000c7b925aa54a5a4c59e5ca2badb7 (diff) | |
QUnicodeTools: port getSentenceBreaks() to QStringIterator
Like getWordBreaks(), this one is a bit more complicated than the
first two, since there's a nested loop. Solve it by using a copy of
the QStringIterator for look-ahead loop.
To see that old and new version are equivalent, observe that qsizetype
`i` and `lookahead` always pointed _onto_, while QStringIterator
always points to just _after_ the last-consumed code-unit.
Pick-to: 6.10 6.8 6.5
Change-Id: Id272b1a1597912eb611acb544b5ef0ac1d13a754
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Diffstat (limited to 'src/corelib/text/qunicodetools.cpp')
| -rw-r--r-- | src/corelib/text/qunicodetools.cpp | 28 |
1 files changed, 8 insertions, 20 deletions
diff --git a/src/corelib/text/qunicodetools.cpp b/src/corelib/text/qunicodetools.cpp index dd751709d8a..56fa41c51ab 100644 --- a/src/corelib/text/qunicodetools.cpp +++ b/src/corelib/text/qunicodetools.cpp @@ -388,16 +388,11 @@ static const uchar breakTable[BAfter + 1][QUnicodeTables::NumSentenceBreakClasse static void getSentenceBreaks(const char16_t *string, qsizetype len, QCharAttributes *attributes) { uchar state = SB::BAfter; // to meet SB1 - for (qsizetype i = 0; i != len; ++i) { - const qsizetype pos = i; - char32_t ucs4 = string[i]; - if (QChar::isHighSurrogate(ucs4) && i + 1 != len) { - ushort low = string[i + 1]; - if (QChar::isLowSurrogate(low)) { - ucs4 = QChar::surrogateToUcs4(ucs4, low); - ++i; - } - } + + QStringIterator it(QStringView{string, len}); + while (it.hasNext()) { + const qsizetype pos = it.index(); + const char32_t ucs4 = it.nextOrRawCodeUnit(); const auto prop = QUnicodeTables::properties(ucs4); QUnicodeTables::SentenceBreakClass ncls = (QUnicodeTables::SentenceBreakClass) prop->sentenceBreakClass; @@ -406,15 +401,8 @@ static void getSentenceBreaks(const char16_t *string, qsizetype len, QCharAttrib state = SB::breakTable[state][ncls]; if (Q_UNLIKELY(state == SB::Lookup)) { // SB8 state = SB::Break; - for (qsizetype lookahead = i + 1; lookahead < len; ++lookahead) { - char32_t ucs4 = string[lookahead]; - if (QChar::isHighSurrogate(ucs4) && lookahead + 1 != len) { - ushort low = string[lookahead + 1]; - if (QChar::isLowSurrogate(low)) { - ucs4 = QChar::surrogateToUcs4(ucs4, low); - ++lookahead; - } - } + for (auto lookahead = it; lookahead.hasNext(); /**/) { + const char32_t ucs4 = lookahead.nextOrRawCodeUnit(); const auto prop = QUnicodeTables::properties(ucs4); QUnicodeTables::SentenceBreakClass tcls = (QUnicodeTables::SentenceBreakClass) prop->sentenceBreakClass; @@ -427,7 +415,7 @@ static void getSentenceBreaks(const char16_t *string, qsizetype len, QCharAttrib case QUnicodeTables::SentenceBreak_Close: continue; case QUnicodeTables::SentenceBreak_Lower: - i = lookahead; + it = lookahead; state = SB::Initial; break; default: |