comphelper/source/misc/string.cxx | 21 ++++--------- compilerplugins/clang/stringview.cxx | 3 + compilerplugins/clang/stringviewparam.cxx | 4 +- compilerplugins/clang/stringviewvar.cxx | 2 - compilerplugins/clang/test/stringviewparam.cxx | 15 +++++++++ i18npool/inc/cclass_unicode.hxx | 2 - i18npool/source/breakiterator/breakiteratorImpl.cxx | 15 ++++----- i18npool/source/characterclassification/cclass_unicode.cxx | 5 +-- include/comphelper/string.hxx | 2 - include/o3tl/string_view.hxx | 2 - o3tl/qa/test-string_view.cxx | 8 ++-- sal/rtl/ustring.cxx | 5 --- vcl/source/gdi/impglyphitem.cxx | 7 ++-- 13 files changed, 50 insertions(+), 41 deletions(-)
New commits: commit 325c571c22231978e5409902b514d2584186e0a0 Author: Noel Grandin <[email protected]> AuthorDate: Fri May 5 12:07:00 2023 +0200 Commit: Noel Grandin <[email protected]> CommitDate: Sat May 6 12:09:48 2023 +0200 update loplugin:stringview* for o3tl::iterateCodePoints And change o3tl::iterateCodePoints to use sal_Int32 for its second param, to integrate better with places where the parameter comes from an UNO API, which cannot use std::size_t Change-Id: I7b9dd2c9bc7f48e6c4a912f039f1b5dae7beae69 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/151421 Tested-by: Jenkins Reviewed-by: Noel Grandin <[email protected]> diff --git a/comphelper/source/misc/string.cxx b/comphelper/source/misc/string.cxx index 0fdd24c83d7e..d5653522a229 100644 --- a/comphelper/source/misc/string.cxx +++ b/comphelper/source/misc/string.cxx @@ -263,13 +263,12 @@ sal_Int32 getTokenCount(std::u16string_view rIn, sal_Unicode cTok) return tmpl_getTokenCount<std::u16string_view, sal_Unicode>(rIn, cTok); } -static sal_uInt32 decimalStringToNumber( - OUString const & str, sal_Int32 nStart, sal_Int32 nLength ) +sal_uInt32 decimalStringToNumber(std::u16string_view str) { sal_uInt32 result = 0; - for( sal_Int32 i = nStart; i < nStart + nLength; ) + for( sal_Int32 i = 0; i < static_cast<sal_Int32>(str.size()); ) { - sal_uInt32 c = str.iterateCodePoints(&i); + sal_uInt32 c = o3tl::iterateCodePoints(str, &i); sal_uInt32 value = 0; if( c <= 0x0039) // ASCII decimal digits, most common value = c - 0x0030; @@ -360,12 +359,6 @@ static sal_uInt32 decimalStringToNumber( return result; } -sal_uInt32 decimalStringToNumber( - OUString const & str ) -{ - return decimalStringToNumber(str, 0, str.getLength()); -} - using namespace ::com::sun::star; // convert between sequence of string and comma separated string @@ -435,8 +428,8 @@ sal_Int32 compareNatural( const OUString & rLHS, const OUString & rRHS, if (nStartsDigitLHS > 0 && nStartsDigitRHS > 0) { - sal_uInt32 nLHS = comphelper::string::decimalStringToNumber(rLHS, 0, nStartsDigitLHS); - sal_uInt32 nRHS = comphelper::string::decimalStringToNumber(rRHS, 0, nStartsDigitRHS); + sal_uInt32 nLHS = comphelper::string::decimalStringToNumber(rLHS.subView(0, nStartsDigitLHS)); + sal_uInt32 nRHS = comphelper::string::decimalStringToNumber(rRHS.subView(0, nStartsDigitRHS)); if (nLHS != nRHS) return nLHS < nRHS ? -1 : 1; @@ -484,8 +477,8 @@ sal_Int32 compareNatural( const OUString & rLHS, const OUString & rRHS, //numbers outside of the normal 0-9 range, e.g. see GetLocalizedChar in //vcl - sal_uInt32 nLHS = comphelper::string::decimalStringToNumber(rLHS, nLHSFirstDigitPos, nLHSChunkLen); - sal_uInt32 nRHS = comphelper::string::decimalStringToNumber(rRHS, nRHSFirstDigitPos, nRHSChunkLen); + sal_uInt32 nLHS = comphelper::string::decimalStringToNumber(rLHS.subView(nLHSFirstDigitPos, nLHSChunkLen)); + sal_uInt32 nRHS = comphelper::string::decimalStringToNumber(rRHS.subView(nRHSFirstDigitPos, nRHSChunkLen)); if (nLHS != nRHS) { diff --git a/compilerplugins/clang/stringview.cxx b/compilerplugins/clang/stringview.cxx index d59d1dc68555..9484f3ace957 100644 --- a/compilerplugins/clang/stringview.cxx +++ b/compilerplugins/clang/stringview.cxx @@ -362,7 +362,8 @@ bool StringView::VisitCXXMemberCallExpr(CXXMemberCallExpr const* expr) || dc.Function("equalsIgnoreAsciiCase") || dc.Function("compareToIgnoreAsciiCase") || dc.Function("matchIgnoreAsciiCase") || dc.Function("trim") || dc.Function("startsWith") || dc.Function("endsWith") || dc.Function("match") - || dc.Function("isEmpty") || dc.Function("getLength")) + || dc.Function("isEmpty") || dc.Function("getLength") + || dc.Function("iterateCodePoints")) { handleSubExprThatCouldBeView(expr->getImplicitObjectArgument()); } diff --git a/compilerplugins/clang/stringviewparam.cxx b/compilerplugins/clang/stringviewparam.cxx index d16c306f6326..9687bd8bbe2c 100644 --- a/compilerplugins/clang/stringviewparam.cxx +++ b/compilerplugins/clang/stringviewparam.cxx @@ -140,8 +140,8 @@ DeclRefExpr const* relevantCXXMemberCallExpr(CXXMemberCallExpr const* expr) if (!(n == "getLength" || n == "getStr" || n == "convertToString" || n == "replace" || n == "replaceAll" || n == "replaceAt" || n == "replaceFirst" || n == "toAsciiLowerCase" || n == "toAsciiUpperCase" || n == "toUtf8" - || n == "iterateCodePoints" || n == "startsWithIgnoreAsciiCase" || n == "toUInt64" - || n == "toFloat" || n == "toBoolean")) + || n == "startsWithIgnoreAsciiCase" || n == "toUInt64" || n == "toFloat" + || n == "toBoolean")) { good = true; } diff --git a/compilerplugins/clang/stringviewvar.cxx b/compilerplugins/clang/stringviewvar.cxx index 06123531e03b..5acc09d3fe78 100644 --- a/compilerplugins/clang/stringviewvar.cxx +++ b/compilerplugins/clang/stringviewvar.cxx @@ -129,7 +129,7 @@ DeclRefExpr const* relevantCXXMemberCallExpr(CXXMemberCallExpr const* expr) || n == "indexOf" || n == "lastIndexOf" || n == "compareTo" || n == "match" || n == "trim" || n == "toInt32" || n == "toInt64" || n == "toDouble" || n == "equalsIgnoreAsciiCase" || n == "compareToIgnoreAsciiCase" || n == "getToken" - || n == "copy") + || n == "copy" || n == "iterateCodePoints") { good = true; } diff --git a/compilerplugins/clang/test/stringviewparam.cxx b/compilerplugins/clang/test/stringviewparam.cxx index 78176fa096e3..abb98797b1dc 100644 --- a/compilerplugins/clang/test/stringviewparam.cxx +++ b/compilerplugins/clang/test/stringviewparam.cxx @@ -95,4 +95,19 @@ void f11(const OUString& f11rString) buf.append(f11rString); } +// expected-error-re@+1 {{replace function parameter of type 'const {{(rtl::)?}}OUString &' with 'std::u16string_view' [loplugin:stringviewparam]}} +sal_uInt32 decimalStringToNumber(OUString const& str, sal_Int32 nStart, sal_Int32 nLength) +{ + sal_uInt32 result = 0; + for (sal_Int32 i = nStart; i < nStart + nLength;) + { + sal_uInt32 c = str.iterateCodePoints(&i); + sal_uInt32 value = 0; + if (c <= 0x0039) + value = c - 0x0030; + result = result * 10 + value; + } + return result; +} + /* vim:set shiftwidth=4 softtabstop=4 expandtab cinoptions=b1,g0,N-s cinkeys+=0=break: */ diff --git a/i18npool/inc/cclass_unicode.hxx b/i18npool/inc/cclass_unicode.hxx index 03c03d5bbc97..3a5f28f124ab 100644 --- a/i18npool/inc/cclass_unicode.hxx +++ b/i18npool/inc/cclass_unicode.hxx @@ -178,7 +178,7 @@ private: void setupInternational( const css::lang::Locale& rLocale ); /// Implementation of getCharacterType() for one single character - static sal_Int32 getCharType( const OUString& Text, sal_Int32 *nPos, sal_Int32 increment); + static sal_Int32 getCharType( std::u16string_view Text, sal_Int32 *nPos, sal_Int32 increment); }; diff --git a/i18npool/source/breakiterator/breakiteratorImpl.cxx b/i18npool/source/breakiterator/breakiteratorImpl.cxx index 71d794e2c9d4..5402caa7a469 100644 --- a/i18npool/source/breakiterator/breakiteratorImpl.cxx +++ b/i18npool/source/breakiterator/breakiteratorImpl.cxx @@ -23,6 +23,7 @@ #include <cppuhelper/supportsservice.hxx> #include <unicode/uchar.h> #include <i18nutil/unicode.hxx> +#include <o3tl/string_view.hxx> #include <com/sun/star/i18n/CharType.hpp> #include <com/sun/star/i18n/ScriptType.hpp> @@ -72,7 +73,7 @@ sal_Int32 SAL_CALL BreakIteratorImpl::previousCharacters( const OUString& Text, #define isZWSP(c) (ch == 0x200B) -static sal_Int32 skipSpace(const OUString& Text, sal_Int32 nPos, sal_Int32 len, sal_Int16 rWordType, bool bDirection) +static sal_Int32 skipSpace(std::u16string_view Text, sal_Int32 nPos, sal_Int32 len, sal_Int16 rWordType, bool bDirection) { sal_uInt32 ch=0; sal_Int32 pos=nPos; @@ -81,7 +82,7 @@ static sal_Int32 skipSpace(const OUString& Text, sal_Int32 nPos, sal_Int32 len, if (bDirection) while (nPos < len) { - ch = Text.iterateCodePoints(&pos); + ch = o3tl::iterateCodePoints(Text, &pos); if (!u_isUWhiteSpace(ch) && !isZWSP(ch)) break; nPos = pos; @@ -89,7 +90,7 @@ static sal_Int32 skipSpace(const OUString& Text, sal_Int32 nPos, sal_Int32 len, else while (nPos > 0) { - ch = Text.iterateCodePoints(&pos, -1); + ch = o3tl::iterateCodePoints(Text, &pos, -1); if (!u_isUWhiteSpace(ch) && !isZWSP(ch)) break; nPos = pos; @@ -99,7 +100,7 @@ static sal_Int32 skipSpace(const OUString& Text, sal_Int32 nPos, sal_Int32 len, if (bDirection) while (nPos < len) { - ch = Text.iterateCodePoints(&pos); + ch = o3tl::iterateCodePoints(Text, &pos); if (!u_isWhitespace(ch) && !isZWSP(ch) && (ch == 0x002E || u_isalnum(ch))) break; nPos = pos; @@ -107,7 +108,7 @@ static sal_Int32 skipSpace(const OUString& Text, sal_Int32 nPos, sal_Int32 len, else while (nPos > 0) { - ch = Text.iterateCodePoints(&pos, -1); + ch = o3tl::iterateCodePoints(Text, &pos, -1); if (!u_isWhitespace(ch) && !isZWSP(ch) && (ch == 0x002E || u_isalnum(ch))) break; nPos = pos; @@ -117,7 +118,7 @@ static sal_Int32 skipSpace(const OUString& Text, sal_Int32 nPos, sal_Int32 len, if (bDirection) while (nPos < len) { - ch = Text.iterateCodePoints(&pos); + ch = o3tl::iterateCodePoints(Text, &pos); if (!u_isUWhiteSpace(ch) && !isZWSP(ch)) break; nPos = pos; @@ -125,7 +126,7 @@ static sal_Int32 skipSpace(const OUString& Text, sal_Int32 nPos, sal_Int32 len, else while (nPos > 0) { - ch = Text.iterateCodePoints(&pos, -1); + ch = o3tl::iterateCodePoints(Text, &pos, -1); if (!u_isUWhiteSpace(ch) && !isZWSP(ch)) break; nPos = pos; diff --git a/i18npool/source/characterclassification/cclass_unicode.cxx b/i18npool/source/characterclassification/cclass_unicode.cxx index f07e9f812951..e3c27e9bc0fa 100644 --- a/i18npool/source/characterclassification/cclass_unicode.cxx +++ b/i18npool/source/characterclassification/cclass_unicode.cxx @@ -27,6 +27,7 @@ #include <breakiteratorImpl.hxx> #include <transliteration_body.hxx> #include <rtl/ref.hxx> +#include <o3tl/string_view.hxx> #include <utility> using namespace ::com::sun::star; @@ -147,10 +148,10 @@ cclass_Unicode::getScript( const OUString& Text, sal_Int32 nPos ) { sal_Int32 -cclass_Unicode::getCharType( const OUString& Text, sal_Int32* nPos, sal_Int32 increment) { +cclass_Unicode::getCharType( std::u16string_view Text, sal_Int32* nPos, sal_Int32 increment) { using namespace ::com::sun::star::i18n::KCharacterType; - sal_uInt32 ch = Text.iterateCodePoints(nPos, increment); + sal_uInt32 ch = o3tl::iterateCodePoints(Text, nPos, increment); switch ( u_charType(ch) ) { // Upper case U_UPPERCASE_LETTER : diff --git a/include/comphelper/string.hxx b/include/comphelper/string.hxx index 8144a7e95e28..2763ded2e9fd 100644 --- a/include/comphelper/string.hxx +++ b/include/comphelper/string.hxx @@ -300,7 +300,7 @@ COMPHELPER_DLLPUBLIC OString join(std::string_view rSeparator, const std::vector @return The value of the string as an int32. */ COMPHELPER_DLLPUBLIC sal_uInt32 decimalStringToNumber( - OUString const & str ); + std::u16string_view str ); COMPHELPER_DLLPUBLIC std::vector<OUString> split(std::u16string_view rString, const sal_Unicode cSeparator); diff --git a/include/o3tl/string_view.hxx b/include/o3tl/string_view.hxx index 5d03b2cfba87..07278be8e529 100644 --- a/include/o3tl/string_view.hxx +++ b/include/o3tl/string_view.hxx @@ -509,7 +509,7 @@ inline double toDouble(std::string_view str) } // Like OUString::iterateCodePoints, but for std::string_view: -inline sal_uInt32 iterateCodePoints(std::u16string_view string, std::size_t* indexUtf16, +inline sal_uInt32 iterateCodePoints(std::u16string_view string, sal_Int32* indexUtf16, sal_Int32 incrementCodePoints = 1) { std::size_t n; diff --git a/o3tl/qa/test-string_view.cxx b/o3tl/qa/test-string_view.cxx index c658da16b5dc..bbe5cbf0eddc 100644 --- a/o3tl/qa/test-string_view.cxx +++ b/o3tl/qa/test-string_view.cxx @@ -736,15 +736,15 @@ private: void testIterateCodePoints() { { - std::size_t i = 1; + sal_Int32 i = 1; auto const c = o3tl::iterateCodePoints(u"\U00010000", &i, 1); - CPPUNIT_ASSERT_EQUAL(std::size_t(2), i); + CPPUNIT_ASSERT_EQUAL(sal_Int32(2), i); CPPUNIT_ASSERT_EQUAL(sal_uInt32(0xDC00), c); } { - std::size_t i = 2; + sal_Int32 i = 2; auto const c = o3tl::iterateCodePoints(u"a\U00010000", &i, -1); - CPPUNIT_ASSERT_EQUAL(std::size_t(1), i); + CPPUNIT_ASSERT_EQUAL(sal_Int32(1), i); CPPUNIT_ASSERT_EQUAL(sal_uInt32(0x10000), c); } } diff --git a/sal/rtl/ustring.cxx b/sal/rtl/ustring.cxx index c863be00512b..2a730f8cb5fa 100644 --- a/sal/rtl/ustring.cxx +++ b/sal/rtl/ustring.cxx @@ -778,11 +778,8 @@ sal_uInt32 SAL_CALL rtl_uString_iterateCodePoints( && o3tl::cmp_less_equal(*indexUtf16, std::numeric_limits<std::size_t>::max())); // using o3tl::cmp_less_equal nicely avoids potential // -Wtautological-constant-out-of-range-compare - std::size_t i = *indexUtf16; auto const cp = o3tl::iterateCodePoints( - std::u16string_view(string->buffer, string->length), &i, incrementCodePoints); - assert(i <= o3tl::make_unsigned(std::numeric_limits<sal_Int32>::max())); - *indexUtf16 = i; + std::u16string_view(string->buffer, string->length), indexUtf16, incrementCodePoints); return cp; } diff --git a/vcl/source/gdi/impglyphitem.cxx b/vcl/source/gdi/impglyphitem.cxx index 6a2c576e1f1a..fed2ee0f3fe7 100644 --- a/vcl/source/gdi/impglyphitem.cxx +++ b/vcl/source/gdi/impglyphitem.cxx @@ -26,6 +26,7 @@ #include <unotools/configmgr.hxx> #include <TextLayoutCache.hxx> #include <officecfg/Office/Common.hxx> +#include <o3tl/string_view.hxx> #include <unicode/ubidi.h> #include <unicode/uchar.h> @@ -238,7 +239,7 @@ SalLayoutGlyphsCache* SalLayoutGlyphsCache::self() return cache.get(); } -static UBiDiDirection getBiDiDirection(const OUString& text, sal_Int32 index, sal_Int32 len) +static UBiDiDirection getBiDiDirection(std::u16string_view text, sal_Int32 index, sal_Int32 len) { // Return whether all character are LTR, RTL, neutral or whether it's mixed. // This is sort of ubidi_getBaseDirection() and ubidi_getDirection(), @@ -247,7 +248,7 @@ static UBiDiDirection getBiDiDirection(const OUString& text, sal_Int32 index, sa UBiDiDirection direction = UBIDI_NEUTRAL; while (index < end) { - switch (u_charDirection(text.iterateCodePoints(&index))) + switch (u_charDirection(o3tl::iterateCodePoints(text, &index))) { // Only characters with strong direction. case U_LEFT_TO_RIGHT: @@ -269,7 +270,7 @@ static UBiDiDirection getBiDiDirection(const OUString& text, sal_Int32 index, sa } static SalLayoutGlyphs makeGlyphsSubset(const SalLayoutGlyphs& source, - const OutputDevice* outputDevice, const OUString& text, + const OutputDevice* outputDevice, std::u16string_view text, sal_Int32 index, sal_Int32 len) { // tdf#149264: We need to check if the text is LTR, RTL or mixed. Apparently
