include/tools/inetmime.hxx | 14 +--
sal/rtl/uri.cxx | 31
-------
sal/rtl/ustrbuf.cxx | 12 ---
sal/textenc/convertbig5hkscs.cxx | 14 +--
sal/textenc/converteuctw.cxx | 10 +-
sal/textenc/convertgb18030.cxx | 13 +--
sal/textenc/convertisciidevangari.cxx | 8 +-
sal/textenc/convertiso2022cn.cxx | 10 +-
sal/textenc/convertiso2022jp.cxx | 8 +-
sal/textenc/convertiso2022kr.cxx | 8 +-
sal/textenc/convertsinglebytetobmpunicode.cxx | 8 +-
sal/textenc/handleundefinedunicodetotextchar.cxx | 4 -
sal/textenc/tcvtutf8.cxx | 13 +--
sal/textenc/unichars.hxx | 24 ------
sax/source/expatwrap/saxwriter.cxx | 40
+++++-----
stoc/source/uriproc/UriSchemeParser_vndDOTsunDOTstarDOTscript.cxx | 9 --
tools/source/inet/inetmime.cxx | 28
-------
vcl/source/gdi/scrptrun.cxx | 10 +-
vcl/win/window/salframe.cxx | 5 -
19 files changed, 87 insertions(+), 182 deletions(-)
New commits:
commit 8b333c76945960fc62a01829666ba234f59a6d94
Author: Mike Kaganski <mike.kagan...@collabora.com>
AuthorDate: Fri Dec 24 15:25:52 2021 +0300
Commit: Mike Kaganski <mike.kagan...@collabora.com>
CommitDate: Fri Dec 24 21:42:08 2021 +0100
Use rtl functions instead of own surrogate checking/combining
Change-Id: I3eb05d8f5b0761bc3b672d4c855eb469f8cc1a29
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/127375
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kagan...@collabora.com>
diff --git a/include/tools/inetmime.hxx b/include/tools/inetmime.hxx
index ce0469fa131a..a3c7de49690c 100644
--- a/include/tools/inetmime.hxx
+++ b/include/tools/inetmime.hxx
@@ -227,16 +227,12 @@ inline int INetMIME::getHexWeight(sal_uInt32 nChar)
inline sal_uInt32 INetMIME::getUTF32Character(const sal_Unicode *& rBegin,
const sal_Unicode * pEnd)
{
- DBG_ASSERT(rBegin && rBegin < pEnd,
+ assert(rBegin && rBegin < pEnd &&
"INetMIME::getUTF32Character(): Bad sequence");
- if (rBegin + 1 < pEnd && rBegin[0] >= 0xD800 && rBegin[0] <= 0xDBFF
- && rBegin[1] >= 0xDC00 && rBegin[1] <= 0xDFFF)
- {
- sal_uInt32 nUTF32 = sal_uInt32(*rBegin++ & 0x3FF) << 10;
- return (nUTF32 | (*rBegin++ & 0x3FF)) + 0x10000;
- }
- else
- return *rBegin++;
+ sal_uInt32 nUTF32 = *rBegin++;
+ if (rBegin < pEnd && rtl::isHighSurrogate(nUTF32) &&
rtl::isLowSurrogate(rBegin[0]))
+ nUTF32 = rtl::combineSurrogates(nUTF32, *rBegin++);
+ return nUTF32;
}
diff --git a/sal/rtl/uri.cxx b/sal/rtl/uri.cxx
index 441a6c69f1e8..0c5479563405 100644
--- a/sal/rtl/uri.cxx
+++ b/sal/rtl/uri.cxx
@@ -219,21 +219,7 @@ namespace {
void writeUcs4(rtl_uString ** pBuffer, sal_Int32 * pCapacity, sal_uInt32
nUtf32)
{
- assert(rtl::isUnicodeCodePoint(nUtf32));
- if (nUtf32 <= 0xFFFF)
- {
- writeUnicode(pBuffer, pCapacity, static_cast< sal_Unicode >(nUtf32));
- }
- else
- {
- nUtf32 -= 0x10000;
- writeUnicode(
- pBuffer, pCapacity,
- static_cast< sal_Unicode >(nUtf32 >> 10 | 0xD800));
- writeUnicode(
- pBuffer, pCapacity,
- static_cast< sal_Unicode >((nUtf32 & 0x3FF) | 0xDC00));
- }
+ rtl_uStringbuffer_insertUtf32(pBuffer, pCapacity, (*pBuffer)->length,
nUtf32);
}
void writeEscapeOctet(rtl_uString ** pBuffer, sal_Int32 * pCapacity,
@@ -284,20 +270,7 @@ bool writeEscapeChar(rtl_uString ** pBuffer, sal_Int32 *
pCapacity,
rtl_UnicodeToTextConverter aConverter
= rtl_createUnicodeToTextConverter(eCharset);
sal_Unicode aSrc[2];
- sal_Size nSrcSize;
- if (nUtf32 <= 0xFFFF)
- {
- aSrc[0] = static_cast< sal_Unicode >(nUtf32);
- nSrcSize = 1;
- }
- else
- {
- aSrc[0] = static_cast< sal_Unicode >(
- ((nUtf32 - 0x10000) >> 10) | 0xD800);
- aSrc[1] = static_cast< sal_Unicode >(
- ((nUtf32 - 0x10000) & 0x3FF) | 0xDC00);
- nSrcSize = 2;
- }
+ sal_Size nSrcSize = rtl::splitSurrogates(nUtf32, aSrc);
char aDst[32]; // FIXME random value
sal_uInt32 nInfo;
diff --git a/sal/rtl/ustrbuf.cxx b/sal/rtl/ustrbuf.cxx
index 53e178e5881a..7c5b187edbcb 100644
--- a/sal/rtl/ustrbuf.cxx
+++ b/sal/rtl/ustrbuf.cxx
@@ -174,17 +174,7 @@ void rtl_uStringbuffer_insertUtf32(
SAL_THROW_EXTERN_C()
{
sal_Unicode buf[2];
- sal_Int32 len;
- OSL_ASSERT(rtl::isUnicodeScalarValue(c));
- if (c <= 0xFFFF) {
- buf[0] = static_cast<sal_Unicode>(c);
- len = 1;
- } else {
- c -= 0x10000;
- buf[0] = static_cast<sal_Unicode>((c >> 10) | 0xD800);
- buf[1] = static_cast<sal_Unicode>((c & 0x3FF) | 0xDC00);
- len = 2;
- }
+ sal_Int32 len = rtl::splitSurrogates(c, buf);
rtl_uStringbuffer_insert(pThis, capacity, offset, buf, len);
}
diff --git a/sal/textenc/convertbig5hkscs.cxx b/sal/textenc/convertbig5hkscs.cxx
index 77484666982e..bf3a2e2cf59d 100644
--- a/sal/textenc/convertbig5hkscs.cxx
+++ b/sal/textenc/convertbig5hkscs.cxx
@@ -131,7 +131,7 @@ sal_Size ImplConvertBig5HkscsToUnicode(void const * pData,
nUnicode = pBig5Data[nRow].mpToUniTrailTab[nChar - n];
if (nUnicode == 0)
nUnicode = 0xFFFF;
- assert(!ImplIsHighSurrogate(nUnicode));
+ assert(!rtl::isHighSurrogate(nUnicode));
}
}
if (nUnicode == 0xFFFF)
@@ -192,11 +192,11 @@ sal_Size ImplConvertBig5HkscsToUnicode(void const * pData,
}
++p;
}
- assert(!ImplIsHighSurrogate(nUnicode));
+ assert(!rtl::isHighSurrogate(nUnicode));
}
if (nUnicode == 0xFFFF)
goto bad_input;
- if (ImplIsHighSurrogate(nUnicode))
+ if (rtl::isHighSurrogate(nUnicode))
if (pDestBufEnd - pDestBufPtr >= 2)
{
nOffset += nLast - nFirst + 1;
@@ -329,19 +329,19 @@ sal_Size ImplConvertUnicodeToBig5Hkscs(void const * pData,
sal_uInt32 nChar = *pSrcBuf++;
if (nHighSurrogate == 0)
{
- if (ImplIsHighSurrogate(nChar))
+ if (rtl::isHighSurrogate(nChar))
{
nHighSurrogate = static_cast<sal_Unicode>(nChar);
continue;
}
- else if (ImplIsLowSurrogate(nChar))
+ else if (rtl::isLowSurrogate(nChar))
{
bUndefined = false;
goto bad_input;
}
}
- else if (ImplIsLowSurrogate(nChar))
- nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
+ else if (rtl::isLowSurrogate(nChar))
+ nChar = rtl::combineSurrogates(nHighSurrogate, nChar);
else
{
bUndefined = false;
diff --git a/sal/textenc/converteuctw.cxx b/sal/textenc/converteuctw.cxx
index edb3c07fa934..8684b643c66e 100644
--- a/sal/textenc/converteuctw.cxx
+++ b/sal/textenc/converteuctw.cxx
@@ -206,7 +206,7 @@ sal_Size ImplConvertEucTwToUnicode(void const * pData,
= pCns116431992Data[nOffset + (nChar - nFirst)];
if (nUnicode == 0xFFFF)
goto bad_input;
- else if (ImplIsHighSurrogate(nUnicode))
+ else if (rtl::isHighSurrogate(nUnicode))
if (pDestBufEnd - pDestBufPtr >= 2)
{
nOffset += nLast - nFirst + 1;
@@ -340,19 +340,19 @@ sal_Size ImplConvertUnicodeToEucTw(void const * pData,
sal_uInt32 nChar = *pSrcBuf++;
if (nHighSurrogate == 0)
{
- if (ImplIsHighSurrogate(nChar))
+ if (rtl::isHighSurrogate(nChar))
{
nHighSurrogate = static_cast<sal_Unicode>(nChar);
continue;
}
- else if (ImplIsLowSurrogate(nChar))
+ else if (rtl::isLowSurrogate(nChar))
{
bUndefined = false;
goto bad_input;
}
}
- else if (ImplIsLowSurrogate(nChar))
- nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
+ else if (rtl::isLowSurrogate(nChar))
+ nChar = rtl::combineSurrogates(nHighSurrogate, nChar);
else
{
bUndefined = false;
diff --git a/sal/textenc/convertgb18030.cxx b/sal/textenc/convertgb18030.cxx
index 88f5a999d87d..d7963762fa2b 100644
--- a/sal/textenc/convertgb18030.cxx
+++ b/sal/textenc/convertgb18030.cxx
@@ -172,10 +172,7 @@ sal_Size ImplConvertGb18030ToUnicode(void const * pData,
if (pDestBufEnd - pDestBufPtr >= 2)
{
nCode -= 189000 - 0x10000;
- *pDestBufPtr++
- =
static_cast<sal_Unicode>(ImplGetHighSurrogate(nCode));
- *pDestBufPtr++
- =
static_cast<sal_Unicode>(ImplGetLowSurrogate(nCode));
+ pDestBufPtr += rtl::splitSurrogates(nCode,
pDestBufPtr);
startOfCurrentChar = nConverted + 1;
}
else
@@ -330,19 +327,19 @@ sal_Size ImplConvertUnicodeToGb18030(void const * pData,
sal_uInt32 nChar = *pSrcBuf++;
if (nHighSurrogate == 0)
{
- if (ImplIsHighSurrogate(nChar))
+ if (rtl::isHighSurrogate(nChar))
{
nHighSurrogate = static_cast<sal_Unicode>(nChar);
continue;
}
- else if (ImplIsLowSurrogate(nChar))
+ else if (rtl::isLowSurrogate(nChar))
{
bUndefined = false;
goto bad_input;
}
}
- else if (ImplIsLowSurrogate(nChar))
- nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
+ else if (rtl::isLowSurrogate(nChar))
+ nChar = rtl::combineSurrogates(nHighSurrogate, nChar);
else
{
bUndefined = false;
diff --git a/sal/textenc/convertisciidevangari.cxx
b/sal/textenc/convertisciidevangari.cxx
index 759328565690..75fcadcf7fe4 100644
--- a/sal/textenc/convertisciidevangari.cxx
+++ b/sal/textenc/convertisciidevangari.cxx
@@ -266,20 +266,20 @@ sal_Size UnicodeToIsciiDevanagari::convert(sal_Unicode
const* pSrcBuf, sal_Size
char cSpecialChar = 0;
if (cHighSurrogate == 0)
{
- if (ImplIsHighSurrogate(c))
+ if (rtl::isHighSurrogate(c))
{
cHighSurrogate = static_cast< sal_Unicode >(c);
continue;
}
- else if (ImplIsLowSurrogate(c))
+ else if (rtl::isLowSurrogate(c))
{
bUndefined = false;
goto bad_input;
}
}
- else if (ImplIsLowSurrogate(c))
+ else if (rtl::isLowSurrogate(c))
{
- c = ImplCombineSurrogates(cHighSurrogate, c);
+ c = rtl::combineSurrogates(cHighSurrogate, c);
}
else
{
diff --git a/sal/textenc/convertiso2022cn.cxx b/sal/textenc/convertiso2022cn.cxx
index 9e89c27486db..60c2b57436ff 100644
--- a/sal/textenc/convertiso2022cn.cxx
+++ b/sal/textenc/convertiso2022cn.cxx
@@ -329,7 +329,7 @@ sal_Size ImplConvertIso2022CnToUnicode(void const * pData,
= pCns116431992Data[nOffset + (nChar - nFirst)];
if (nUnicode == 0xFFFF)
goto bad_input;
- else if (ImplIsHighSurrogate(nUnicode))
+ else if (rtl::isHighSurrogate(nUnicode))
if (pDestBufEnd - pDestBufPtr >= 2)
{
nOffset += nLast - nFirst + 1;
@@ -556,19 +556,19 @@ sal_Size ImplConvertUnicodeToIso2022Cn(void const * pData,
sal_uInt32 nChar = *pSrcBuf++;
if (nHighSurrogate == 0)
{
- if (ImplIsHighSurrogate(nChar))
+ if (rtl::isHighSurrogate(nChar))
{
nHighSurrogate = static_cast<sal_Unicode>(nChar);
continue;
}
- else if (ImplIsLowSurrogate(nChar))
+ else if (rtl::isLowSurrogate(nChar))
{
bUndefined = false;
goto bad_input;
}
}
- else if (ImplIsLowSurrogate(nChar))
- nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
+ else if (rtl::isLowSurrogate(nChar))
+ nChar = rtl::combineSurrogates(nHighSurrogate, nChar);
else
{
bUndefined = false;
diff --git a/sal/textenc/convertiso2022jp.cxx b/sal/textenc/convertiso2022jp.cxx
index 4024653f04e5..935bc8515d77 100644
--- a/sal/textenc/convertiso2022jp.cxx
+++ b/sal/textenc/convertiso2022jp.cxx
@@ -375,19 +375,19 @@ sal_Size ImplConvertUnicodeToIso2022Jp(void const * pData,
sal_uInt32 nChar = *pSrcBuf++;
if (nHighSurrogate == 0)
{
- if (ImplIsHighSurrogate(nChar))
+ if (rtl::isHighSurrogate(nChar))
{
nHighSurrogate = static_cast<sal_Unicode>(nChar);
continue;
}
- else if (ImplIsLowSurrogate(nChar))
+ else if (rtl::isLowSurrogate(nChar))
{
bUndefined = false;
goto bad_input;
}
}
- else if (ImplIsLowSurrogate(nChar))
- nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
+ else if (rtl::isLowSurrogate(nChar))
+ nChar = rtl::combineSurrogates(nHighSurrogate, nChar);
else
{
bUndefined = false;
diff --git a/sal/textenc/convertiso2022kr.cxx b/sal/textenc/convertiso2022kr.cxx
index 5c7971ba5af6..5aea5c66c877 100644
--- a/sal/textenc/convertiso2022kr.cxx
+++ b/sal/textenc/convertiso2022kr.cxx
@@ -353,19 +353,19 @@ sal_Size ImplConvertUnicodeToIso2022Kr(void const * pData,
sal_uInt32 nChar = *pSrcBuf++;
if (nHighSurrogate == 0)
{
- if (ImplIsHighSurrogate(nChar))
+ if (rtl::isHighSurrogate(nChar))
{
nHighSurrogate = static_cast<sal_Unicode>(nChar);
continue;
}
- else if (ImplIsLowSurrogate(nChar))
+ else if (rtl::isLowSurrogate(nChar))
{
bUndefined = false;
goto bad_input;
}
}
- else if (ImplIsLowSurrogate(nChar))
- nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
+ else if (rtl::isLowSurrogate(nChar))
+ nChar = rtl::combineSurrogates(nHighSurrogate, nChar);
else
{
bUndefined = false;
diff --git a/sal/textenc/convertsinglebytetobmpunicode.cxx
b/sal/textenc/convertsinglebytetobmpunicode.cxx
index b948ba0fbc5c..43c2b9cf39ce 100644
--- a/sal/textenc/convertsinglebytetobmpunicode.cxx
+++ b/sal/textenc/convertsinglebytetobmpunicode.cxx
@@ -111,17 +111,17 @@ sal_Size rtl_textenc_convertBmpUnicodeToSingleByte(
bool undefined = true;
sal_uInt32 c = *srcBuf++;
if (highSurrogate == 0) {
- if (ImplIsHighSurrogate(c)) {
+ if (rtl::isHighSurrogate(c)) {
highSurrogate = static_cast< sal_Unicode >(c);
continue;
}
- else if (ImplIsLowSurrogate(c))
+ else if (rtl::isLowSurrogate(c))
{
undefined = false;
goto bad_input;
}
- } else if (ImplIsLowSurrogate(c)) {
- c = ImplCombineSurrogates(highSurrogate, c);
+ } else if (rtl::isLowSurrogate(c)) {
+ c = rtl::combineSurrogates(highSurrogate, c);
} else {
undefined = false;
goto bad_input;
diff --git a/sal/textenc/handleundefinedunicodetotextchar.cxx
b/sal/textenc/handleundefinedunicodetotextchar.cxx
index 320562495e97..76aed03e1881 100644
--- a/sal/textenc/handleundefinedunicodetotextchar.cxx
+++ b/sal/textenc/handleundefinedunicodetotextchar.cxx
@@ -106,7 +106,7 @@ bool sal::detail::textenc::handleUndefinedUnicodeToTextChar(
/* Surrogates Characters should result in */
/* one replacement character */
- if (ImplIsHighSurrogate(c))
+ if (rtl::isHighSurrogate(c))
{
if ( ((*ppSrcBuf) + 1) == pEndSrcBuf )
{
@@ -115,7 +115,7 @@ bool sal::detail::textenc::handleUndefinedUnicodeToTextChar(
}
c = *((*ppSrcBuf)+1);
- if (ImplIsLowSurrogate(c))
+ if (rtl::isLowSurrogate(c))
(*ppSrcBuf)++;
else
{
diff --git a/sal/textenc/tcvtutf8.cxx b/sal/textenc/tcvtutf8.cxx
index ca29156c418f..f210b654d57f 100644
--- a/sal/textenc/tcvtutf8.cxx
+++ b/sal/textenc/tcvtutf8.cxx
@@ -199,10 +199,7 @@ sal_Size ImplConvertUtf8ToUnicode(
else
goto no_output;
else if (pDestBufEnd - pDestBufPtr >= 2)
- {
- *pDestBufPtr++ =
static_cast<sal_Unicode>(ImplGetHighSurrogate(nUtf32));
- *pDestBufPtr++ =
static_cast<sal_Unicode>(ImplGetLowSurrogate(nUtf32));
- }
+ pDestBufPtr += rtl::splitSurrogates(nUtf32, pDestBufPtr);
else
goto no_output;
}
@@ -349,18 +346,18 @@ sal_Size ImplConvertUnicodeToUtf8(
sal_uInt32 nChar = *pSrcBufPtr++;
if (nHighSurrogate == 0)
{
- if (ImplIsHighSurrogate(nChar) && !bJavaUtf8)
+ if (rtl::isHighSurrogate(nChar) && !bJavaUtf8)
{
nHighSurrogate = static_cast<sal_Unicode>(nChar);
continue;
}
- else if (ImplIsLowSurrogate(nChar) && !bJavaUtf8)
+ else if (rtl::isLowSurrogate(nChar) && !bJavaUtf8)
{
goto bad_input;
}
}
- else if (ImplIsLowSurrogate(nChar) && !bJavaUtf8)
- nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
+ else if (rtl::isLowSurrogate(nChar) && !bJavaUtf8)
+ nChar = rtl::combineSurrogates(nHighSurrogate, nChar);
else
goto bad_input;
diff --git a/sal/textenc/unichars.hxx b/sal/textenc/unichars.hxx
index 8ca1021da005..e627d3c95c3e 100644
--- a/sal/textenc/unichars.hxx
+++ b/sal/textenc/unichars.hxx
@@ -31,34 +31,10 @@
bool ImplIsControlOrFormat(sal_uInt32 nUtf32);
-// All code points that are high-surrogates, as of Unicode 3.1.1.
-inline bool ImplIsHighSurrogate(sal_uInt32 nUtf32) { return nUtf32 >= 0xD800
&& nUtf32 <= 0xDBFF; }
-
-// All code points that are low-surrogates, as of Unicode 3.1.1.
-inline bool ImplIsLowSurrogate(sal_uInt32 nUtf32) { return nUtf32 >= 0xDC00 &&
nUtf32 <= 0xDFFF; }
-
bool ImplIsPrivateUse(sal_uInt32 nUtf32);
bool ImplIsZeroWidth(sal_uInt32 nUtf32);
-inline sal_uInt32 ImplGetHighSurrogate(sal_uInt32 nUtf32)
-{
- assert(nUtf32 >= 0x10000);
- return ((nUtf32 - 0x10000) >> 10) | 0xD800;
-}
-
-inline sal_uInt32 ImplGetLowSurrogate(sal_uInt32 nUtf32)
-{
- assert(nUtf32 >= 0x10000);
- return ((nUtf32 - 0x10000) & 0x3FF) | 0xDC00;
-}
-
-inline sal_uInt32 ImplCombineSurrogates(sal_uInt32 nHigh, sal_uInt32 nLow)
-{
- assert(ImplIsHighSurrogate(nHigh) && ImplIsLowSurrogate(nLow));
- return (((nHigh & 0x3FF) << 10) | (nLow & 0x3FF)) + 0x10000;
-}
-
#endif
/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/sax/source/expatwrap/saxwriter.cxx
b/sax/source/expatwrap/saxwriter.cxx
index 37eb58f099db..e19a31211d98 100644
--- a/sax/source/expatwrap/saxwriter.cxx
+++ b/sax/source/expatwrap/saxwriter.cxx
@@ -447,20 +447,22 @@ bool SaxWriterHelper::convertToXML(const sal_Unicode*
pStr, sal_Int32 nStrLen,
}
// Deal with other unicode cases
- if (c >= 0xd800 && c < 0xdc00)
+ if (rtl::isHighSurrogate(c))
{
// 1. surrogate: save (until 2. surrogate)
- OSL_ENSURE(nSurrogate == 0, "left-over Unicode surrogate");
- nSurrogate = ((c & 0x03ff) + 0x0040);
+ if (nSurrogate != 0) // left-over lone 1st Unicode surrogate
+ {
+ OSL_FAIL("left-over Unicode surrogate");
+ bRet = false;
+ }
+ nSurrogate = c;
}
- else if (c >= 0xdc00 && c < 0xe000)
+ else if (rtl::isLowSurrogate(c))
{
// 2. surrogate: write as UTF-8
- OSL_ENSURE(nSurrogate != 0, "lone 2nd Unicode surrogate");
-
- nSurrogate = (nSurrogate << 10) | (c & 0x03ff);
- if (rtl::isUnicodeScalarValue(nSurrogate) && nSurrogate >=
0x00010000)
+ if (nSurrogate) // can only be 1st surrogate
{
+ nSurrogate = rtl::combineSurrogates(nSurrogate, c);
sal_Int8 aBytes[] = { sal_Int8(0xF0 | ((nSurrogate >> 18)
& 0x0F)),
sal_Int8(0x80 | ((nSurrogate >> 12)
& 0x3F)),
sal_Int8(0x80 | ((nSurrogate >> 6) &
0x3F)),
@@ -479,7 +481,7 @@ bool SaxWriterHelper::convertToXML(const sal_Unicode* pStr,
sal_Int32 nStrLen,
rPos++;
}
}
- else
+ else // lone 2nd surrogate
{
OSL_FAIL("illegal Unicode character");
bRet = false;
@@ -526,13 +528,18 @@ bool SaxWriterHelper::convertToXML(const sal_Unicode*
pStr, sal_Int32 nStrLen,
rPos = writeSequence();
// reset left-over surrogate
- if ((nSurrogate != 0) && (c < 0xd800 || c >= 0xdc00))
+ if ((nSurrogate != 0) && !rtl::isHighSurrogate(c))
{
- OSL_ENSURE(nSurrogate != 0, "left-over Unicode surrogate");
+ OSL_FAIL("left-over Unicode surrogate");
nSurrogate = 0;
bRet = false;
}
}
+ if (nSurrogate != 0) // trailing lone 1st surrogate
+ {
+ OSL_FAIL("left-over Unicode surrogate");
+ bRet = false;
+ }
return bRet;
}
@@ -951,16 +958,15 @@ sal_Int32 SaxWriterHelper::calcXMLByteLength(const
OUString& rStr, bool bDoNorma
}
// Deal with other unicode cases
- if (c >= 0xd800 && c < 0xdc00)
+ if (rtl::isHighSurrogate(c))
{
// save surrogate
- nSurrogate = ((c & 0x03ff) + 0x0040);
+ nSurrogate = c;
}
- else if (c >= 0xdc00 && c < 0xe000)
+ else if (rtl::isLowSurrogate(c))
{
// 2. surrogate: write as UTF-8 (if range is OK
- nSurrogate = (nSurrogate << 10) | (c & 0x03ff);
- if (rtl::isUnicodeScalarValue(nSurrogate) && nSurrogate >=
0x00010000)
+ if (nSurrogate)
nOutputLength += 4;
nSurrogate = 0;
}
@@ -975,7 +981,7 @@ sal_Int32 SaxWriterHelper::calcXMLByteLength(const
OUString& rStr, bool bDoNorma
}
// surrogate processing
- if ((nSurrogate != 0) && (c < 0xd800 || c >= 0xdc00))
+ if ((nSurrogate != 0) && !rtl::isHighSurrogate(c))
nSurrogate = 0;
}
diff --git a/stoc/source/uriproc/UriSchemeParser_vndDOTsunDOTstarDOTscript.cxx
b/stoc/source/uriproc/UriSchemeParser_vndDOTsunDOTstarDOTscript.cxx
index 1614f44b2f89..054c6ca3d76c 100644
--- a/stoc/source/uriproc/UriSchemeParser_vndDOTsunDOTstarDOTscript.cxx
+++ b/stoc/source/uriproc/UriSchemeParser_vndDOTsunDOTstarDOTscript.cxx
@@ -115,14 +115,7 @@ OUString parsePart(
{
break;
}
- if (encoded <= 0xFFFF) {
- buf.append(static_cast< sal_Unicode >(encoded));
- } else {
- buf.append(static_cast< sal_Unicode >(
- (encoded >> 10) | 0xD800));
- buf.append(static_cast< sal_Unicode >(
- (encoded & 0x3FF) | 0xDC00));
- }
+ buf.appendUtf32(encoded);
} else {
break;
}
diff --git a/tools/source/inet/inetmime.cxx b/tools/source/inet/inetmime.cxx
index f7265523b6e2..85f03cfce3e2 100644
--- a/tools/source/inet/inetmime.cxx
+++ b/tools/source/inet/inetmime.cxx
@@ -131,30 +131,6 @@ std::unique_ptr<sal_Unicode[]> convertToUnicode(const char
* pBegin,
return pBuffer;
}
-/** Put the UTF-16 encoding of a UTF-32 character into a buffer.
-
- @param pBuffer Points to a buffer, must not be null.
-
- @param nUTF32 A UTF-32 character, must be in the range 0..0x10FFFF.
-
- @return A pointer past the UTF-16 characters put into the buffer
- (i.e., pBuffer + 1 or pBuffer + 2).
- */
-sal_Unicode * putUTF32Character(sal_Unicode * pBuffer,
- sal_uInt32 nUTF32)
-{
- DBG_ASSERT(rtl::isUnicodeCodePoint(nUTF32), "putUTF32Character(): Bad
char");
- if (nUTF32 < 0x10000)
- *pBuffer++ = sal_Unicode(nUTF32);
- else
- {
- nUTF32 -= 0x10000;
- *pBuffer++ = sal_Unicode(0xD800 | (nUTF32 >> 10));
- *pBuffer++ = sal_Unicode(0xDC00 | (nUTF32 & 0x3FF));
- }
- return pBuffer;
-}
-
void writeUTF8(OStringBuffer & rSink, sal_uInt32 nChar)
{
// See RFC 2279 for a discussion of UTF-8.
@@ -1386,9 +1362,7 @@ OUString INetMIME::decodeHeaderFieldBody(const OString&
rBody)
if (translateUTF8Char(pUTF8End, pEnd, nCharacter))
{
appendISO88591(sDecoded, pCopyBegin, p - 1);
- sal_Unicode aUTF16Buf[2];
- sal_Int32 nUTF16Len = putUTF32Character(aUTF16Buf,
nCharacter) - aUTF16Buf;
- sDecoded.append(aUTF16Buf, nUTF16Len);
+ sDecoded.appendUtf32(nCharacter);
p = pUTF8End;
pCopyBegin = p;
}
diff --git a/vcl/source/gdi/scrptrun.cxx b/vcl/source/gdi/scrptrun.cxx
index c0da5c5322ed..fbc68ca41bf1 100644
--- a/vcl/source/gdi/scrptrun.cxx
+++ b/vcl/source/gdi/scrptrun.cxx
@@ -37,6 +37,10 @@
* This file is largely copied from the ICU project,
* under folder source/extra/scrptrun/scrptrun.cpp
*/
+
+#include <sal/config.h>
+
+#include <rtl/character.hxx>
#include <unicode/utypes.h>
#include <unicode/uscript.h>
@@ -160,14 +164,14 @@ UBool ScriptRun::next()
// if the character is a high surrogate and it's not the last one
// in the text, see if it's followed by a low surrogate
- if (high >= 0xD800 && high <= 0xDBFF && scriptEnd < charLimit - 1)
+ if (rtl::isHighSurrogate(high) && scriptEnd < charLimit - 1)
{
UChar low = charArray[scriptEnd + 1];
// if it is followed by a low surrogate,
// consume it and form the full character
- if (low >= 0xDC00 && low <= 0xDFFF) {
- ch = (high - 0xD800) * 0x0400 + low - 0xDC00 + 0x10000;
+ if (rtl::isLowSurrogate(low)) {
+ ch = rtl::combineSurrogates(high, low);
scriptEnd += 1;
}
}
diff --git a/vcl/win/window/salframe.cxx b/vcl/win/window/salframe.cxx
index 6593b2fbebb6..95b9a36693ad 100644
--- a/vcl/win/window/salframe.cxx
+++ b/vcl/win/window/salframe.cxx
@@ -38,6 +38,7 @@
#include <osl/file.hxx>
#include <osl/process.h>
+#include <rtl/character.hxx>
#include <rtl/string.h>
#include <rtl/ustring.h>
#include <sal/log.hxx>
@@ -122,8 +123,6 @@ const unsigned int WM_USER_SYSTEM_WINDOW_ACTIVATED =
RegisterWindowMessageW(L"SY
bool WinSalFrame::mbInReparent = false;
// Macros for support of WM_UNICHAR & Keyman 6.0
-//#define Uni_UTF32ToSurrogate1(ch) (((unsigned long) (ch) - 0x10000) /
0x400 + 0xD800)
-#define Uni_UTF32ToSurrogate2(ch) ((static_cast<tools::ULong>(ch) - 0x10000)
% 0x400 + 0xDC00)
#define Uni_SupplementaryPlanesStart 0x10000
static void UpdateFrameGeometry( HWND hWnd, WinSalFrame* pFrame );
@@ -3472,7 +3471,7 @@ static bool ImplHandleKeyMsg( HWND hWnd, UINT nMsg,
nLastVKChar = 0;
pFrame->CallCallback( SalEvent::KeyInput, &aKeyEvt );
pFrame->CallCallback( SalEvent::KeyUp, &aKeyEvt );
- wParam = static_cast<sal_Unicode>(Uni_UTF32ToSurrogate2( wParam
));
+ wParam = rtl::getLowSurrogate( wParam );
}
aKeyEvt.mnCharCode = static_cast<sal_Unicode>(wParam);
