summaryrefslogtreecommitdiffstats
path: root/src/corelib/codecs
diff options
context:
space:
mode:
authorJason Barron <jbarron@trolltech.com>2009-07-24 09:45:33 (GMT)
committerJason Barron <jbarron@trolltech.com>2009-07-27 13:04:30 (GMT)
commit3643028959f0b38350e57e60ba4000435b75e592 (patch)
treec129e4dee11487abd437ab8ebd993ba261e06fa6 /src/corelib/codecs
parentcf66c667a97c0079141eb3f2d9e997b7378ae792 (diff)
parentc36139c665e61866aff4bf8572890a735167a7d0 (diff)
downloadQt-3643028959f0b38350e57e60ba4000435b75e592.zip
Qt-3643028959f0b38350e57e60ba4000435b75e592.tar.gz
Qt-3643028959f0b38350e57e60ba4000435b75e592.tar.bz2
Merge commit 'qt/master-stable'
Conflicts: configure.exe qmake/Makefile.unix qmake/generators/makefile.cpp src/corelib/global/qglobal.h src/corelib/kernel/kernel.pri src/corelib/kernel/qcoreevent.cpp src/corelib/kernel/qsharedmemory_unix.cpp src/gui/graphicsview/qgraphicsscene.cpp src/gui/kernel/qaction.cpp src/gui/kernel/qaction.h src/gui/kernel/qaction_p.h src/gui/kernel/qapplication.cpp src/gui/kernel/qapplication.h src/gui/kernel/qwidget.cpp src/gui/kernel/qwidget.h src/gui/kernel/qwidget_mac.mm src/gui/painting/qgraphicssystemfactory.cpp src/gui/styles/qwindowsstyle.cpp src/gui/text/qfontengine_qpf.cpp src/gui/widgets/qabstractscrollarea_p.h src/network/access/qnetworkaccessdebugpipebackend.cpp src/network/socket/qlocalsocket_unix.cpp src/network/socket/qnativesocketengine_p.h src/network/socket/qnativesocketengine_unix.cpp src/openvg/qpaintengine_vg.cpp tests/auto/q3sqlcursor/tst_q3sqlcursor.cpp tests/auto/qcssparser/qcssparser.pro tests/auto/qdir/tst_qdir.cpp tests/auto/qfile/tst_qfile.cpp tests/auto/qobject/tst_qobject.cpp tests/auto/qpathclipper/qpathclipper.pro tests/auto/qprocess/tst_qprocess.cpp tests/auto/qsettings/tst_qsettings.cpp tests/auto/qsharedpointer/qsharedpointer.pro tests/auto/qsqlquerymodel/qsqlquerymodel.pro tests/auto/qsqlrelationaltablemodel/qsqlrelationaltablemodel.pro tests/auto/qsqltablemodel/qsqltablemodel.pro tests/auto/qsqlthread/qsqlthread.pro tests/auto/qwidget/tst_qwidget.cpp
Diffstat (limited to 'src/corelib/codecs')
-rw-r--r--src/corelib/codecs/qiconvcodec.cpp6
-rw-r--r--src/corelib/codecs/qtextcodec.cpp51
-rw-r--r--src/corelib/codecs/qtextcodec_p.h27
-rw-r--r--src/corelib/codecs/qutfcodec.cpp349
-rw-r--r--src/corelib/codecs/qutfcodec_p.h53
5 files changed, 280 insertions, 206 deletions
diff --git a/src/corelib/codecs/qiconvcodec.cpp b/src/corelib/codecs/qiconvcodec.cpp
index 1bf76ea..188ac8c 100644
--- a/src/corelib/codecs/qiconvcodec.cpp
+++ b/src/corelib/codecs/qiconvcodec.cpp
@@ -225,10 +225,11 @@ QString QIconvCodec::convertToUnicode(const char* chars, int len, ConverterState
char *inBytes = const_cast<char *>(chars);
#endif
+ QByteArray in;
if (remainingCount) {
// we have to prepend the remaining bytes from the previous conversion
inBytesLeft += remainingCount;
- QByteArray in(inBytesLeft, Qt::Uninitialized);
+ in.resize(inBytesLeft);
inBytes = in.data();
memcpy(in.data(), remainingBuffer, remainingCount);
@@ -362,9 +363,10 @@ QByteArray QIconvCodec::convertFromUnicode(const QChar *uc, int len, ConverterSt
inBytes = const_cast<char *>(reinterpret_cast<const char *>(uc));
inBytesLeft = len * sizeof(QChar);
+ QByteArray in;
if (convState && convState->remainingChars) {
// we have one surrogate char to be prepended
- QByteArray in(sizeof(QChar) + len, Qt::Uninitialized);
+ in.resize(sizeof(QChar) + len);
inBytes = in.data();
QChar remaining = convState->state_data[0];
diff --git a/src/corelib/codecs/qtextcodec.cpp b/src/corelib/codecs/qtextcodec.cpp
index bca36a6..354f29b 100644
--- a/src/corelib/codecs/qtextcodec.cpp
+++ b/src/corelib/codecs/qtextcodec.cpp
@@ -104,6 +104,10 @@ Q_GLOBAL_STATIC_WITH_ARGS(QFactoryLoader, loader,
(QTextCodecFactoryInterface_iid, QLatin1String("/codecs")))
#endif
+static char qtolower(register char c)
+{ if (c >= 'A' && c <= 'Z') return c + 0x20; return c; }
+static bool qisalnum(register char c)
+{ return (c >= '0' && c <= '9') || ((c | 0x20) >= 'a' && (c | 0x20) <= 'z'); }
static bool nameMatch(const QByteArray &name, const QByteArray &test)
{
@@ -116,21 +120,21 @@ static bool nameMatch(const QByteArray &name, const QByteArray &test)
// if the letters and numbers are the same, we have a match
while (*n != '\0') {
- if (isalnum((uchar)*n)) {
+ if (qisalnum(*n)) {
for (;;) {
if (*h == '\0')
return false;
- if (isalnum((uchar)*h))
+ if (qisalnum(*h))
break;
++h;
}
- if (tolower((uchar)*n) != tolower((uchar)*h))
+ if (qtolower(*n) != qtolower(*h))
return false;
++h;
}
++n;
}
- while (*h && !isalnum((uchar)*h))
+ while (*h && !qisalnum(*h))
++h;
return (*h == '\0');
}
@@ -246,9 +250,9 @@ QString QWindowsLocalCodec::convertToUnicode(const char *chars, int length, Conv
return QString();
const int wclen_auto = 4096;
- WCHAR wc_auto[wclen_auto];
+ wchar_t wc_auto[wclen_auto];
int wclen = wclen_auto;
- WCHAR *wc = wc_auto;
+ wchar_t *wc = wc_auto;
int len;
QString sp;
bool prepend = false;
@@ -288,7 +292,7 @@ QString QWindowsLocalCodec::convertToUnicode(const char *chars, int length, Conv
} else {
wclen = MultiByteToWideChar(CP_ACP, MB_PRECOMPOSED,
mb, mblen, 0, 0);
- wc = new WCHAR[wclen];
+ wc = new wchar_t[wclen];
// and try again...
}
} else if (r == ERROR_NO_UNICODE_TRANSLATION) {
@@ -354,7 +358,7 @@ QString QWindowsLocalCodec::convertToUnicodeCharByChar(const char *chars, int le
const char *next = 0;
QString s;
while((next = CharNextExA(CP_ACP, mb, 0)) != mb) {
- WCHAR wc[2] ={0};
+ wchar_t wc[2] ={0};
int charlength = next - mb;
int len = MultiByteToWideChar(CP_ACP, MB_PRECOMPOSED|MB_ERR_INVALID_CHARS, mb, charlength, wc, 2);
if (len>0) {
@@ -1059,16 +1063,10 @@ QList<int> QTextCodec::availableMibs()
This might be needed for some applications that want to use their
own mechanism for setting the locale.
- Setting this codec is not supported on DOS based Windows.
-
\sa codecForLocale()
*/
void QTextCodec::setCodecForLocale(QTextCodec *c)
{
-#ifdef Q_WS_WIN
- if (QSysInfo::WindowsVersion& QSysInfo::WV_DOS_based)
- return;
-#endif
localeMapper = c;
if (!localeMapper)
setupLocaleMapper();
@@ -1555,9 +1553,13 @@ QTextCodec *QTextCodec::codecForHtml(const QByteArray &ba, QTextCodec *defaultCo
}
/*!
- \overload
+ \overload
- If the codec cannot be detected, this overload returns a Latin-1 QTextCodec.
+ Tries to detect the encoding of the provided snippet of HTML in
+ the given byte array, \a ba, by checking the BOM (Byte Order Mark)
+ and the content-type meta header and returns a QTextCodec instance
+ that is capable of decoding the html to unicode. If the codec cannot
+ be detected, this overload returns a Latin-1 QTextCodec.
*/
QTextCodec *QTextCodec::codecForHtml(const QByteArray &ba)
{
@@ -1569,10 +1571,13 @@ QTextCodec *QTextCodec::codecForHtml(const QByteArray &ba)
Tries to detect the encoding of the provided snippet \a ba by
using the BOM (Byte Order Mark) and returns a QTextCodec instance
- that is capable of decoding the text to unicode. If the codec
+ that is capable of decoding the text to unicode. If the codec
cannot be detected from the content provided, \a defaultCodec is
returned.
+ The behavior of this function is undefined if \a ba is not
+ encoded in unicode.
+
\sa codecForHtml()
*/
QTextCodec *QTextCodec::codecForUtfText(const QByteArray &ba, QTextCodec *defaultCodec)
@@ -1610,9 +1615,17 @@ QTextCodec *QTextCodec::codecForUtfText(const QByteArray &ba, QTextCodec *defaul
}
/*!
- \overload
+ \overload
- If the codec cannot be detected, this overload returns a Latin-1 QTextCodec.
+ Tries to detect the encoding of the provided snippet \a ba by
+ using the BOM (Byte Order Mark) and returns a QTextCodec instance
+ that is capable of decoding the text to unicode. If the codec
+ cannot be detected, this overload returns a Latin-1 QTextCodec.
+
+ The behavior of this function is undefined if \a ba is not
+ encoded in unicode.
+
+ \sa codecForHtml()
*/
QTextCodec *QTextCodec::codecForUtfText(const QByteArray &ba)
{
diff --git a/src/corelib/codecs/qtextcodec_p.h b/src/corelib/codecs/qtextcodec_p.h
index 499c0f9..5c82735 100644
--- a/src/corelib/codecs/qtextcodec_p.h
+++ b/src/corelib/codecs/qtextcodec_p.h
@@ -77,6 +77,33 @@ struct QTextCodecUnalignedPointer
}
};
+#else
+
+class QTextCodec
+{
+public:
+ enum ConversionFlag {
+ DefaultConversion,
+ ConvertInvalidToNull = 0x80000000,
+ IgnoreHeader = 0x1,
+ FreeFunction = 0x2
+ };
+ Q_DECLARE_FLAGS(ConversionFlags, ConversionFlag)
+
+ struct ConverterState {
+ ConverterState(ConversionFlags f = DefaultConversion)
+ : flags(f), remainingChars(0), invalidChars(0), d(0) { state_data[0] = state_data[1] = state_data[2] = 0; }
+ ~ConverterState() { }
+ ConversionFlags flags;
+ int remainingChars;
+ int invalidChars;
+ uint state_data[3];
+ void *d;
+ private:
+ Q_DISABLE_COPY(ConverterState)
+ };
+};
+
#endif //QT_NO_TEXTCODEC
QT_END_NAMESPACE
diff --git a/src/corelib/codecs/qutfcodec.cpp b/src/corelib/codecs/qutfcodec.cpp
index abae6f7..abcc07c 100644
--- a/src/corelib/codecs/qutfcodec.cpp
+++ b/src/corelib/codecs/qutfcodec.cpp
@@ -44,23 +44,19 @@
#include "qendian.h"
#include "qchar.h"
-#ifndef QT_NO_TEXTCODEC
-
QT_BEGIN_NAMESPACE
-QUtf8Codec::~QUtf8Codec()
-{
-}
+enum { Endian = 0, Data = 1 };
-QByteArray QUtf8Codec::convertFromUnicode(const QChar *uc, int len, ConverterState *state) const
+QByteArray QUtf8::convertFromUnicode(const QChar *uc, int len, QTextCodec::ConverterState *state)
{
uchar replacement = '?';
int rlen = 3*len;
int surrogate_high = -1;
if (state) {
- if (state->flags & ConvertInvalidToNull)
+ if (state->flags & QTextCodec::ConvertInvalidToNull)
replacement = 0;
- if (!(state->flags & IgnoreHeader))
+ if (!(state->flags & QTextCodec::IgnoreHeader))
rlen += 3;
if (state->remainingChars)
surrogate_high = state->state_data[0];
@@ -71,7 +67,7 @@ QByteArray QUtf8Codec::convertFromUnicode(const QChar *uc, int len, ConverterSta
uchar* cursor = (uchar*)rstr.data();
const QChar *ch = uc;
int invalid = 0;
- if (state && !(state->flags & IgnoreHeader)) {
+ if (state && !(state->flags & QTextCodec::IgnoreHeader)) {
*cursor++ = 0xef;
*cursor++ = 0xbb;
*cursor++ = 0xbf;
@@ -133,7 +129,7 @@ QByteArray QUtf8Codec::convertFromUnicode(const QChar *uc, int len, ConverterSta
rstr.resize(cursor - (const uchar*)rstr.constData());
if (state) {
state->invalidChars += invalid;
- state->flags |= IgnoreHeader;
+ state->flags |= QTextCodec::IgnoreHeader;
state->remainingChars = 0;
if (surrogate_high >= 0) {
state->remainingChars = 1;
@@ -143,18 +139,18 @@ QByteArray QUtf8Codec::convertFromUnicode(const QChar *uc, int len, ConverterSta
return rstr;
}
-void QUtf8Codec::convertToUnicode(QString *target, const char *chars, int len, ConverterState *state) const
+QString QUtf8::convertToUnicode(const char *chars, int len, QTextCodec::ConverterState *state)
{
bool headerdone = false;
- QChar replacement = QChar::ReplacementCharacter;
+ ushort replacement = QChar::ReplacementCharacter;
int need = 0;
int error = -1;
uint uc = 0;
uint min_uc = 0;
if (state) {
- if (state->flags & IgnoreHeader)
+ if (state->flags & QTextCodec::IgnoreHeader)
headerdone = true;
- if (state->flags & ConvertInvalidToNull)
+ if (state->flags & QTextCodec::ConvertInvalidToNull)
replacement = QChar::Null;
need = state->remainingChars;
if (need) {
@@ -170,40 +166,28 @@ void QUtf8Codec::convertToUnicode(QString *target, const char *chars, int len, C
headerdone = true;
}
- int originalLength = target->length();
- QString &result = *target;
- result.resize(originalLength + len + 1); // worst case
- QChar *qch = result.data() + originalLength;
+ QString result(need + len + 1, Qt::Uninitialized); // worst case
+ ushort *qch = (ushort *)result.unicode();
uchar ch;
int invalid = 0;
- for (int i=0; i<len; i++) {
+ for (int i = 0; i < len; ++i) {
ch = chars[i];
if (need) {
if ((ch&0xc0) == 0x80) {
uc = (uc << 6) | (ch & 0x3f);
- need--;
+ --need;
if (!need) {
// utf-8 bom composes into 0xfeff code point
if (!headerdone && uc == 0xfeff) {
// dont do anything, just skip the BOM
} else if (uc > 0xffff && uc < 0x110000) {
// surrogate pair
- uc -= 0x10000;
- unsigned short high = uc/0x400 + 0xd800;
- unsigned short low = uc%0x400 + 0xdc00;
-
- // resize if necessary
- long where = qch - result.unicode();
- if (where + 2 >= result.length()) {
- result.resize(where + 2);
- qch = result.data() + where;
- }
-
- *qch++ = QChar(high);
- *qch++ = QChar(low);
+ Q_ASSERT((qch - (ushort*)result.unicode()) + 2 < result.length());
+ *qch++ = QChar::highSurrogate(uc);
+ *qch++ = QChar::lowSurrogate(uc);
} else if ((uc < min_uc) || (uc >= 0xd800 && uc <= 0xdfff) || (uc >= 0xfffe)) {
- // error
+ // error: overlong sequence, UTF16 surrogate or BOM
*qch++ = replacement;
++invalid;
} else {
@@ -221,7 +205,7 @@ void QUtf8Codec::convertToUnicode(QString *target, const char *chars, int len, C
}
} else {
if (ch < 128) {
- *qch++ = QLatin1Char(ch);
+ *qch++ = ushort(ch);
headerdone = true;
} else if ((ch & 0xe0) == 0xc0) {
uc = ch & 0x1f;
@@ -255,57 +239,35 @@ void QUtf8Codec::convertToUnicode(QString *target, const char *chars, int len, C
++invalid;
}
}
- result.truncate(qch - result.unicode());
+ result.truncate(qch - (ushort *)result.unicode());
if (state) {
state->invalidChars += invalid;
state->remainingChars = need;
if (headerdone)
- state->flags |= IgnoreHeader;
+ state->flags |= QTextCodec::IgnoreHeader;
state->state_data[0] = need ? uc : 0;
state->state_data[1] = need ? min_uc : 0;
}
-}
-
-QString QUtf8Codec::convertToUnicode(const char *chars, int len, ConverterState *state) const
-{
- QString result;
- convertToUnicode(&result, chars, len, state);
return result;
}
-QByteArray QUtf8Codec::name() const
-{
- return "UTF-8";
-}
-
-int QUtf8Codec::mibEnum() const
-{
- return 106;
-}
-
-enum { Endian = 0, Data = 1 };
-
-QUtf16Codec::~QUtf16Codec()
-{
-}
-
-QByteArray QUtf16Codec::convertFromUnicode(const QChar *uc, int len, ConverterState *state) const
+QByteArray QUtf16::convertFromUnicode(const QChar *uc, int len, QTextCodec::ConverterState *state, DataEndianness e)
{
- Endianness endian = e;
+ DataEndianness endian = e;
int length = 2*len;
- if (!state || (!(state->flags & IgnoreHeader))) {
+ if (!state || (!(state->flags & QTextCodec::IgnoreHeader))) {
length += 2;
}
- if (e == Detect) {
- endian = (QSysInfo::ByteOrder == QSysInfo::BigEndian) ? BE : LE;
+ if (e == DetectEndianness) {
+ endian = (QSysInfo::ByteOrder == QSysInfo::BigEndian) ? BigEndianness : LittleEndianness;
}
QByteArray d;
d.resize(length);
char *data = d.data();
- if (!state || !(state->flags & IgnoreHeader)) {
+ if (!state || !(state->flags & QTextCodec::IgnoreHeader)) {
QChar bom(QChar::ByteOrderMark);
- if (endian == BE) {
+ if (endian == BigEndianness) {
data[0] = bom.row();
data[1] = bom.cell();
} else {
@@ -314,7 +276,7 @@ QByteArray QUtf16Codec::convertFromUnicode(const QChar *uc, int len, ConverterSt
}
data += 2;
}
- if (endian == BE) {
+ if (endian == BigEndianness) {
for (int i = 0; i < len; ++i) {
*(data++) = uc[i].row();
*(data++) = uc[i].cell();
@@ -328,35 +290,35 @@ QByteArray QUtf16Codec::convertFromUnicode(const QChar *uc, int len, ConverterSt
if (state) {
state->remainingChars = 0;
- state->flags |= IgnoreHeader;
+ state->flags |= QTextCodec::IgnoreHeader;
}
return d;
}
-QString QUtf16Codec::convertToUnicode(const char *chars, int len, ConverterState *state) const
+QString QUtf16::convertToUnicode(const char *chars, int len, QTextCodec::ConverterState *state, DataEndianness e)
{
- Endianness endian = e;
+ DataEndianness endian = e;
bool half = false;
uchar buf = 0;
bool headerdone = false;
if (state) {
- headerdone = state->flags & IgnoreHeader;
- if (endian == Detect)
- endian = (Endianness)state->state_data[Endian];
+ headerdone = state->flags & QTextCodec::IgnoreHeader;
+ if (endian == DetectEndianness)
+ endian = (DataEndianness)state->state_data[Endian];
if (state->remainingChars) {
half = true;
buf = state->state_data[Data];
}
}
- if (headerdone && endian == Detect)
- endian = (QSysInfo::ByteOrder == QSysInfo::BigEndian) ? BE : LE;
+ if (headerdone && endian == DetectEndianness)
+ endian = (QSysInfo::ByteOrder == QSysInfo::BigEndian) ? BigEndianness : LittleEndianness;
QString result(len, Qt::Uninitialized); // worst case
QChar *qch = (QChar *)result.unicode();
while (len--) {
if (half) {
QChar ch;
- if (endian == LE) {
+ if (endian == LittleEndianness) {
ch.setRow(*chars++);
ch.setCell(buf);
} else {
@@ -364,17 +326,17 @@ QString QUtf16Codec::convertToUnicode(const char *chars, int len, ConverterState
ch.setCell(*chars++);
}
if (!headerdone) {
- if (endian == Detect) {
- if (ch == QChar::ByteOrderSwapped && endian != BE) {
- endian = LE;
- } else if (ch == QChar::ByteOrderMark && endian != LE) {
+ if (endian == DetectEndianness) {
+ if (ch == QChar::ByteOrderSwapped && endian != BigEndianness) {
+ endian = LittleEndianness;
+ } else if (ch == QChar::ByteOrderMark && endian != LittleEndianness) {
// ignore BOM
- endian = BE;
+ endian = BigEndianness;
} else {
if (QSysInfo::ByteOrder == QSysInfo::BigEndian) {
- endian = BE;
+ endian = BigEndianness;
} else {
- endian = LE;
+ endian = LittleEndianness;
ch = QChar((ch.unicode() >> 8) | ((ch.unicode() & 0xff) << 8));
}
*qch++ = ch;
@@ -396,7 +358,7 @@ QString QUtf16Codec::convertToUnicode(const char *chars, int len, ConverterState
if (state) {
if (headerdone)
- state->flags |= IgnoreHeader;
+ state->flags |= QTextCodec::IgnoreHeader;
state->state_data[Endian] = endian;
if (half) {
state->remainingChars = 1;
@@ -409,72 +371,21 @@ QString QUtf16Codec::convertToUnicode(const char *chars, int len, ConverterState
return result;
}
-int QUtf16Codec::mibEnum() const
-{
- return 1015;
-}
-
-QByteArray QUtf16Codec::name() const
-{
- return "UTF-16";
-}
-
-QList<QByteArray> QUtf16Codec::aliases() const
-{
- return QList<QByteArray>();
-}
-
-int QUtf16BECodec::mibEnum() const
-{
- return 1013;
-}
-
-QByteArray QUtf16BECodec::name() const
-{
- return "UTF-16BE";
-}
-
-QList<QByteArray> QUtf16BECodec::aliases() const
-{
- QList<QByteArray> list;
- return list;
-}
-
-int QUtf16LECodec::mibEnum() const
-{
- return 1014;
-}
-
-QByteArray QUtf16LECodec::name() const
-{
- return "UTF-16LE";
-}
-
-QList<QByteArray> QUtf16LECodec::aliases() const
-{
- QList<QByteArray> list;
- return list;
-}
-
-QUtf32Codec::~QUtf32Codec()
-{
-}
-
-QByteArray QUtf32Codec::convertFromUnicode(const QChar *uc, int len, ConverterState *state) const
+QByteArray QUtf32::convertFromUnicode(const QChar *uc, int len, QTextCodec::ConverterState *state, DataEndianness e)
{
- Endianness endian = e;
+ DataEndianness endian = e;
int length = 4*len;
- if (!state || (!(state->flags & IgnoreHeader))) {
+ if (!state || (!(state->flags & QTextCodec::IgnoreHeader))) {
length += 4;
}
- if (e == Detect) {
- endian = (QSysInfo::ByteOrder == QSysInfo::BigEndian) ? BE : LE;
+ if (e == DetectEndianness) {
+ endian = (QSysInfo::ByteOrder == QSysInfo::BigEndian) ? BigEndianness : LittleEndianness;
}
QByteArray d(length, Qt::Uninitialized);
char *data = d.data();
- if (!state || !(state->flags & IgnoreHeader)) {
- if (endian == BE) {
+ if (!state || !(state->flags & QTextCodec::IgnoreHeader)) {
+ if (endian == BigEndianness) {
data[0] = 0;
data[1] = 0;
data[2] = (char)0xfe;
@@ -487,7 +398,7 @@ QByteArray QUtf32Codec::convertFromUnicode(const QChar *uc, int len, ConverterSt
}
data += 4;
}
- if (endian == BE) {
+ if (endian == BigEndianness) {
for (int i = 0; i < len; ++i) {
uint cp = uc[i].unicode();
if (uc[i].isHighSurrogate() && i < len - 1)
@@ -511,59 +422,59 @@ QByteArray QUtf32Codec::convertFromUnicode(const QChar *uc, int len, ConverterSt
if (state) {
state->remainingChars = 0;
- state->flags |= IgnoreHeader;
+ state->flags |= QTextCodec::IgnoreHeader;
}
return d;
}
-QString QUtf32Codec::convertToUnicode(const char *chars, int len, ConverterState *state) const
+QString QUtf32::convertToUnicode(const char *chars, int len, QTextCodec::ConverterState *state, DataEndianness e)
{
- Endianness endian = e;
+ DataEndianness endian = e;
uchar tuple[4];
int num = 0;
bool headerdone = false;
if (state) {
- headerdone = state->flags & IgnoreHeader;
- if (endian == Detect) {
- endian = (Endianness)state->state_data[Endian];
+ headerdone = state->flags & QTextCodec::IgnoreHeader;
+ if (endian == DetectEndianness) {
+ endian = (DataEndianness)state->state_data[Endian];
}
num = state->remainingChars;
memcpy(tuple, &state->state_data[Data], 4);
}
- if (headerdone && endian == Detect)
- endian = (QSysInfo::ByteOrder == QSysInfo::BigEndian) ? BE : LE;
+ if (headerdone && endian == DetectEndianness)
+ endian = (QSysInfo::ByteOrder == QSysInfo::BigEndian) ? BigEndianness : LittleEndianness;
QString result;
result.resize((num + len) >> 2 << 1); // worst case
QChar *qch = (QChar *)result.unicode();
-
+
const char *end = chars + len;
while (chars < end) {
tuple[num++] = *chars++;
if (num == 4) {
if (!headerdone) {
- if (endian == Detect) {
- if (endian == Detect) {
- if (tuple[0] == 0xff && tuple[1] == 0xfe && tuple[2] == 0 && tuple[3] == 0 && endian != BE) {
- endian = LE;
+ if (endian == DetectEndianness) {
+ if (endian == DetectEndianness) {
+ if (tuple[0] == 0xff && tuple[1] == 0xfe && tuple[2] == 0 && tuple[3] == 0 && endian != BigEndianness) {
+ endian = LittleEndianness;
num = 0;
continue;
- } else if (tuple[0] == 0 && tuple[1] == 0 && tuple[2] == 0xfe && tuple[3] == 0xff && endian != LE) {
- endian = BE;
+ } else if (tuple[0] == 0 && tuple[1] == 0 && tuple[2] == 0xfe && tuple[3] == 0xff && endian != LittleEndianness) {
+ endian = BigEndianness;
num = 0;
continue;
} else if (QSysInfo::ByteOrder == QSysInfo::BigEndian) {
- endian = BE;
+ endian = BigEndianness;
} else {
- endian = LE;
+ endian = LittleEndianness;
}
}
- } else if (((endian == BE) ? qFromBigEndian<quint32>(tuple) : qFromLittleEndian<quint32>(tuple)) == QChar::ByteOrderMark) {
+ } else if (((endian == BigEndianness) ? qFromBigEndian<quint32>(tuple) : qFromLittleEndian<quint32>(tuple)) == QChar::ByteOrderMark) {
num = 0;
continue;
}
}
- uint code = (endian == BE) ? qFromBigEndian<quint32>(tuple) : qFromLittleEndian<quint32>(tuple);
+ uint code = (endian == BigEndianness) ? qFromBigEndian<quint32>(tuple) : qFromLittleEndian<quint32>(tuple);
if (code >= 0x10000) {
*qch++ = QChar::highSurrogate(code);
*qch++ = QChar::lowSurrogate(code);
@@ -574,10 +485,10 @@ QString QUtf32Codec::convertToUnicode(const char *chars, int len, ConverterState
}
}
result.truncate(qch - result.unicode());
-
+
if (state) {
if (headerdone)
- state->flags |= IgnoreHeader;
+ state->flags |= QTextCodec::IgnoreHeader;
state->state_data[Endian] = endian;
state->remainingChars = num;
memcpy(&state->state_data[Data], tuple, 4);
@@ -585,6 +496,113 @@ QString QUtf32Codec::convertToUnicode(const char *chars, int len, ConverterState
return result;
}
+
+#ifndef QT_NO_TEXTCODEC
+
+QUtf8Codec::~QUtf8Codec()
+{
+}
+
+QByteArray QUtf8Codec::convertFromUnicode(const QChar *uc, int len, ConverterState *state) const
+{
+ return QUtf8::convertFromUnicode(uc, len, state);
+}
+
+void QUtf8Codec::convertToUnicode(QString *target, const char *chars, int len, ConverterState *state) const
+{
+ *target += QUtf8::convertToUnicode(chars, len, state);
+}
+
+QString QUtf8Codec::convertToUnicode(const char *chars, int len, ConverterState *state) const
+{
+ return QUtf8::convertToUnicode(chars, len, state);
+}
+
+QByteArray QUtf8Codec::name() const
+{
+ return "UTF-8";
+}
+
+int QUtf8Codec::mibEnum() const
+{
+ return 106;
+}
+
+QUtf16Codec::~QUtf16Codec()
+{
+}
+
+QByteArray QUtf16Codec::convertFromUnicode(const QChar *uc, int len, ConverterState *state) const
+{
+ return QUtf16::convertFromUnicode(uc, len, state, e);
+}
+
+QString QUtf16Codec::convertToUnicode(const char *chars, int len, ConverterState *state) const
+{
+ return QUtf16::convertToUnicode(chars, len, state, e);
+}
+
+int QUtf16Codec::mibEnum() const
+{
+ return 1015;
+}
+
+QByteArray QUtf16Codec::name() const
+{
+ return "UTF-16";
+}
+
+QList<QByteArray> QUtf16Codec::aliases() const
+{
+ return QList<QByteArray>();
+}
+
+int QUtf16BECodec::mibEnum() const
+{
+ return 1013;
+}
+
+QByteArray QUtf16BECodec::name() const
+{
+ return "UTF-16BE";
+}
+
+QList<QByteArray> QUtf16BECodec::aliases() const
+{
+ QList<QByteArray> list;
+ return list;
+}
+
+int QUtf16LECodec::mibEnum() const
+{
+ return 1014;
+}
+
+QByteArray QUtf16LECodec::name() const
+{
+ return "UTF-16LE";
+}
+
+QList<QByteArray> QUtf16LECodec::aliases() const
+{
+ QList<QByteArray> list;
+ return list;
+}
+
+QUtf32Codec::~QUtf32Codec()
+{
+}
+
+QByteArray QUtf32Codec::convertFromUnicode(const QChar *uc, int len, ConverterState *state) const
+{
+ return QUtf32::convertFromUnicode(uc, len, state, e);
+}
+
+QString QUtf32Codec::convertToUnicode(const char *chars, int len, ConverterState *state) const
+{
+ return QUtf32::convertToUnicode(chars, len, state, e);
+}
+
int QUtf32Codec::mibEnum() const
{
return 1017;
@@ -633,7 +651,6 @@ QList<QByteArray> QUtf32LECodec::aliases() const
return list;
}
+#endif //QT_NO_TEXTCODEC
QT_END_NAMESPACE
-
-#endif //QT_NO_TEXTCODEC
diff --git a/src/corelib/codecs/qutfcodec_p.h b/src/corelib/codecs/qutfcodec_p.h
index 749f5be..4f8f92e 100644
--- a/src/corelib/codecs/qutfcodec_p.h
+++ b/src/corelib/codecs/qutfcodec_p.h
@@ -54,9 +54,35 @@
//
#include "QtCore/qtextcodec.h"
+#include "private/qtextcodec_p.h"
QT_BEGIN_NAMESPACE
+enum DataEndianness
+{
+ DetectEndianness,
+ BigEndianness,
+ LittleEndianness
+};
+
+struct QUtf8
+{
+ static QString convertToUnicode(const char *, int, QTextCodec::ConverterState *);
+ static QByteArray convertFromUnicode(const QChar *, int, QTextCodec::ConverterState *);
+};
+
+struct QUtf16
+{
+ static QString convertToUnicode(const char *, int, QTextCodec::ConverterState *, DataEndianness = DetectEndianness);
+ static QByteArray convertFromUnicode(const QChar *, int, QTextCodec::ConverterState *, DataEndianness = DetectEndianness);
+};
+
+struct QUtf32
+{
+ static QString convertToUnicode(const char *, int, QTextCodec::ConverterState *, DataEndianness = DetectEndianness);
+ static QByteArray convertFromUnicode(const QChar *, int, QTextCodec::ConverterState *, DataEndianness = DetectEndianness);
+};
+
#ifndef QT_NO_TEXTCODEC
class QUtf8Codec : public QTextCodec {
@@ -73,13 +99,8 @@ public:
class QUtf16Codec : public QTextCodec {
protected:
- enum Endianness {
- Detect,
- BE,
- LE
- };
public:
- QUtf16Codec() { e = Detect; }
+ QUtf16Codec() { e = DetectEndianness; }
~QUtf16Codec();
QByteArray name() const;
@@ -90,12 +111,12 @@ public:
QByteArray convertFromUnicode(const QChar *, int, ConverterState *) const;
protected:
- Endianness e;
+ DataEndianness e;
};
class QUtf16BECodec : public QUtf16Codec {
public:
- QUtf16BECodec() : QUtf16Codec() { e = BE; }
+ QUtf16BECodec() : QUtf16Codec() { e = BigEndianness; }
QByteArray name() const;
QList<QByteArray> aliases() const;
int mibEnum() const;
@@ -103,21 +124,15 @@ public:
class QUtf16LECodec : public QUtf16Codec {
public:
- QUtf16LECodec() : QUtf16Codec() { e = LE; }
+ QUtf16LECodec() : QUtf16Codec() { e = LittleEndianness; }
QByteArray name() const;
QList<QByteArray> aliases() const;
int mibEnum() const;
};
class QUtf32Codec : public QTextCodec {
-protected:
- enum Endianness {
- Detect,
- BE,
- LE
- };
public:
- QUtf32Codec() { e = Detect; }
+ QUtf32Codec() { e = DetectEndianness; }
~QUtf32Codec();
QByteArray name() const;
@@ -128,12 +143,12 @@ public:
QByteArray convertFromUnicode(const QChar *, int, ConverterState *) const;
protected:
- Endianness e;
+ DataEndianness e;
};
class QUtf32BECodec : public QUtf32Codec {
public:
- QUtf32BECodec() : QUtf32Codec() { e = BE; }
+ QUtf32BECodec() : QUtf32Codec() { e = BigEndianness; }
QByteArray name() const;
QList<QByteArray> aliases() const;
int mibEnum() const;
@@ -141,7 +156,7 @@ public:
class QUtf32LECodec : public QUtf32Codec {
public:
- QUtf32LECodec() : QUtf32Codec() { e = LE; }
+ QUtf32LECodec() : QUtf32Codec() { e = LittleEndianness; }
QByteArray name() const;
QList<QByteArray> aliases() const;
int mibEnum() const;