fix utf8 codec

This commit is contained in:
espie 2007-03-31 23:10:18 +00:00
parent a30e3ddd6c
commit 395dac0562
3 changed files with 137 additions and 3 deletions

View File

@ -1,4 +1,4 @@
# $OpenBSD: Makefile,v 1.34 2007/03/26 21:22:13 espie Exp $
# $OpenBSD: Makefile,v 1.35 2007/03/31 23:10:18 espie Exp $
COMMENT-main= "C++ X11 GUI toolkit"
COMMENT-examples= "examples for qt4"
@ -8,13 +8,13 @@ COMMENT-mysql= "MySQL plugin for qt4"
COMMENT-sqlite2= "sqlite2 plugin for qt4"
COMMENT-sqlite= "sqlite plugin for qt4"
PKGNAME= qt4-${VERSION}p0
PKGNAME= qt4-${VERSION}
PKGNAME-mysql= qt4-mysql-${VERSION}
PKGNAME-postgresql= qt4-postgresql-${VERSION}
PKGNAME-examples= qt4-examples-${VERSION}
PKGNAME-sqlite2= qt4-sqlite2-${VERSION}
PKGNAME-sqlite= qt4-sqlite-${VERSION}
PKGNAME-main= qt4-${VERSION}
PKGNAME-main= qt4-${VERSION}p0
FULLPKGNAME-html= qt4-html-${VERSION}
SHARED_LIBS= Qt3Support 7.0 \
QtCore 6.2 \

View File

@ -0,0 +1,92 @@
$OpenBSD: patch-src_corelib_codecs_qutfcodec_cpp,v 1.1 2007/03/31 23:10:18 espie Exp $
--- src/corelib/codecs/qutfcodec.cpp.orig Wed Feb 21 10:58:41 2007
+++ src/corelib/codecs/qutfcodec.cpp Sat Mar 31 15:01:10 2007
@@ -127,15 +127,19 @@ QString QUtf8Codec::convertToUnicode(const char *chars
bool headerdone = false;
QChar replacement = QChar::ReplacementCharacter;
int need = 0;
+ int error = -1;
uint uc = 0;
+ uint min_uc = 0;
if (state) {
if (state->flags & IgnoreHeader)
headerdone = true;
if (state->flags & ConvertInvalidToNull)
replacement = QChar::Null;
need = state->remainingChars;
- if (need)
+ if (need) {
uc = state->state_data[0];
+ min_uc = state->state_data[1];
+ }
}
if (!headerdone && len > 3
&& (uchar)chars[0] == 0xef && (uchar)chars[1] == 0xbb && (uchar)chars[2] == 0xbf) {
@@ -152,7 +156,7 @@ QString QUtf8Codec::convertToUnicode(const char *chars
int invalid = 0;
for (int i=0; i<len; i++) {
- ch = *chars++;
+ ch = chars[i];
if (need) {
if ((ch&0xc0) == 0x80) {
uc = (uc << 6) | (ch & 0x3f);
@@ -163,14 +167,27 @@ QString QUtf8Codec::convertToUnicode(const char *chars
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.size()) {
+ result.resize(where + 2);
+ qch = result.data() + where;
+ }
+
*qch++ = QChar(high);
*qch++ = QChar(low);
+ } else if ((uc < min_uc) || (uc >= 0xd800 && uc <= 0xdfff) || (uc >= 0xfffe)) {
+ // error
+ *qch++ = QChar::ReplacementCharacter;
+ ++invalid;
} else {
*qch++ = uc;
}
}
} else {
// error
+ i = error;
*qch++ = QChar::ReplacementCharacter;
++invalid;
need = 0;
@@ -181,12 +198,22 @@ QString QUtf8Codec::convertToUnicode(const char *chars
} else if ((ch & 0xe0) == 0xc0) {
uc = ch & 0x1f;
need = 1;
+ error = i;
+ min_uc = 0x80;
} else if ((ch & 0xf0) == 0xe0) {
uc = ch & 0x0f;
need = 2;
+ error = i;
+ min_uc = 0x800;
} else if ((ch&0xf8) == 0xf0) {
uc = ch & 0x07;
need = 3;
+ error = i;
+ min_uc = 0x10000;
+ } else {
+ // error
+ *qch++ = QChar::ReplacementCharacter;
+ ++invalid;
}
}
}
@@ -197,6 +224,7 @@ QString QUtf8Codec::convertToUnicode(const char *chars
if (headerdone)
state->flags |= IgnoreHeader;
state->state_data[0] = need ? uc : 0;
+ state->state_data[1] = need ? min_uc : 0;
}
return result;
}

View File

@ -0,0 +1,42 @@
$OpenBSD: patch-src_corelib_tools_qstring_cpp,v 1.1 2007/03/31 23:10:18 espie Exp $
--- src/corelib/tools/qstring.cpp.orig Wed Feb 21 10:58:43 2007
+++ src/corelib/tools/qstring.cpp Sat Mar 31 15:01:10 2007
@@ -3352,6 +3352,7 @@ QString QString::fromUtf8(const char *str, int size)
result.resize(size); // worst case
ushort *qch = result.d->data;
uint uc = 0;
+ uint min_uc = 0;
int need = 0;
int error = -1;
uchar ch;
@@ -3369,6 +3370,12 @@ QString QString::fromUtf8(const char *str, int size)
ushort low = uc%0x400 + 0xdc00;
*qch++ = high;
*qch++ = low;
+ } else if ((uc < min_uc) || (uc >= 0xd800 && uc <= 0xdfff) || (uc >= 0xfffe)) {
+ // overlong seqence, UTF16 surrogate or BOM
+ i = error;
+ qch = addOne(qch, result);
+ *qch++ = 0xdbff;
+ *qch++ = 0xde00 + ((uchar)str[i]);
} else {
*qch++ = uc;
}
@@ -3391,14 +3398,17 @@ QString QString::fromUtf8(const char *str, int size)
uc = ch & 0x1f;
need = 1;
error = i;
+ min_uc = 0x80;
} else if ((ch & 0xf0) == 0xe0) {
uc = ch & 0x0f;
need = 2;
error = i;
+ min_uc = 0x800;
} else if ((ch&0xf8) == 0xf0) {
uc = ch & 0x07;
need = 3;
error = i;
+ min_uc = 0x10000;
} else {
// Error
qch = addOne(qch, result);