| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |\
| |
| |
| |
| |
| |
| | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-water-staging:
Mac: pixmaps dont treat alpha exactly the same as other platforms
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
From fixing qwidget autotest, it seems to be the case that filling
a pixmap with a solid color on mac changes the hasAlpha status. This
is different from X11 (and raster engine). That is, the following
code will produce different result:
QPixmap pixmap(100, 100);
pixmap.setAlphaChannel(QPixmap(100, 100));
qDebug() << "Has alpha after setting channel?" << pixmap.hasAlpha();
pixmap.fill(Qt::red);
qDebug() << "Has alpha after fill?" << pixmap.hasAlpha();
So, this patch remove the possibility of converting an already alpha
supporting pixmap to non-alpha upon fill. NB: somewhat related, the
following code works the same on both x11, raster and mac (prints
'false'), but still seems wrong:
QPixmap pixmap2(pixmap);
qDebug() << "pixmap2 has alpha initially?" << pixmap2.hasAlpha();
Rev-By: Fabien Freling
|
| | | |
|
| |\ \
| |/
|/|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Only test control character handling in X11
Compile when qreal != double
Make application font family names locale sensitive in X11
Rename QTextBlockFormat::AtLeastHeight -> MinimumHeight
Add line spacing to QTextBlockFormat, get/set functions, CSS support
Optimize texture glyph cache
Disable certain GPOS features by default
Move advances for control character in Indic syllable shaping
|
| | |\
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Only test control character handling in X11
Compile when qreal != double
Make application font family names locale sensitive in X11
Rename QTextBlockFormat::AtLeastHeight -> MinimumHeight
Add line spacing to QTextBlockFormat, get/set functions, CSS support
Optimize texture glyph cache
Disable certain GPOS features by default
Move advances for control character in Indic syllable shaping
|
| | | |\
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Compile when qreal != double
Make application font family names locale sensitive in X11
Rename QTextBlockFormat::AtLeastHeight -> MinimumHeight
Add line spacing to QTextBlockFormat, get/set functions, CSS support
Optimize texture glyph cache
Disable certain GPOS features by default
Move advances for control character in Indic syllable shaping
|
| | | | |\
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Compile when qreal != double
Make application font family names locale sensitive in X11
Rename QTextBlockFormat::AtLeastHeight -> MinimumHeight
Add line spacing to QTextBlockFormat, get/set functions, CSS support
Optimize texture glyph cache
Disable certain GPOS features by default
Move advances for control character in Indic syllable shaping
|
| | | | | |\ |
|
| | | | | |\ \ |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Compile fix for devices where qreals are floats.
Reviewed-by: Fabien Freling
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
So that they will match the family names returned by
QFontDatabase::families. Because the family names returned
by FcFreeTypeQueryFace are not sorted with locale as the
names returned by FcFontList, we have to find out the
family name matching the system language in the former case.
Task-number: QTBUG-14269
Reviewed-by: Eskil
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Make some alterations to merge request 2305:
1. Rename enum for consistency with other APIs in Qt.
2. Correct documentation with actual version number of features.
3. Remove unstandard css syntax in line-height specification.
Reviewed-by: Jiang Jiang
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
The patch adds line spacing options to Qt, as in all modern word
processors. I primarily checked OpenOffice as a reference for how to
position the lines based on the various types of line spacing as well
as where to do the page breaks. The following functions were added:
void QTextBlockFormat::setLineHeight(qreal height, int heightType)
qreal QTextBlockFormat::lineHeight()
int QTextBlockFormat::lineHeightType()
qreal QTextBlockFormat::lineHeight(qreal scriptLineHeight, qreal scaling)
Here are the HeightTypes available:
QTextBlockFormat::SingleHeight
QTextBlockFormat::ProportionalHeight
QTextBlockFormat::FixedHeight
QTextBlockFormat::AtLeastHeight
QTextBlockFormat::LineDistanceHeight
The following CSS syntax was added:
line-height:150%
line-height:40px
line-height:normal
line-height:40al
line-height:40ld
Merge-request: 2305
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
To avoid calling the bounding box function for the same non-printable
characters in every call, we cache them as empty in the coords table
and bail out before painting them. This patch also fixes some possible
reads of uninitialized memory (reference to return value) and reference
to a default constructed object consisting of uninitialized pods.
Reviewed-by: Samuel
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
According to OpenType Layout tag registry, these features should be
turned off by default. Leaving them on will cause some text layout
problems in common UI widgets -- they are usually designed for better
positioning in print publishing or full featured text layout programs,
not for general usage.
There is no way to turn them on in HarfBuzz now, it should be
addressed later. Features like 'valt' are intended for vertical
writing mode are also turned off, since HarfBuzz doesn't support it
properly yet.
Task-number: QTBUG-15754
Reviewed-by: Lars Knoll
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
When handling control characters (ZWJ, ZWNJ) in Indic syllables,
we only moved glyphs and attributes of the item, advances should
also be moved forward, otherwise we will use the advance of ZWJ/
ZWNJ for the part of syllable, which is normally zero, it results
some positioning problems.
Task-number: QTBUG-14204
Reviewed-by: Lars Knoll
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
QHash is slightly faster than QMap and should be preferred where
the key order has no meaning;
take() is faster than value() + remove()
Reviewed-by: Olivier Goffart
Merge-request: 1017
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Merge-request: 906
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
depending on platform, qBinaryFind might be up to 3 times faster than bsearch
or a bit bit slower at the worst case; so why don't simply use qBinaryFind instead of bsearch and be an optimist? ;)
also we know that all color names are in the lower case already and thus
we can avoid lowercasing of them again and again, and again;
the name to find needs to be lowercased separately instead.
some color-by-name-lookup numbers:
before:
0.000737 msecs per iteration (total: 3,684, iterations: 5000000)
after:
0.000154 msecs per iteration (total: 771, iterations: 5000000)
Merge-request: 906
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
warnings
SQLLEN is unnecessarily an unsigned int
Merge-request: 955
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Merge-request: 955
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
on win, when used in FD_SET, FD_ISSET, etc., socketDescriptor assumed to be
SOCKET which is unsigned int
Merge-request: 955
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
sizeof produces signed int result; QT_SOCKOPTLEN_T is int
Merge-request: 955
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
v.s_addr = 0; does the same
Merge-request: 955
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
by setting the proper macro-guards;
regroup macro-guards for better readability (superfluous change, I know);
fix typos in the comments
was not splited into two commits to make it more clear why and how
that warning is fixed now
Merge-request: 955
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
sysV shm isn't used in QSystemSemaphore
Merge-request: 998
Reviewed-by: Olivier Goffart
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
as POSIX man says, if semop() is interrupted by a signal,
it shall return -1 and set errno to EINTR.
in qcore_unix_p.h, we have EINTR_LOOP helper macro exactly for such cases ;)
Task-number: QTBUG-14434
Merge-request: 998
Reviewed-by: Olivier Goffart
|
| |/ / / / / /
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
like how we do that for ISO format
Reviewed-by: ddenis
Merge-request: 2536
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
according to the write(2) docs:
When write requests greater than {PIPE_BUF} bytes to a pipe that
has available space at least 1 byte, if O_NONBLOCK is set,
write transfers what it can and returns the number of bytes written.
When write requests of {PIPE_BUF} or less bytes to a pipe that has no
enough space, or write requests for greater than {PIPE_BUF} bytes
to a pipe that has no space, if O_NONBLOCK is set,
write returns -1 and sets errno to EAGAIN.
Merge-request: 997
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| |/ / / / /
| | | | |
| | | | |
| | | | |
| | | | | |
Merge-request: 997
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
|
| |/ / / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This is mostly the same as for unix.
Merge-request: 2543
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
| | | | |
| | | |
| | | |
| | | | |
Reviewed-by: Bradley T. Hughes
|
| |\ \ \ \
| |_|_|/
|/| | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/oslo-staging-1:
Add /DYNAMICBASE /NXCOMPAT to linker options in Visual Studio 2005 and higher
Fix compilation error on Solaris
Check elapsed time only once
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
d_type in dirent is a non-standard extension to POSIX available on some
platforms (namely BSDs, Mac, Linux, Symbian).
Taking the conservative approach and using this information only on
platforms known to support it.
Reviewed-by: Shane Kearns
|
| | | | |
| | | |
| | | |
| | | |
| | | | |
Task-number: QTBUG-16262
Reviewed-by: Bradley T. Hughes
|
| |\ \ \ \
| |/ / /
|/| | | |
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
QImage::fill(Qt::transparent) pre-4.8 is a bug, also using Format_ARGB32
is very unoptimal compared to Format_ARGB32_Premultiplied.
Task-number: QTBUG-16439
Reviewed-by: Eskil Abrahamsen Blomfeldt
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Change 284211ccbd2cbd recently introduced a fix for EGL, to prevent
flushing when nothing has been rendered into the back buffer. However,
the skip should only be done when there's no partial update support in
the window surface. If there is partial update support we can still
flush as usual.
Reviewed-by: Gunnar Sletta
|
| | |\ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/qt-s60-public:
Fix KERN-EXEC 0 panic on exit when bearer is searching for WLANs
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The access point scanner cancels itself in the destructor.
This requires the handle to be valid, but it was closed in the symbian
engine destructor immediately before deleting the AP scanner.
Because of the way symbian active objects work, the crashing function
is only called if there was an asynchronous request in progress.
So it could be missed in cases where the scan completes faster than the
test case.
Task-number: QTBUG-16484
Reviewed-by: Markus Goetz
|
| | |\ \ \ \
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Designer: Block QEvent::WinIdChange.
QUrl::setUrl should call detach
qkeymapper_x11.cpp: fix compilation with LSB and without XKB
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Task-number: QTBUG-16425
Reviewed-by: Gabriel
Reviewed-by: Markus Goetz
|
| | | |/ / /
| | | | |
| | | | |
| | | | |
| | | | | |
Task-number: QTBUG-16312
Reviewed-by: Marius Storm-Olsen
|
| |\ \ \ \ \
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-releng-staging:
Remove unnecessarily frozen functions from QtOpenGL WINSCW DEF file
Updated QTOpenGL DEF files for WINSCW and ARMV5
Updated DEF files for QtOpenGL for WINSCW and ARMV5
|
| | |\ \ \ \ \
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-releng-staging:
Remove unnecessarily frozen functions from QtOpenGL WINSCW DEF file
Updated QTOpenGL DEF files for WINSCW and ARMV5
Updated DEF files for QtOpenGL for WINSCW and ARMV5
|
| | | |\ \ \ \ \ |
|
| | | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
Reviewed-by: TrustMe
|
| | | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
Marked more functions absent
Reviewed-by: TrustMe
|
| | | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
Marked QGLContextResource::value absent
Reviewed-by: TrustMe
|
| |\ \ \ \ \ \ \ \
| |/ / / / / / /
|/| | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
scm.dev.nokia.troll.no:qt/qt-integration into master-integration
* 'qt-master-from-4.7' of scm.dev.nokia.troll.no:qt/qt-integration:
PathView crashed when the path is provided with undefined values.
PathView: update modelCount before attempting to regenerate delegates.
Update docs - calling overloaded functions from QML is now supported
Doc fixes for introduction page and Item docs
|
