| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
the internal ucstricmp and ucstrcmp contains different bugs if the
strings are not 0-terminated, as it is with QStringRef.
- in ucstricmp, even if the pointer are the same, the lenght could
be different
- we used to deference the 'end' pointer, that would be 0 if the
string ends with 0, but we cannot do that in the general case
Task-number: QTBUG-10404
Reviewed-by: Denis
|
| |
|
|
|
| |
Task-number: QTBUG-11039
Reviewed-by: Thierry
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Details:
Fixes http://bugreports.qt.nokia.com/browse/QTBUG-10596
Patch taken from
https://bugzilla.novell.com/show_bug.cgi?id=374066
https://bugzillafiles.novell.org/attachment.cgi?id=232442
and
http://code.google.com/p/gentoo-taiwan/source/browse/trunk/x11-libs/qt-gui/files/synthetic-bold-4.5.diff?spec=svn298&r=298
and slightly edited.
Moving the two lines concerning the underlining is unrelated to the
emboldening but seems to make sense anyway, the comment "//underline
metrics" seems to be misplaced before the "if (FT_IS_SCALABLE(face)) {|
The hunk:
@@ -789,6 +794,7 @@ QFontEngineFT::Glyph *QFontEngineFT::loadGlyphMetrics(QGlyphSet *set, uint glyph
}
FT_GlyphSlot slot = face->glyph;
+ if (embolden) FT_GlyphSlot_Embolden(slot);
int left = slot->metrics.horiBearingX;
int right = slot->metrics.horiBearingX + slot->metrics.width;
int top = slot->metrics.horiBearingY;
seems to be logical but it is apparently not used when running
my test program, i.e. synthetic emboldening in my test program
works even without this hunk.
Merge-request: 635
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
|
| |
|
|
|
|
| |
The function is marked internal for now.
Reviewed-by: Gunnar
|
| |\
| |
| |
| |
| |
| |
| |
| | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Fix compilation on WinXP MinGW32;
Add a new qconfig feature GESTURES
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Without this change, I get an error when building plug QODBC:
drivers\odbc\qsql_odbc.cpp: In function 'QString qODBCWarn(const QODBCPrivate*, int*)':
drivers\odbc\qsql_odbc.cpp:264: error: 'class QStringBuilder<QStringBuilder<QStringBuilder<QStringBuilder<QString, QLatin1Char>, QString>, QLatin1Char>, QString>' has no member named 'simplified'
drivers\odbc\qsql_odbc.cpp: In function 'QString qODBCWarn(const QODBCDriverPrivate*, int*)':
drivers\odbc\qsql_odbc.cpp:270: error: 'class QStringBuilder<QStringBuilder<QString, QLatin1Char>, QString>' has no member named 'simplified'
drivers\odbc\qsql_odbc.cpp: In function 'QString qGetStringData(void*, int, int, bool)':
drivers\odbc\qsql_odbc.cpp:403: warning: comparison between signed and unsigned integer expressions
drivers\odbc\qsql_odbc.cpp:444: warning: comparison between signed and unsigned integer expressions
Env:
Windows XP SP2
MinGW (GCC 4.4.0 from trolls ftp)
Merge-request: 2402
Reviewed-by: Andreas Aardal Hanssen <andreas.aardal.hanssen@nokia.com>
|
| | |
| |
| |
| |
| | |
Merge-request: 535
Reviewed-by: Andreas Aardal Hanssen <andreas.aardal.hanssen@nokia.com>
|
| |\ \
| |/
|/|
| |
| |
| | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Introduced getter and setter for the visual X cursor position.
Defines whether the cursor should keep its current position.
|
| | |\
| | |
| | |
| | |
| | |
| | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Introduced getter and setter for the visual X cursor position.
Defines whether the cursor should keep its current position.
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The mechanism allows the cursor to move up and down on a visually
straight line with proportional fonts, and to gently "jump"
over short lines.
Reviewed-by: mae
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Defines whether the cursor should keep its position when
text gets inserted at the current position of the cursor.
For example, we don't want QTextCursor to extend the selection
when inserting characters at the end of an extra selection
representing a user-type but we do want the selection to grow
when rewriting a region of code (e.g. when using our QuickFix engine).
Task-number: QTBUG-11075
Reviewed-by: mae
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Fixes one minor regression in the last commit to QGraphicsIten::scroll:
if the expose region doesn't intersect with the cache pixmap, then
there's no point in removing it.
Reviewed-by: Alexis Menard <alexis.menard@nokia.com>
|
| |\ \ \
| |/ / |
|
| | |\ \
| | |/
| |/|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Doc: Fixed a typo
Doc: QAbstractItemModel: note unexpected behavior with beginMoveRows
Added environment variable graphicssystem switch.
Fixed a problem where QPixmaps where re-bound in non-sharing contexts.
Fix bug in QGraphicsItem::scroll() when called with no QRectF argument.
Make sure we don't pull inn /usr/X11 stuff in the qws mkspek conf
Avoid unnecessary detach / deep copy in QGraphicsItem::scroll().
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The meaning of the destinationChild parameter differs when moving
down in the same parent, contrary to statements by the previous docs.
Reviewed-by: Olivier Goffart
Merge-Request: 600
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Add environment variable switch for graphicssystem so a default can be set
at runtime without passing a command-line parameter.
Merge-request: 622
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
If you have two non-sharing GL contexts and try to draw the same
QPixmap in both of them, they will wind up competing for a spot in the
texture cache. Make the context group a part of the cache key to avoid
that.
Task-number: QT-3462
Reviewed-by: Kim
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
With a null rect argument, QGraphicsItem::scroll() is supposed to scroll
the whole item. When using ItemCoordinateCache, currently the only
supported mode for scroll optimization, we simple scrolled the whole
cache pixmap. Problem: The cache pixmap has a border of 2 pixels. So we
scroll the contents _and_ the border. This leaves white/transparent
horizontal and vertical line artifacts when scrolling.
This change unifies the two cases - partial and full scrolling - into
one (shorter) approach that works without scrolling the margin as well.
Reviewed-by: Alexis Menard <alexis.menard@nokia.com>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
QGraphicsItem::scroll() scrolls the cache pixmap when cacheMode is
enabled (for ItemCoordinateCache only). Because the pixmap exists both
in the cache and in a temp QPixmap copy, the ref count is 2, so the
scroll operation has to do a deep copy. To avoid this, we remove the
pixmap from the cache before and reinsert it again after calling
QPixmap::scroll().
Reviewed-by: Alexis Menard <alexis.menard@nokia.com>
|
| | | | |
|
| | | |
| | |
| | |
| | | |
Reviewed-by: Martin Jones
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
and QGraphicsWidgets.
For QGraphicsWidgets use palette as per QWidget. For QML items
leave the color unset so that the default pen is used.
Task-number: QTBUG-10997
Reviewed-by: Sami Merila
|
| | |\ \ |
|
| | | |\ \ |
|
| | | | |\ \
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Fix example code and image
Fix doc
Fix some bugs in the QML Web Browser buttons
Fix zoom bug in QML Web Browser demo
Fix Typo
Various doc fixes and improvements
Document issues with rectangle border width of 1 where clipping is used
Simplify selection setting. Make TextInput more like TextEdit.
|
| | | | | |\ \ |
|
| | | | | | |\ \
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Various doc fixes and improvements
Document issues with rectangle border width of 1 where clipping is used
Simplify selection setting. Make TextInput more like TextEdit.
|
| | | | | | |/ /
| | | | |/| | |
|
| | | | | | | | |
|
| | | | | | | | |
|
| | | | |\ \ \ \
| | | | |_|_|/
| | | |/| | |
| | | | | | |
| | | | | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-webkit:
Updated WebKit to c58dc2f491a824ac56e31c440fcf7fe16dab09c4
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Integrate https://trac.webkit.org/changeset/60435 - [Qt] Escape backslashes in the .pro files by Ossi.
|
| | | | |_|/ /
| | |/| | |
| | | | | |
| | | | | | |
Task-number: QTBUG-11054
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
The view position is reset to 0 after clear()ing, but this results
in refill() being called, effectively undoing the clear(). Fix is to make
the model invalid before resetting position.
Task-number: QTBUG-11105
|
| | |\ \ \ \ \
| | |/ / / / |
|
| | | |\ \ \ \ |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
By making selectionStart/End read-only, and adding adding select().
Task-number: QTBUG-11056
|
| | |\ \ \ \ \ \
| | | |/ / / /
| | |/| | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
QTextControl::updateRequest() with empty rect means update all.
Task-number: QTBUG-11013
|
| | |_|_|/ / /
|/| | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Since the OpenGL2 paint engine supports transforming the prerendered
glyphs rather than rasterizing the glyphs with the transformation
applied, we don't need to recalculate the QStaticText layout whenever
the transformation changes. This means that we can do fast animated
transforms for QStaticText on this paint engine. A quick test yields
something like 100x speed-up on Windows. This also give visually
better results, as we previously would animate the hinting of the
glyphs, thus causing jittering.
The autotest has been updated to reflect the fact that drawText() and
drawStaticText() now go through identical paths on GL, also when
transforms are set on the painter. However, the scale was changed in
one test, because it was so great that drawText() would fall back to
paths. With QStaticText the idea is speed, so you'll get a poor, but
fast result instead, which is better than tricking people.
Reviewed-by: Samuel
|
| |\ \ \ \ \ \
| |_|_|_|_|/
|/| | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2: (21 commits)
Doc: Fixed an off-by-one error in an example.
Fix QT_NO_TEXTHTMLPARSER
Fix QT_NO_DOM
Fix QT_NO_TEXTSTREAM compilation errors.
Fix QT_NO_VALIDATOR compilation.
Doc: Removed a misleading sentence about a class constructor.
Doc: correcting docs in QDrag::exec
Doc: Adding note about QDrag::exec()
Revert behavior of QTextLayout::boundingRect() when line width is set
Fix unreasonably large width of QTextLayout::boundingRect()
Doc: replacing old image
Fix compilation with stricts compilers.
Fix QT_NO_DIRMODEL build error in QtDeclarative.
Revert "Added native key support to QSharedMemory API."
Revert "Improved documentation for QSharedMemory's key/setKey functions."
Fix a simple mistake in QXmlStreamReader::atEnd() docs.
Improved documentation for QSharedMemory's key/setKey functions.
Added native key support to QSharedMemory API.
Make test work with shadow builds again.
Doc: Corrected the documentation about the compression threshold.
...
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Merge-request: 644
Reviewed-by: Andreas Aardal Hanssen <andreas.aardal.hanssen@nokia.com>
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Merge-request: 2395
Reviewed-by: Andreas Aardal Hanssen <andreas.aardal.hanssen@nokia.com>
|
| | |\ \ \ \ \
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Merge-request: 649
Reviewed-by: Andreas Aardal Hanssen <andreas.aardal.hanssen@nokia.com>
|
| | | | | | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Merge-request: 647
Reviewed-by: Andreas Aardal Hanssen <andreas.aardal.hanssen@nokia.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Merge-request: 648
Reviewed-by: Andreas Aardal Hanssen <andreas.aardal.hanssen@nokia.com>
|
| | | | | | | | |
|
| | |\ \ \ \ \ \ |
|
| | | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
In change 817469519a784b1cc84f89cb3cb84f7560874f8e, there was a
behavioral change which can cause regressions, as the bounding rect of
the QTextLayout would previously return the set line width when this
was greater than the calculated natural text width. We revert to this
behavior to avoid regressions and add an autotest for it. When the
line width is not set (and si.width is equal to QFIXED_MAX), then we
will still return the natural text width.
Reviewed-by: Lars
|
