| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
Reviewed-by: thiago
(cherry picked from commit 855ce69f17f570bc91e02bfb34d9e3f1699b3b18)
|
|
|
|
|
|
|
|
|
|
| |
Calling setParentItem is causing the previous opacity/visible updates
to be discarded because the dirty flags were not propagated to the new parent.
Also removed some unnecessary 'markDirty' and 'update' calls.
Task-number: QTBUG-6738
Reviewed-by: bnilsen
|
|
|
|
|
|
| |
We are only interested in getting the posted MetaCall events delivered,
so we can get away with sendPostedEvents() instead of processEvents().
Makes the test also pass on Mac g++ carbon 32.
|
|
|
|
| |
Reviewed-by: Trust Me
|
|
|
|
|
|
|
| |
Avoid traversing the whole child hierarchy when opacity changes unless
there are children with graphics effects.
Reviewed-by: Bjørn Erik Nilsen
|
|
|
|
|
|
|
| |
We need to invalidate the graphics source pixmap cache for both child
items and parent items when changing the opacity of a graphics item.
Reviewed-by: Bjørn Erik Nilsen
|
|
|
|
|
|
| |
Simply skip the updates that are outside the replay widget's bounds.
Reviewed-by: Gunnar Sletta
|
|
|
|
|
|
| |
Don't repeatedly update the pen / brush if no pen / brush is set.
Reviewed-by: Gunnar Sletta
|
|
|
|
|
|
|
|
|
| |
That rare case when we are deleting the last span was not being taken care
of. Unbelievable but true. Auto-test included.
Reviewed-by: Thierry
Reviewed-by: leo
Task-number: QTBUG-6004
|
|
|
|
|
|
|
|
|
|
|
| |
These were processed immediately, so there was a fair chance that we
could end up doing a virtual function call on items that were not fully
constructed. This patch is also an optimization, since we never remove
anything from the vector.
Auto-test included.
Reviewed-by: Jan-Arve
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem occured when doing something in the polishEvent() which
eventually ended up as an update(). The problem was that in
QGraphicsScene::addItem we scheduled a polish event after scheduling
an update, resulting in update requests being processed before polish
requests.
Auto-test included.
Task-number: QTBUG-6956
Reviewed-by: alexis
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Provide configure flag to enable/disable the use of DEF files on Symbian.
A useful side-effect was that it cleaned up how we control DEF files from
qbase.pri and in WebKit.
-nokia-developer still disables DEF files, as it triggers the autotest
exports, which are not frozen into the DEF files.
Disabling DEF files means that there is no BC with previously released
versions of Qt, so this should only be used for development purposes.
.pro files can specify custom locations for DEF files by setting
defFilePath.
Task-number: QTBUG-6556
Reviewed-by: Jason Barron
|
|
|
|
| |
Reviewed-by: Trond
|
|
|
|
|
|
| |
We can't call QCOMPARE from within a nested function, because the
return statement will just exit that function.
VERIFY_COLOR can't be turned into a function this way.
|
|
|
|
| |
Reviewed-by: Sami Merila
|
|
|
|
|
|
|
|
| |
Also use HighlightAllOccurrences to have highlighting during type and
search. Some more refactoring.
Task-number: QTBUG-3335
Reviewed-by: ck
|
|
|
|
|
|
|
|
|
|
|
|
| |
There is a lot of code depending on that pixmaps are flipped
upside down in the gl graphicssystem, so toggling this requires
extensive testing. Since we're anyway questioning the relevance
of this feature (compared to raster + GL viewport) its simply not
worth the effort to fix it properly right now.
Revert "Fixed y-inverted pixmaps on N900."
This reverts commit 57473d5d2a7bd6ae3117f61ff29264a1b790bb01.
|
|
|
|
|
|
|
|
| |
Need to bind the PMV matrix's attributes to their indexes in the
simple shader, which is created in a seperate code path to all the
other shaders. This should fix the qgl autotest failures.
Reviewed-By: TrustMe
|
| |
|
|
|
|
|
|
|
| |
If Avkon components support transparency then dialog background is transparent
enabling rounded corners.
Reviewed-by: Sami Merila
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QmlGraphicsItem doesn't need any parent change notifactions so we can
call the helper class (QGraphicsItemPrivate::setParentItemHelper)
direclty from QmlGraphicsItem::setParentItem. This avoids a lot of unnecessary
instructions related to QVariant constructions as well as virtual
function calls.
I've made the variant pointers explicit in the declaration of
setParentItemHelper so that we don't accidentally call setParentItemHelper
from places where we need parent change notifications.
Task-number: QTBUG-6877
Reviewed-by: alexis
|
|
|
|
|
|
|
|
|
| |
We need this change in order to bypass some of the QVariant itemChange
notifications from QGraphicsItem::setParentItem. All this is internal
stuff and we know what we do, so I don't consider the change too ugly.
Task-number: QTBUG-6877
Reviewed-by: alexis
|
|
|
|
|
|
|
|
|
| |
We don't have to do a full blown QGraphicsItem::setFlag call from
QGraphicsItem::setFlags, only to change the ItemStacksBehindParent
bits. We can do it directly (with care).
Task-number: QTBUG-6877
Reviewed-by: alexis
|
|
|
|
|
|
|
|
| |
Make sure we do cheap tests before the more expensive ones.
This function is called from QGraphicsScene::addItem.
Task-number: QTBUG-6877
Reviewed-by: alexis
|
|
|
|
|
|
|
|
|
| |
The biggest optimization here is "updateAncestorFlags()". It's much
faster to update all the flags rather than trying to enable/disable
certain flags according to the current state.
Task-number: QTBUG-6877
Reviewed-by: alexis
|
|
|
|
|
|
| |
This makes GLSL dumps _significantly_ easier to read.
Reviewed-By: TrustMe
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This has several advantages: First, updating an attribute value
seems to be cheaper than updating a uniform. Second, vertex atribute
values are independent of shader program, which means they persist
across changing of the shader program. This makes code simpler and
reduces GL state changes. Note: Credit goes to Samuel for finding
this little gem. :-)
For the 25920 solid QGraphicsRectItem test case, this gives 10%
improvement on desktop and 27% on the SGX.
Reviewed-By: Kim
|
|
|
|
| |
Reviewed-By: Samuel
|
|
|
|
| |
Task-number: QTBUG-7418
|
| |
|
|
|
|
| |
Reviewed-by: Eskil
|
|
|
|
|
|
|
|
|
| |
if the pro file for the translations lives in a sibling tree of the
actual source tree, messages from included headers wouldn't have been
collected, as they were not considered part of the project.
Task-number: QTBUG-7495
(cherry picked from commit f300f5cfdf3b32e687191b071c30e14ac2721fc8)
|
|
|
|
|
| |
this has always been kind of an omission ...
(cherry picked from commit 2a8fd7584590016dbe22fb74f440e71a2578dc81)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
previously, the semantics of nesting projects were pretty bizarre:
the list of TS files grew ever as the subdirs were visited. the
source files otoh were cleared for every project. this meant that
some TS files were updated over and over again, each time with different
data.
this is obviously total nonsense, so there is no compatibilty to keep.
so here come the new semantics:
- each project is scanned separately. it has separate include paths and
may override the inherited CODECFORSRC.
- the messages from all sub-projects are merged
- if a sub-project has a TRANSLATIONS entry, it is "detached" from its
parent project, thus starting an own merged translator. it is also
possible to name an empty set of TS files to simply exclude the
sub-project.
- CODECFORTR can be specified in each project with TRANSLATIONS
- if TS files are specified on the command line, they override
TRANSLATIONS from the top level project and stop processing of
detached projects alltogether. if multiple top-level projects are
specified, they are all merged.
this is somewhat slower, as now includes are re-scanned per project,
while previously all sources were simply accumulated and scanned as one
project. this can be fixed retroactively if needed.
(cherry picked from commit b62b87d6ae3ced4466c1a1b085b51cec757e24f5)
|
|
|
|
| |
(cherry picked from commit df9a6a9bf9d7bf3609ac034e705d076b82572333)
|
|
|
|
|
|
| |
lrelease needs only the TRANSLATIONS value anyway, so there is hardly
anything to share for real
(cherry picked from commit b0a454fd09e0cbf74304aabd160ea25c2c83b4b9)
|
|
|
|
| |
(cherry picked from commit d74f8fd8d7d378bbf23872ddc24ddfbfd948e295)
|
|
|
|
| |
(cherry picked from commit 20c871370727630b83adce1e96ac0437802c433a)
|
|
|
|
|
| |
really no point in doing so ...
(cherry picked from commit d97a3e97f54d1e1d028fd7bc2b46171c6b0be966)
|
|
|
|
|
|
|
|
|
|
|
|
| |
Patch_capabilities.pl script now also patches any embedded sis
file entries in pkg to embed corresponding *_selfsigned.sis instead.
Also the script now removes dependencies to other packages from
patched pkg to reduce unnecessary warnings about missing dependencies
when said dependencies were installed via patched sises.
Reviewed-by: Janne Koskinen
(cherry picked from commit 0b2bde7af1ec6cec5c1bc674c9049b12fe0deb20)
|
|
|
|
|
| |
Reviewed-by: TrustMe
(cherry picked from commit c513782f7cbad45b7c91e069ae81552a4bbb2a5f)
|
|
|
|
|
|
|
|
|
|
| |
QS60Style does not regard layout orientation when calculating subrects
for QComboBox. This leads to a situation where button is not visible
on UI with RightToLeft orientation.
Task-number: QTBUG-7584
Reviewed-by: Alessandro Portale
(cherry picked from commit 78bb23107d9fcde431cd2ba83c1a9e9886b6d51a)
|
|
|
|
|
|
|
|
|
|
| |
When the line edit was refactored into a line control this regression
was introduced. This regression was introduced by change
fb7d86cf23227302d48db279ec589221d11a1f6a.
Task-number: QTBUG-4789
Reviewed-by: Alan Alpert
(cherry picked from commit 06ee48f7ad7f5ab8bdbaed765d301fa2c0b417fb)
|
|
|
|
|
|
|
|
|
|
| |
Modal Mac/Carbon dialogs do not have the close, minimize and resize
window title buttons enabled, which makes it very hard to close modal
dialogs.
Task-number: QTBUG-7481
Reviewed-by: Kim
(cherry picked from commit c8f4319b3a2ddacacd3bca67861a41e22dd1ada0)
|
|
|
|
| |
(cherry picked from commit 03c77ad02d5a3c69f4edfdfd2f3ef7cd7dc67aaa)
|
|
|
|
|
|
| |
Task-number: QTBUG-7530
Reviewed-by: Kim
(cherry picked from commit 0a3ce6fa04ed2b085fe72b89d302a3a5f9e82948)
|
|
|
|
|
|
|
| |
This is a very operation in DirectFB and saves a fair bit of overhead.
Reviewed-by: Donald Carr <donald.carr@nokia.com>
(cherry picked from commit e5e6d9e7fe8f5c3baa5caf609198db1184f59ef9)
|
|
|
|
|
|
|
|
| |
Added missing #include.
Task-number: QTBUG-6371
Reviewed-by: Alessandro Portale
(cherry picked from commit 8c29bbe3dcd192024d2dbfcf2faa72077b77107d)
|
|
|
|
|
|
|
|
|
|
| |
If long path (or any other long string) is added to a combobox, it
grows outside of screen area. QS60Style needs to check that widget
will not grow wider than screen area.
Task-number: QTBUG-6371
Reviewed-by: Alessandro Portale
(cherry picked from commit e571e0f012e3e27ffff56493755999da1960318c)
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently only keypad navigation focus is drawn to RadioButtons and
CheckBoxes in non-touch devices. However, in S60 the native widgets
have themed highlight (similar to lists). This task will add that
highlight support for these widgets regardless if keypad navigation
is active or not.
Task-number: QTBUG-5536
Reviewed-by: Alessandro Portale
(cherry picked from commit ce64d3c7e0b93fa6792513f804b42d31acccc68e)
|