This is faster because it just swaps the pointers stored instead of touching
the refcounts. It's the same as qMove()ing the right-hand-side, but benefits
C++98, too.
Change-Id: I6b7e3d0e5c7eb81f88fe9069d6662335d24aa86c
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Check whether a detach happened, and also check that non-consecutive duplicates
are properly removed.
Change-Id: Iac7a3d8fe5ed5a69a9a0b55ddbf95308f11122c2
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
A function pointer requires aggressive optimization to inline. A function object
otoh is inlined by compilers two decades old.
Even shaves off 544B in text size off of a -O3 GCC 4.7 QtCore build.
Change-Id: Ibfd41654257360fa0f118701f502e6c23a6c28b3
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Don't call that->size() if we have already saved its value in a local temp.
Surprisingly, this saves 56B in text size even on a -O3 GCC build, proving
that the compiler does _not_ have enough context to const-fold the repeated
size() evaluation.
Change-Id: I2c41ddc4376f4e1534fc2f0d4789e42b984d3ba6
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
QStringList << QList<QString> already compiled, but was interpreted as
QStringList << QStringList(QList<QString>), which involves the QList
copy ctor. Adding the overload saves that copy.
Cannot use a using declaration here, since the return type is different.
Change-Id: I119cc98e7e2df24549a1abb158543b729edc30ef
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
The compiler-generated one is just fine, and enables the move constructor
inherited from QList, too.
This is BC and SC since QStringList isn't exported.
Change-Id: Ic49000d9feb66504d7ec7e5d519e7a13ca9e5d9e
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The new function just converts the job-sheets cups option argument,
reducing the number of almost-identical switch statements to one.
Effects on a stripped Linux AMD64 GCC 4.7 -std=c++11 -O2 build:
text: -1008B
data: +-0B
relocs: +-0
Change-Id: I56843fa963e098e5c8ae203584f34bf5798667e6
Reviewed-by: Martin Klapetek <mklapetek@kde.org>
Reviewed-by: John Layt <jlayt@kde.org>
The new function just calculates the job-hold-until cups option argument,
reducing the number of setCupsOption() calls down to one.
Effects on a stripped Linux AMD64 GCC 4.7 c++11 release build:
text: -552B
data: +-0B
relocs: +-0
Change-Id: I4aa4118c3493cd302dd8fd7f4985c7e0f70ef071
Reviewed-by: Martin Klapetek <mklapetek@kde.org>
Reviewed-by: John Layt <jlayt@kde.org>
The new flag ItemContainsChildrenInShape is similar to the existing
flag ItemClipsChildrenToShape. Setting the new flag makes QGraphicsScene
assume that children are drawn within the shape of the current item
but this is not enforced by clipping. When an application manually
ensures this clipping boundary, setting the new flag removes the
overhead of enforcing the clip with ItemClipsChildrenToShape, while
still allowing other routines to behave more optimially by assuming
children are within the shape of the current item.
[ChangeLog][QtWidgets][QGraphicsItem] Added the
ItemContainsChildrenInShape flag that enables using optimizations
of ItemClipsChildrenToShape without the overhead of enforcing the clip.
Change-Id: I5496fe1ca331b77fd51e0df8a3ace2b8e939eaf2
Reviewed-by: Andreas Aardal Hanssen <andreas@hanssen.name>
This update fixes a bug introduced by the extensive changes
for QTBUG-35377. Three node subtypes - Group, Module, and
QML module, were promoted to be first line node types in the
update for QTBUG-35377. This broke the file name construction
routine for those node subtypes, which used to have the DocNode
node type. This caused empty ref attributes to appear for those
keyword elements in the qhp file. The file name construction
routine has now been corrected to account for this.
Task-number: QTBUG-37658
Change-Id: I307979255fdfd48493b3a4cebaf996b2130bc2c7
Reviewed-by: Martin Smith <martin.smith@digia.com>
kThis update fixes a bug introduced by the extensive changes
for QTBUG-35377. The name Qt represents two namespaces, one
in C++ and one in QML. The name "Window" is used in both of
them, so the link \l{Qt::Window} would cause a collision in
the single tree qdoc. In the multiple tree qdoc, there is
no collision, but in this case the link should have gone to
the C++ page and it went to the QML page instead. The fix
involved correcting the way qdoc searches for link targets.
Task-number: QTBUG-37633
Change-Id: Ib9b209eced937a0be0d3299f300ebf22b2776012
Reviewed-by: Martin Smith <martin.smith@digia.com>
This update fixes a bug introduced by the extensive changes
for QTBUG-35377. For a QML base type loaded from an index file,
its QML base type was not being resolved. This resulted in the
"All members" page for some QML types to be incomplete because
the pointer to the base type was 0 when it should have been set.
This change also introduces the concept of "just in time"
resolution for base type pointers, which appears to speed up
qdoc a little.
Task-number: QTBUG-37326
Change-Id: I5f09336ec70ba84029b44b245c56f7f8fe349757
Reviewed-by: Martin Smith <martin.smith@digia.com>
The current child concept was added to the Name
Collision Node to heuristically provide better
linking when a link operation reached a collision
node. It is doubtful that this improved linking
much, but now that qdoc uses multiple trees, it
is much less likely that collision nodes will
occur. In fact, there are none in in the current
Qt5. Therefore, the current child code is hereby
removed from qdoc.
Task-number: QTBUG-37067
Change-Id: I33aea5d550afb7ceaf941d49112e02c21d44f6dc
Reviewed-by: Martin Smith <martin.smith@digia.com>
With this update, qdoc is now ready for testing
with multiple trees. In making this change to using
multiple trees, it has become clear that qdoc does
not really need trees the way it currently uses them.
Each C++ class or namespace, or QML type is naturally
a tree tree structure, but above that level, what we
currently call a tree in qdoc should really be called
a collection of maps. This change has moved qdoc in
that direction. It remains to replace the Tree class
with a class that encapsulates a set of maps, one for
each major node type. That can be implemented later.
Task-number: QTBUG-35377
Change-Id: I39068a0cb26c01f14ec0e4621742d727efb913bf
Reviewed-by: Martin Smith <martin.smith@digia.com>
qdoc now knows how to search the forrest of node
trees in an optimal order. But there remain some
problems with specific searches that cross module
boundaries. These include group membership and C++
and QML module membership, as well ass C++ base
class resolution. Part 3 will be concerned with
fixing these remaining bugs.
With this update, qdoc now takes less time to
generate the docs for Qt 5. Testing indicates
that qdoc run time has dropped from about 14
minutes to about 7.5 minutes on an iMac.
Task-number: QTBUG-35377
Change-Id: I6bded6ef54124b4f6e5914cad4548f0b600209b0
Reviewed-by: Martin Smith <martin.smith@digia.com>
qdoc now builds a separate Node Tree for each index
file it parsed. The main Node Tree now contains only
the Nodes of things being documented in the current
module. This should make qdoc run a little faster.
qdoc now uses these separate trees to make intra-module
and inter-module linking more robust by searching the
trees in an order that depends on the type of link it
is searching for. The tree for the current module is
always searched first. Then qdoc searches the trees
for either the C++ modules or the QML modules, depending
on whether it is looking for a C++ link or a QML link.
In preparation for this update, qdoc was also simplified
a lot. Many functions became obsolete and were removed.
Others were combined.
Task-number: QTBUG-35377
Change-Id: Iea4e49869ff6a6ff0f4d53090728770d40d892f3
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
WinRT distinguishes between regular capabilities and device
capabilities. For now the latter section are location, microphone,
proximity and webcam. Hence we must add those properly to the generated
manifest.
However, Windows Phone currently combines all of these into the
Capability section, so add a warning if someone uses devicecapabilities
for Windows Phone.
Task-number: QTBUG-37932
Change-Id: I8e9550f29b6afdea3737cc85bdc68344fc04223d
Reviewed-by: Andrew Knight <andrew.knight@digia.com>
The following adjustments are made:
- A precompiled path, defaulting to a QRC path, is added for looking
up precompiled shaders
- The standard service source/binary paths are created if needed, in
order to avoid fast-fails when the service cannot create the structure
itself fast enough.
Change-Id: I966e54c0b35bafdaf0b3a32b76eb896308aca6db
Reviewed-by: Oliver Wolff <oliver.wolff@digia.com>
Some data was never serialized (styleName, hintingPreference) and
some was incorrectly serialized (styleStrategy). This change also adds
auto tests for every applicable QDataStream version.
[ChangeLog][QtGui][QFont] QFont now serializes its data properly,
without any data loss.
Task-number: QTBUG-22814
Task-number: QTBUG-22946
Change-Id: I34e61b10662b7ad6c57054dacc7e1f522f5b5c5d
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
Otherwise, touch update/end events were ignored and the gestures
were never triggered (unless a native widget was used).
Task-number: QTBUG-37759
Change-Id: I4b0145ffe535065e1458164b67bc8d9e33a4af97
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
This finishes 800232e1d3, which encovered
a tricky issue in QFontEngineMultiQPA::createMultiFontEngine().
Task-number: QTBUG-37836
Change-Id: I6d432e09e755f5d9ded09752c4c092f4857ad224
Reviewed-by: Pierre Rossi <pierre.rossi@gmail.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Just a C&P typo, of course we have to compare lists, not their sizes.
Change-Id: I40542035b87f5bb8d75207cb02c0826cc3a2a413
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
The QFontEngineMulti's `base` font engine always non-null.
Change-Id: I5092b66bc839a6a3216ffa0b0afe55a8b62be620
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
* Report correct pre-edit information when calling updateSelection()
on endBatchEdit()
* Fix getExtractedText() to report the correct offset and cursor position,
* Fix setSelection() to use the correct cursor position.
Task-number: QTBUG-35689
Change-Id: I7e8427d0f5a18abf18ba2faf9d510756ddf6044b
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
The current implementation of QOpenGLContext sharing assumes that the
contexts form a tree and that leaf-nodes are destroyed before their
parents.
We build on this assumption and keep track of the default FBOs for
windows in the root context of the tree. This allows two shared
contexts to both makeCurrent() on the same window surface without
resulting in two FBOs being set up (which doesn't work on iOS due
to the CEAGLLayer already being tied to another render-buffer).
Change-Id: Ib9f8c597effe488480fe99e10846be22c257f490
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
It also has a broken declaration of posix_memalign
Change-Id: Ie8f245564f80b04901425729b46953828204efaf
Reviewed-by: Andy Nichols <andy.nichols@digia.com>
This patch aims to provide an updated test that follows changes started
in 10.7: new rule is that only root can access SystemScope settings. It
also disables the sync() workaround code path which is at least not
executed during the tst_QSettings execution and returns wrong value to
the test.
From Apple's documentation:
"Note that modification of some preferences domains (those not
belonging to the “Current User”) requires root privileges (or Admin
privileges prior to OS X v10.6)—see Authorization Services Programming
Guide for information on how to gain suitable privileges"
https://developer.apple.com/library/mac/documentation/CoreFoundation/Reference/CFPreferencesUtils/Reference/reference.html
[ChangeLog][QtCore][QSettings] QSettings now returns the correct value
for isWritable() when using SystemScope settings.
Task-number: QTBUG-9824
Task-number: QTBUG-21062
Task-number: QTBUG-22745
Change-Id: Ib6a1490ec596b99d189ec4de9a0f28ecfd684172
Reviewed-by: Liang Qi <liang.qi@digia.com>
Basically a copy of the inline documentation inside the prf file.
Task-number: QTBUG-37788
Change-Id: Ie6df1597297223be778c748aad525f5521232cb7
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@digia.com>
In some circumstances NSScreen screens can be empty yet there is still a
valid main screen so we prevent any problems arising by ensuring we work
with the main screen in that instance.
Task-number: QTBUG-37876
Change-Id: I5827009a9894eb8c1d4f5addc6c6f59e1c50c3d0
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
The current logic requires primary->output to not be specified, or
match the output number. This fails when the output by that number
doesn't currently have a display attached, as non-connected outputs
are skipped. This means setting the display to :0.1 for Qt5 will put
the window on :0.0 and it can't be moved to :0.1 after the fact. The
solution is to pick the first output on the requested screen, then
override if an output exactly matches the primary output value. That
way it will at least be on the requested screen.
Tested with Xephyr, ATI, and nVidia (proprietary driver), where the
nVidia had the primary output disconnected.
Task-number: QTBUG-15418
Change-Id: I7951dcef74cb135dc12a6e133a71c5852eae9778
Reviewed-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Asserting is only relevant for debug builds, and there were a couple
of other places that a segfault would typically occur before the assert
had a chance.
Change-Id: I1abc82eb3ecfa91050117fab1525f4cbd82ff486
Task-number: QTBUG-37876
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
I found a useless else-if statement in qwidgettextcontrol.cpp.
It is duplicating.Needless to say,it should be removed.
Change-Id: Ic70b995196f7199e4aa5263a4937c1678b130357
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
In tst_QEvent, add a check that exhausts the available event type ids and verifies
that registerEventTypes() returns -1 in that case, as documented.
I haven't found a way to test the other case in which -1 is returned.
Since this test operates on a global write-only registry the new test case needs
to keep track of whether the earlier test cases have run successfully. If they
didn't, skip this test case.
Change-Id: I68ea9d17d10dcec22175994aba269dd09c9adf43
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
A test for a hint > QEvent::MaxUser was missing.
Added.
Change-Id: I00c1c45f6673f35d4203df15b4ffc35f57273e46
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Apparently, the function before returned const-&, but now returns by-value.
Use the member data variable directly, the function is const anyway.
Change-Id: I324bc675fd7c9b5bfb28c475eba0c0275379132c
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
> error: 'REFKNOWNFOLDERID' does not name a type
> error: 'GetKnownFolderPath' does not name a type
> error: 'SHGetKnownFolderPath' was not declared in this scope
REFKNOWNFOLDERID is a typedef for `const KNOWNFOLDERID &` and
KNOWNFOLDERID is a typedef for `GUID`.
Change-Id: I2371f49aba6e20b0bd44aad423dd822e5ee152d4
Reviewed-by: Samuel Gaist <samuel.gaist@edeltech.ch>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
The native PDF strokes can not be used with constant opacity as any
overlapping parts of the path will end up being drawn multiple times
getting the wrong final opacity. A constant opacity should therefore
be treated the same as opacity on the pen color, and trigger the
fallback drawing.
Task-number: QTBUG-37101
Change-Id: I44f70a8b494567ee89804a5fa6333304a5abe91e
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@jollamobile.com>
While the user is touching the screen, postpone scrolling
until we get a touch release. Scrolling in the middle of
a touch sequence will change the coordinates under the
touch, and cause some artefacts.
Change-Id: I02ef420abaab780a459f014d4b4cfb75c8fbb725
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
On iOS we have set the style hint 'SetFocusOnTouchRelease'. This is in
conflict with the 'hide keyboard' gesture, since a control can
try to regain focus (and open the keyboard) if the gesture ends
on top of it. So we need some extra work-around code to prevent this
from happening.
The correct way would probably be to cancel the touch sequence once
the gesture triggers, but this is not well implemented in Qt yet,
especially in combination with widgets and mouse synthesis.
Since usage of the gesture behaves really bad in some cases (e.g
if using the TextEdit example) we need to apply this for now.
Change-Id: Ib3327c0bd94d722b4c4793bc6d152d6d19810e4b
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
If you show/hide the keyboard quickly while we scroll the
screen, the scroll will appear to jump. The reason is that
the screen animation will start from where the model layer
is at, and not the presentation layer. So specify that
the animation should start from the current state of the
presentation layer.
Change-Id: I3db87ab11aab583eb50784b0c0a03a9a07c8b822
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
If you resign/become first responder several times
while the keyboard is animating (e.g changing focus between
focus objects while the keyboard is animating), iOS will
short-cut the whole animation, and jump directly to
keyboard end-state. For that reason, we always need to handle
keyboardRectChanged, and not bail out early. This is
fine, since the guard we had was really only meant for
keyboardWillShow/Hide in the first place.
Change-Id: I3a3d1e7061962286c538360029ed38410dc0f347
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
