the commit 53a420a4d1 merged translate()
functions. The simple versions need to be merged too.
Change-Id: Ie873483beb8ed0b911ae0568e97b427f4c6b74e3
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
This is for behaviour consistency with Qt on unix, as well as the
socket close issues described in the task.
Task-number: QTBUG-4465
Change-Id: Ida95650d8a9bd7b5bc3d3926d22e20a6d7eeb30b
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
For high resolution or extremely large font sizes, the advance
cached here is likely to overflow, since FreeType returns 26.6
fixed point value and we only take signed char here for advance.
In those cases we should skip caching because there won't be
that many big glyphs after all.
Also move the metrics caching block a bit down to take glyph
embolden and oblique into account.
As a result we also don't need to increase the linearAdvance
size because any linearAdvance less than 128 should fit in
the old 10.6 fixed format.
Change-Id: Ic4920ada49954ce1e0a8673c9f33f30e385e3046
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
This allows QML to emit signals without having to built lazily created
metaobjects.
Change-Id: If8068fb3cb35d79dc8f3ef79253d9c2eb7c93205
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Reviewed-by: Kent Hansen <kent.hansen@nokia.com>
This is slightly faster, and also avoids accessing the object's meta
object that would cause any lazily created meta objects from being built.
Change-Id: I0a78e09511c120bdbe707a1efc91ba480ab7680c
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Reviewed-by: Kent Hansen <kent.hansen@nokia.com>
This change allows us to delay creating the actual meta object
until it is actually required.
Change-Id: I1c4a4226bd82fa606b206dd60322f49b49c32463
Reviewed-by: Kent Hansen <kent.hansen@nokia.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
We no longer have code doing MMX or 3dNow! on Qt 5, so nothing defines
those macros. If nothing defines them, we don't need to undef
them. Note that the SSE case is a mislabel: the old "SSE" code on Qt 4
wasn't real SSE, it was MMX with some instructions added at the same
time as SSE was added. It's MMX that Windows 64 doesn't support.
As for QT_HAVE_ARMV6, this macro is replaced by the sub-arch detection
in qprocessordetection.h.
Change-Id: Ic3b00e1533e6b4cea32ba7824233de0a5c0fb32b
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
In at least one case found in our testfarm, the compiler runs into an
ICE (Internal Compiler Error) compiling the new AVX code. The error it
reported was:
fatal error: error in backend: Cannot select: 0x7fbf1aa42210: f64 = sint_to_fp 0x7fbf1aa88a10 [ORD=1936] [ID=37]
0x7fbf1aa88a10: i32,ch = CopyFromReg 0x7fbf19538768, 0x7fbf1a9f2610 [ORD=1936] [ID=27]
0x7fbf1a9f2610: i32 = Register %vreg38 [ORD=1936] [ID=9]
As is the nature of ICEs, juggling the code around will probably make
it pass. But since I have plenty more AVX changes pending, which make
the code even more complex, it's also very likely that this issue will
happen in other places. For that reason, I choose to blacklist the
compiler instead.
Change-Id: Ide3201f0cc49c7ceb63e966f6de65a8315cbea4b
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
The CI system is now using the parallel_test flag to run tests in
parallel. This test has become flaky, or at least more flaky than it
was previously. Mark it to no longer run in parallel.
Change-Id: I47bca3be620a8f648a0eb9c9b9f26d2d925efc01
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
-O2 is already the default optimization level for release builds,
as defined in gcc-base.conf. In addition, it shouldn't be set for
debug builds.
Change-Id: Idd7406b0d135d9579676b389050fd6e5776e722c
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
Compiler flags like CPU architecture and FPU should be set on
QMAKE_CFLAGS instead of QMAKE_CFLAGS_RELEASE, as the latter only
applies to release builds
Change-Id: I2e729a9e413934e904fc2810394e118940b8557f
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
The atlas it assumed to have 8 cursors per row. All cursors have
to be square.
Change-Id: I7ffbad4662be450b146f84032bb26187894d528f
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Presumably, this was set for QSound, which has since moved to QtMultimedia (and
no longer uses nas, anyway).
Configure also no longer seems to have logic for setting nas, as of
4535913c4f.
Change-Id: Ie5b351844bd169ad0548b0d29513adbf6f5d9a12
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Donald Carr <donald.carr@nokia.com>
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
This significantly reduces the size of the generated code
in places where we don't need exceptions.
The -(no-)exceptions configure flag has been removed in the
process, as there is now a fine grained way to control this
on a per module level, and Qt is being compiled without
exceptions in most places.
Change-Id: I99a15c5d03339db1fbffd4987935d0d671cdbc32
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QTextLine::cursorToX returned the line width for cursor positions
outside the width of a wrapped right to left line because the
leading space width was always calculated as 0.
Returning a non-zero width for the leading space does cause
problems for other uses of QTextEngine::alignLine() though
as the textAdvance already doesn't include the leading/trailing
space so subtracting it there double accounts for it.
Task-number: QTBUG-24801
Change-Id: I56cbb139814c32813bebb49de8c045b29154a958
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
This change will generate working links between all modules in qtbase.
Some testing needs to be done on the other modules.
Change-Id: Ic65a9c753f891ac51427ca7c1cdcab13611d2f5b
Reviewed-by: Martin Smith <martin.smith@nokia.com>
Maintain the consistency of QWizardPrivate's two members:
QVector<QWizardField> fields;
QMap<QString, int> fieldIndexMap;
during and after calls to QWizardPrivate's
void _q_handleFieldObjectDestroyed(QObject *)
member function. The failure to maintain this consistency
caused an out of bounds access and core dump in
QWizard's field(const QString &name) member function.
QWizard's field(const QString &name) member function expects
the values in the QMap fieldIndexMap to be indexes into the
QVector fields. Prior to this change
_q_handleFieldObjectDestroyed only removed the appropriate
entry from the map and erased it from the vector. It did
not decrement by one all the indexes greater than the index
that was removed from the map and erased from the vector
in the rest of the map.
For example ...
So if initially have the following mapping ...
"field0" -> 0,
"field1" -> 1, and
"field2" -> 2
with fields of size 3. After destruction of "field1" have ...
"field0" -> 0, and
"field2" -> 2
with fields of size 2.
Now attempts to look up "field2" using QWizard::field will
have an out of bounds error and possibly core dump or trigger
an internal Qt assert because an attempt to access
this->fields[2] will be made. It should be accessing
this->fields[1], but does not because the map is no longer
consistent with the vector.
This change adds a decrement by one for all the indexes
greater than the index that was removed from the map and
erased from the vector.
Task-number: QTBUG-25691
Change-Id: Ia2a41027628a65faec4ecdd5da235ddd19746a57
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
e.g. in QStaticText, the data is used just to get the line's y-position
and re-calculates just after the loop to determine the bounding rect and to draw the text;
in QWidgetLineControl, the data re-calculated over and over while the result
is seems to remain the same; probably the caching is needed here too
Change-Id: I0f7eb291532f63eccb9c5f749daebb73ff90632f
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
A case was found where a method inherited from a
QML type marked abstract had the same name as a
method in the inheriting class, and these two
methods received the same quid. This was fixed
by including the QML type name in the guid for
QML methods.
Change-Id: I110eb254b3c6be014cb67fdc5b57b5aa2f575220
Reviewed-by: Casper van Donderen <casper.vandonderen@nokia.com>
The current alternative is to define QT_WIDGETS_LIB before including
qtest.h, but this is not convenient/intuitive when using other build
systems than qmake. If one forgets the define, crashes happen when
using QApplication-related code.
Use <QTestWidgets> in one of the widgets autotests, for testing.
Change-Id: Id96be4976723aea3e8a28c9d0d594daab25a6d90
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
The Mac platform requires populating the font database to build the
fallback list for font families.
Change-Id: I5bd63cd3c5fa6216c312d439390681f160a716fb
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
Required for Cocoa platform menus support, we need a
way to update state after focusWindow() result changes.
Change-Id: Idc573888c3d75bcbff2252e243c4b57b15fc2fcd
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
cursor-atlas.png was generated from existing cursor images
(qttools/src/shared/qtpropertybrowser/images/)
Change-Id: Ic4b396590eaec93e14a4b0915b15f735f5b1a5f5
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
On Windows, Unicode command line arguments are re-created from
the original command line filtering out the known arguments.
To avoid having to hard-code all arguments of derived application
classes, keep the original argv-array and use that to verify if
an argument is still present.
Task-number: QTBUG-25724
Change-Id: I5d7bbd9530b1b74e1dcd22a0edc4f323ef687d23
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The Unicode non-characters are 32 characters, from U+FDD0 to
U+FDEF. The code matching these comments was fixed in
9327bc87c3, but the comment wasn't
fixed.
Change-Id: I5bde0ab9d70c1c6623893de36d31235cbd9fb152
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
This adds the Visual Studio 11 mkspec
and the corresponding changes to configure
and qmake makesystem.
Change-Id: I3a7e82a6f7f90aa0a94dedd493ebaa66bf100923
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
The basic config file includes the macros file already if needed.
Change-Id: I8b03360ce1e9fff4a3be5270f659dbe22fc1b295
Reviewed-by: Clinton Stimpson <clinton@elemtech.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
This is required if Qt is installed into a directory with spaces.
Change-Id: I1d6874265558d712ac98a3aef670c2934a632ab1
Reviewed-by: Clinton Stimpson <clinton@elemtech.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Copying events is a bad idea but it is permitted, used at least in the
state machine framework and QApplication, which somehow found it
amusing to clone events. We can't forbid it because it would be
source-incompatible with Qt 4, and other ill-advised developer may be
doing this.
In the new copy functions and in the destructor, ensure that the d
pointer is null. We can't copy it if it isn't. The exception is for
DeferredDelete events, which use the d pointer to store the loop level
count. Such value must not be deleted.
In the future, if QEvent::d is used at the QEvent level, make sure to
adapt QCoreApplication::postEvent to store the counter somewhere else.
Task-number: QTBUG-25070
Change-Id: I1f2d3f3cfc891ec216df2e8b7dbe531524d21b26
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Try to use a builtin codec as codecForLocale() if possible
first. Fall back and instantiate the iconv codec only
if that failed.
In addition, make sure we initialize the locale correctly
before we try to setup the codec.
Change-Id: I86d635f9d11e8ff93093f162e79fb37f3d85731b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The parser is recursive and too deeply nested json would
cause it to exhaust the available stack space leading to
crashes.
We now abort parsing with a DeepNesting parse error if the
document is too deeply nested. The current nesting limit
is set to 1024, which should be more then enough for any
real JSON data set.
Change-Id: I4adea3fd727149f7342536d73cf4530361a0a3a1
Reviewed-by: Jamey Hicks <jamey.hicks@nokia.com>
Reviewed-by: Denis Dzyubenko <denis.dzyubenko@nokia.com>
In rare cases, if the section is empty, we're reading the whole object
into memory because size -1 = UINT_MAX.
Change-Id: Ibf9a1534159ce626e4f2327536076d0cc1ebf0ba
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
This reverts commit a17523805e56511465550a6a93a88b3fc3c8325a
The compatibility overloads were removed in change Icf108a80177155f21bb73c165fb8ab5d4e997bc2.
Change-Id: I4861f281451d66a1aa5f3525eea1773dfef0540e
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Because the QPA font database would query fallback families inside
findFont(), support for requesting multiple font families in order
of preference (like QFont("Times New Roman, Arial")) did not work,
because the Arial fallback was never attempted. To fix this, we
pass in the queried fallbacks and make sure they are tried before
any platform specific fallbacks.
Task-number: QTBUG-20986
Change-Id: Idb2b717856f013ce2874f00a8debaff60176d2fc
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
+ QChar::LastValidCodePoint enum value that supercede the UNICODE_LAST_CODEPOINT macro
replace uses of hardcoded values with the new API; remove leftovers
Change-Id: I1395c9840b85fcb6b08e241b131794a98773c952
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
inline all non-static members to a static ones (declared with QT_FASTCALL),
ushort converts automatically to uint and the conversion cost is minimal.
Task-Number: QTBUG-13052
Change-Id: I189a6f205736766adcd3de2d61cee71f30cc64f3
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
texts for the textKeys are stored each in a separate COM(ment) section
so that the maximum text's size is almost 65KB
Task-number: QTBUG-10568
Task-number: QTBUG-111
Change-Id: I7d693741e10e5d78d497cb0af448160077350bb2
Reviewed-by: aavit <qt_aavit@ovi.com>
When qt_ntfs_permission_lookup is used, QFile::permissions failed
for files with long filenames.
Also created a test case for this API, which revealed another bug.
Task-number: QTBUG-25629
Change-Id: I73b7676a9d059c0e782b3f701b2e6bbc92f671ed
Reviewed-by: Prasanth Ullattil <prasanth.ullattil@nokia.com>
This can be enabled again now. The sender transfer test is still not
re-enabled since the test would take to long to run if sending
enough data to overwhelm the reciever's kernel buffers.
Change-Id: I4056fdca53ec8ebbcc53dfdc814d8bfdbc73f7ce
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
Reviewed-by: Prasanth Ullattil <prasanth.ullattil@nokia.com>
Incase there is no bytesAvailable on the socket there is no need to
try to read.
Change-Id: Ied9e4b8d86854b356a733fab4948739db206b6ad
Reviewed-by: Prasanth Ullattil <prasanth.ullattil@nokia.com>
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
When we get a remoteHostClosed we should try to read everything from
the socket before we close the channel.
Change-Id: Iaa87d79ea16d69735f6ba3e8b3b4a0f86fbd5f73
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
Reviewed-by: Prasanth Ullattil <prasanth.ullattil@nokia.com>
Reviewed-by: Richard J. Moore <rich@kde.org>
