PCRE doesn't like illegal Unicode sequences (it's explicitely
documented in pcreunicode(3) that they trigger undefined behaviour,
and the program may crash). Therefore, we always let PCRE check
the validity of both the pattern and the subject string.
However, when performing global matching, the subject string
can be checked only once: subsequent matches can safely skip the check
and avoid a huge performance hit of scanning the whole subject
string for each match (!).
This patch implements that behaviour internally -- it's still
not possible for the user to skip the sanity check. On large
subject strings, this gives a terrific performance benefit.
Change-Id: Ia44cf18782e07966c9cd6ec4ccfef081ed131763
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Since libjpeg 7, decompression scaling factors of [1,16]/8 have
been supported. Upscaling is slower, but using the entire range
of downscaling factors is significantly faster for sizes between
the power-of-two factors.
Time to decompress a 5184x3456 image and scale to 1900x1200,
slightly less than 3/8ths, changes from 251ms to 203ms.
libjpeg versions prior to 7 will round up to the next largest
factor they support, and continue to work as before.
Change-Id: I00a0655df2ef057e739927a643bfe0b0cabd5602
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: aavit <qt_aavit@ovi.com>
Snap points are not needed if the overshoot condition is met; therefore,
move the overshoot handling before the computation of snap points.
Change-Id: I37556b10c1b1124eaed53b0f2a330085acf70776
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Compute the time needed for the scrolling operation before computing the
movement on the X and Y components. This ensures (except when
overshooting or snapping to snap points) that the scrolling on the X and
Y components takes the same time. That is, scrolling occurs always along
a straight line.
Task-number: QTBUG-23227
Change-Id: Ic3b78b8611dbcbf19711c3dd96485a0d53accaed
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Separate PPS based navigator event notification handling and implementation
of the actual event handling.
Like the similar refactoring for screen events, this will allow to reuse the
same event handler class for BPS driven event processing.
Change-Id: I29d4a082cb01ff7943b7822ea9aa8ad622fda593
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Add a toplevel manual.pro and other .pro-files to build them and
fix the build.
Change-Id: Ibc98a27b39dd1304edfa8a6894d62e77ce7ef387
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
I can't say for sure why q*linecontrol passes an empty rect to
the updateNeeded() signal when an input mask is set; presumably
the empty rect at some point has meant "full update", but there
are a few problems with this. Surely a full update is wrong,
even if the semantics have been lost in translation somewhere
(likely the qlinecontrol refactoring). This fix ensures
that empty rects from updateNeeded() are interpreted as a request
to update the whole widget. A further improvement would be to
ensure the line control doesn't request a full update when an
input mask is set. The cursor is usually wider when a mask is
set but because of QLineEdit::paintEvent()'s implementation,
there is currently a mismatch between the cursor width as seen
by q*linecontrol and what is actually drawn, which causes
rendering artifacts if updateNeeded() sends the cursorRect().
Since QLineEdit and Q*LineControl aren't actively developed, it's
best to keep this fix minimal, although the performance cost of
updating the whole line edit when an input mask is set is
unfortunate.
Task-number: QTBUG-7174
Change-Id: Ie51e015d760915e07b0220b770f04fc958d93a12
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
Using g++ 4.6.1, we get warnings like below:
qprocess_unix.cpp:1376:69: warning: ignoring return value of
‘int chdir(const char*)’, declared with attribute warn_unused_result [-Wunused-result]
g++ is pretty adamant and prints the warning even if you explicitly
ignore using (void). So, just check for error and print a warning.
Change-Id: Ifd6f3b6bb9e17d44aa235815b06a762131ca8751
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Don't pick the modality type based on the presences of a parent widget,
use QWidget::windowModality() instead. If modality hasn't been
specified, then use presence of the parent to chose an appropriate
modality type.
Change-Id: Ib39c166ba3293174c55748f63222367e3e1c56bc
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
libudev extends the base functionality of these plugins but should not
be mandatory as it is missing on many embedded devices that still
actually require input devices (keyboard and/or mouse and/or
touchscreen) support
Change-Id: Ieeb949f1af5e774578f689a63f47a8c48f546ac1
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
Add mkspec for the Trident's Shiner platform to be used in conjunction
with the -device support in configure. This allows you to build Qt
Change-Id: I68d8e8eb78f073e98f72207e9c5de6a0da0ed35d
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Add mkspec for the STMicroelectronic's ST7108 platform to be used in
conjunction with the -device support in configure. This allows you
to build Qt with the application libraries provided by STM SDK 35.
Change-Id: I36aa174200f034ccd6ee285e3f8a8ffec99b3e70
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
Add mkspec for the STMicroelectronic's ST7540 platform to be used in
conjunction with the -device support in configure. This allows you
to build Qt with the application libraries provided by STM SDK 36.
Change-Id: I4369b23b9a0c9063417e5cc71969f65e967e4019
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
The suggested behavioural change had the potential to break existing
code, so the change won't be made for Qt 5.
Task-number: QTBUG-25119
Change-Id: Ie03271d12b21a800c998e073eeb9ca1cd03ffe19
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
mDelegate keeps the pointer to a QNSFontPanelDelegate, which reacts
on the NSFontPanel used by this dialog helper. It has to be created
before it can be used. On a read-only access, this has been fixed to
return a default-constructed value (like QFont()). For writing access
(like setting the font) the delegate was already created.
The same applies to mDelegate to QNSColorPanelDelegate respective.
Change-Id: I36b89c16d98db9275aa31d399fe094b1d56d800d
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Change-Id: I21dd1e3186e0dbbc60294f807de0db6aad8b9eee
Reviewed-by: James Turner <james.turner@kdab.com>
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Reviewed-by: Christoph Schleifenbaum <christoph.schleifenbaum@kdab.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
If VC++ 2010 Express and VC# 2010 Express are installed, then the
installation path is written in two places into the registry.
We're now filtering detected installations with the same installation
path.
Task-number: QTBUG-24956
Change-Id: I401430e7aa81d96c523d8172d2a2e9d40ebdb3ce
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
- The crash on XCB can no longer be reproduced.
- Use QFINDTESTDATA instead of SRCDIR defines.
- Remove Windows CE specific profile section.
Task-number: QTBUG-20756
Change-Id: I6077b3a0daacb15ab440a90c7bda247aa3756fa5
Reviewed-by: Jason McDonald <jason.mcdonald@nokia.com>
0xfdef-0xfdd0 is definitely 31 and not 15 :)
also fix all copy-pastes of this code (greping for '0xfdd0' helps ;)
Change-Id: I8f3bd4fd9d85f9de066f0f5df378b9188c12bd48
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Denis Dzyubenko <denis.dzyubenko@nokia.com>
The change discussed in the comment doesn't have to be done in a major
release if it isn't source-incompatible.
Task-number: QTBUG-25103
Change-Id: I50036ab13611871ede01b7b7a17ce4c325476b00
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Crash on XCB can no longer be reproduced.
Task-number: QTBUG-20756
Change-Id: I057231a397573f2a28a1325c6d6f728735ebbee6
Reviewed-by: Jason McDonald <jason.mcdonald@nokia.com>
Neither of these comments will be actioned for Qt 5: the first because
would be source-incompatible with Qt 4, the second becuase it would be
a significant behavioural change.
Task-number: QTBUG-25090
Change-Id: I5f8f7cce3007c3188b2f0184138fa8e55a165654
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Currently, for host builds, pkg-config usage is autodetected based
on it's availability in the mkspec or the PATH. For xcompile builds,
pkg-config is disabled unless -force-pkg-config is passed.
-force-pkg-config is poorly named since it doesn't reflect the fact
that it applies only to xplatform builds. It is in fact the only way to
enable pkg-config in xcompile builds. And when passed, it doesn't actually
force anything since all it does is check env variables. To add to the
confusion, it prints a warning even if the env variables are setup correctly.
This patch remedies the situation. It adds (-no)-pkg-config. The flag works
for both host and xcompile builds.
By default, the value is 'auto'. In this mode, it will try try to detect pkg-config
from the path. If found, it will be used. For xcompiled builds, we use some heuristics
to determine if the pkg-config is actually usable:
1. if -sysroot is not set and the environment variables PKG_CONFIG_LIBDIR or
PKG_CONFIG_SYSROOT_DIR are not set, we disable pkg-config.
2. if -sysroot is set, then we setup PKG_CONFIG_LIBDIR and PKG_CONFIG_SYSROOT_DIR
automatically (provided $SYSROOT/usr/lib/pkgconfig exists).
If the value is 'yes', configure will error if it's heuristics fail to detect a usable
pkg-config.
If the value is 'no', pkg-config usage is disabled.
If the value is 'force', configure will skip it's heuristics and use pkg-config anyway.
This mode is useful, for example, when compiling for 32-bit on 64-bit systems.
This change also removes references to PKG_CONFIG_SYSROOT (PKG_CONFIG_SYSROOT_DIR
is the correct environment variable).
Change-Id: I07fc8d48603c65a60de0336fc6276e90fcb41430
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Apply the state in QWidgetWindow and send an event to
the widget unless the code is triggered by
QWidget::setWindowState().
Change-Id: Ibf2f4e730384e41636841b9216eecfdff35b7bcb
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
- Avoid sending Window State change events from
WM_STATE/NET_WM_STATE changes irrelevant to Qt::WindowState.
- Introduce QFlags for the NetWmState getter and setter to
avoid passing QVector<> around.
Change-Id: I74730928c7fffca0fa1cab3b90ded90b06304c06
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
If a modal top-level widget has WA_DontShowOnScreen set, we need to call
QGuiApplicationPrivate::showModalWindow() and hideModalWindow()
ourselves, since we will not be calling QWindow::setVisible() (which
would normally do the call for us).
Change-Id: I1b22dd177c5956a2290f3ee031c95ab50d88f153
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
This tests that modalWindow() returns the expected value and that
QEvent::WindowBlocked and QEvent::WindowUnblocked are sent correctly
when modal windows are hidden and shown.
Change-Id: I872f35e0240c928566ab35fa5764fad6cfda6db6
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
QWindow already has windowModality() and setWindowModality() as part of
its API from commit 516f4e283b. Platform
plugins can use this already to setup modality hints on windows that
they create, but it's not enough to implement modality fully.
QGuiApplication gets a modalWindow() static method, which is similar to
QApplication::activeModalWidget() in that it returns the last modal
window to be shown.
The modal window "stack" moves from QApplicationPrivate to
QGuiApplicationPrivate. The enterModal*() and leaveModal*() functions in
QApplicationPrivate are removed and replaced by
QGuiApplicationPrivate::showModalWindow() and hideModalWindow(), which
are called by QWindow::setVisible() just before calling
QPlatformWindow::setVisible().
The virtual QGuiApplicationPrivate::isWindowBlocked() will tell us if a
window is blocked by a modal window (and tell which modal window for any
interested callers). The default implementation works on the QWindow
level. QApplicationPrivate reimplements isWindowBlocked() and adds popup
and WA_GroupLeader support.
QGuiApplication uses the state set from isWindowBlocked() to block
user-input events: mouse press, mouse move, mouse release, wheel, key
presses, key releases, enter/leave events, close events, and touch
begin, update, and end events.
Note also that the modality helper functions in QtWidgets and
QApplicationPrivate are left in place and working as they always have.
The behavior of QWidget in the presence of modal windows/dialogs should
not change.
Change-Id: I2c89e6026d40160387787a6e009ae1fdc12dfd69
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
This was accidentally removed in the big change regarding searching in
the internal QDoc tree.
Change-Id: I2496d7497d239f1ec5fbd01be6a918c1ef29fc95
Reviewed-by: Casper van Donderen <casper.vandonderen@nokia.com>
Now the default for a QML property is writable. If qdoc
can't detect the actual read-only status, writable is
assumed. There were some cases where qdoc could not
determine the actual read-only/writable status for a
QML property. In these cases, qdoc reported read-only
because the default was read-only, which was not optimal.
Change-Id: I55aeb2bedcde92a414f4d48a8d995e5e9dbca5da
Reviewed-by: Casper van Donderen <casper.vandonderen@nokia.com>
by reordering and regrouping conditions so that they lead to result earlier
in most-common usecases (l.letters, spaces and puncts, u.letters, other);
there are no title cased letters in range [0..127] -> use this in isTitleCase();
test for 0xa0 (nbsp) early in isSpace() as it is quite common in HTML, etc.;
add early test to isNumber().
Change-Id: Ib415f34cb1212d9ccf8753de2d1beaece1aa2243
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Windows ships with a minimal set of CA roots.
When using windows API to verify a certificate, it will fetch the
root certificate from windows update (assuming it is part of the
Microsoft trust program).
As we are using openssl, this does not happen transparently.
If SSL errors occur which indicate a broken chain then attempt
to fix it using the windows API before emitting sslErrors.
If the system CA certs are not in use (a CA bundle has been set
on the socket or as the global configuration), then this is skipped.
This is so an application can continue to use its own cert bundle
rather than trusting the system certs.
Key usage is specified, so that windows will return not trusted
status if the root is not suitable for SSL (server auth or
client auth OID).
Testability:
- to test, must delete the CA cert(s) from the "third party
root certification authorities" section of the cert store
using mmc.exe.
- If the workaround of installing the windows XP cert bundle was
performed, then you also need to delete certs from the "trusted
root certification authorities" section.
This is dangerous, be careful not to delete the required
certificates which are documented on MS website
- Naturally, modifying these areas of the cert store requires
elevated privilege.
Task-number: QTBUG-24827
Change-Id: I5cfe71c8a10595731f6bbbbabaaefa3313496654
Reviewed-by: Richard J. Moore <rich@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Removed some MacOS source code files from iOS build. Use unix standard
paths for now (iOS-specific implementation will come later).
Change-Id: I8b2731b431b3a379a1ec4ec07d227e886209e3e9
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
Initial bits to enable figuring out the EGL display and context.
Change-Id: I4b578e356dceb40b4456f0590d32c8df1f51fa53
Reviewed-by: Andy Nichols <andy.nichols@nokia.com>
KMS plugin deals with FBO IDs not equal to zero (the default FBO),
so return the correct ID.
Change-Id: I904fc0b8d732f856b4526bd7f73cc48c358c8441
Reviewed-by: Andy Nichols <andy.nichols@nokia.com>
Threaded applications like qmlscene could not bind the EGL context if
the context is already bound in the other thread.
Change-Id: Ia75ef9e76ebff48aa2c9b348101ab2f388e18c5e
Reviewed-by: Andy Nichols <andy.nichols@nokia.com>
Sync up the KMS QPA plugin with the plugin system changes.
Change-Id: Ifaa8be6f11aeb93acc63643c62ca15db4e9bc38f
Reviewed-by: Andy Nichols <andy.nichols@nokia.com>
The naming convention for these libraries says that libGLES_CM is to be
used when EGL is included while libGLESv1_CM should be used when EGL is
not included. Since we have a seperate variable for libEGL, it
makes sense to have this variable represent the non-EGL version of the
library.
Change-Id: I9147c116da7be4a296a0ebeac39762b46725f10e
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
The comment relates to a change that is neither source- nor
binary-incompatible, so the change can be done in any minor release.
Task-number: QTBUG-25117
Change-Id: Ifba3ef53241f9bf2504c573066e4cfa5fbfe679e
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
There are two problems with the current design:
1. if (hooks) hooks->foo() doesn't work in debug mode when no platform hook
is defined. The problem doesn't arise in release mode because the compiler
optimizes away the if (hooks) into a no-op since hooks is NULL when no
platform hook is defined.
2. Adding a new hook requires changing every platform's hook implementation.
New approach:
1. Define QEglFSHooks as a class with virtual functions. A stub file provides
the default implementation.
2. Platform hooks derive from above class and reimplement whatever is needed.
The filenames and variables have been changed to be more in line with the
Qt style.
Change-Id: I2eaaa5ad7c8b48a06361c4747d4f210c428c983f
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
This removes a couple of functions. Two of them are unused and the
last one has its (now) very simple implementation inlined in the
only caller. The last function was called something with spans and
we would like to get away from using the word 'span' since we no
longer uses spans.
Change-Id: Icef95166289d52bd958400cba70daceb6fa75913
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
