| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
This patch fixes the deployment of files when using Qt For Symbian SDK
on Linux.
Merge-request: 839
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
| |
|
|
|
|
| |
Reviwed-by: Simon Hausmann <simon.hausmann@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Commit 3ed9e2788e8 changed the logic from using lookahead to checking
if the library already existed in the arguments parsed so far. This
broke the case -lFoo -lBar -lFoo where Bar depends on symbols in Foo.
Since we were reducing this to -lFoo -lBar, the linker would not know
which missing symbols to include from Foo when encountering the -lFoo
argument.
We now keep the library order when building the final argument list,
by making sure the position of a duplicated library is always that
of the last instance of that library.
Task-number: QTBUG-13944
Reviewed-by: Simon Hausmann <simon.hausmann@nokia.com>
|
|
|
|
|
|
|
|
| |
there is no point in &&-ing it with a possibly present signature command
- putting it on a line of its own works just fine - with the added
benefit of not breaking if it contains operators itself.
Reviewed-by: mariusSO
|
|
|
|
|
|
|
|
|
|
| |
instead of glueing the commands with "&&" and having a monster hack
(\r\h) for the case where they cannot be and-ed, glue them with newlines
and error checks. this ensures that we a) never have excessively long
lines, b) don't need the monster hack and c) commands always have proper
error checks, just like in makefiles.
Reviewed-by: mariusSO
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1: (34 commits)
Add declarative changelog entries for 4.7.1
connman backend
My changelog entries for 4.7.1
updated changes-4.7.1
Update frameGeometry when the unified toolbar visiblity is toggled
Update changelog
my contributions to dist/changes-4.7.1
My 4.7.1 changes.
added my and Jan-Arve's change to changelog for 4.7.1
Changes done for 4.7.1
Updated changes-4.7.1
My changes.
Fix for bug QTMOBILITY-448 to list the default network configuration in Bearermonitor example
Initialise pointer variables.
An aborted QNetworkSession on Maemo must emit SessionAbortedError.
Autotest: also enable testing of QtDeclarative
Add Qt 4.7.0 baseline symbols to the 4.7 branch
Call createpackage without the explicit .bat suffix
Improve Qt/Symbian compatibility with unix shells
Ukrainian translation updated
...
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Remove an "echo." call that doesn't work on unix, and make the
create_temps target work with unix shell syntax (QMAKE_CHK_DIR_EXISTS
requires an || before the command to execute on unix shells).
Merge-request: 823
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
|/
|
|
|
|
|
|
|
|
|
|
| |
Previously manual editing of the pkg file was required to publish
application using smart installer. Now a proper app_installer.pkg
will be created as long as application has protected range UID.
Also changed "make installer_sis" to always generate the application
sis as publishing process supports signing both application and its
smart installer wrapper packages in single step.
Task-number: QTBUG-13991
Reviewed-by: axis
|
|
|
|
|
|
|
|
|
|
|
|
| |
Qmake's smart library merge would not take Xarch into account, so lines
like -Xarch_i386 -foo -Xarch_ppc -foo would be reduced to -Xarch_i386
-Xarch_ppc -foo, due to the "duplicate" -foo.
The explicit addition of the -Xarch option for each entry of the
architecture specific options is intentional, to make cases like
"-Xarch_i386 -framework Foo" magically work.
Reviewed-by: Simon Hausmann <simon.hausmann@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QT_RVCT_VERSION environment variable can be used to specify
the desired RVCT version for symbian-sbsv2 Makefiles:
E.g. QT_RVCT_VERSION=4.0, release-armv6 -> sbs -c arm.v6.urel.rvct4_0
If no version is specified, generic sbsv2 rvct aliases are used just
like before:
E.g. release-armv6 -> sbs -c armv6_urel
Version specific make targets are also generated for each detected
compiler:
E.g. release-armv6-rvct4.0 -> sbs -c arm.v6.urel.rvct4_0
Also optimized finding the installed compiler versions.
Task-number: QTBUG-13499
Reviewed-by: axis
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Commit 019032c93b7397c7239ec99ec44b4923df31533b introduced a lot of
annoying error messages when configuring Qt. It makes qmake complain
that it cannot find rcc and uic, which aren't yet built at this point.
We're now checking for the existence of the depend_command if it is
given as absolute path. This is the case for uic and rcc.
Yeah this is a weak solution for sick code.
Task-number: QTBUG-13366
Reviewed-by: ossi
|
|
|
|
|
|
|
| |
this magic is a bit braindead anyway, but whatever.
Reviewed-by: joerg
Task-number: QTBUG-13366
|
|\
| |
| |
| |
| |
| |
| |
| |
| | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/qt-s60-public:
Let's not duplicate Symbian softkey menu top-level actions in submenu.
Remove some .pro statements left behind after IAP usage cleanup
Ignore MAKEFILE variable for Symbian abld and sbsv2 builds.
|
| |
| |
| |
| |
| |
| |
| |
| | |
Renaming makefile makes little sense for Symbian toolchains, when
we generate so many other files filenames derived from other sources.
Task-number: QTBUG-13336
Reviewed-by: Janne Anttila
|
|\ \
| |/
|/|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
qmake/Makefile.win32
src/corelib/io/qfsfileengine_win.cpp
src/corelib/kernel/qeventdispatcher_win.cpp
src/gui/dialogs/qfiledialog_win.cpp
src/gui/inputmethod/qcoefepinputcontext_s60.cpp
src/gui/text/qfontdatabase_win.cpp
src/gui/util/qsystemtrayicon_win.cpp
src/script/utils/qscriptdate.cpp
tests/auto/qinputcontext/tst_qinputcontext.cpp
tests/auto/qscriptengine/tst_qscriptengine.cpp
|
| |
| |
| |
| | |
Task-number: QT-3825
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This was a security hole that has been there for a while, but the
public awareness have recently rised so the threat is more imminent
now.
The solution is to fix all places where we dynamically load system
libraries. More specifically, we now load all system libraries with
an absolute path that points to a library in the system directory
(usually c:\windows\system32).
We therefore introduce a small class named QSystemLibrary that only loads
libraries located in the system path. This shares some of the API with
QLibrary (in order to make the patch as small as possible).
We don't fix QLibrary due to risk of regressions. In
addition, applications can fix the code that calls QLibrary themselves.
The problem does not apply to Windows CE, since the search order is
documented as not searching in the current directory.
However, it touches some CE-specific code - therefore QSystemLibrary
is sometimes used on WinCE (however, it will just do a normal
LoadLibrary() since its safe anyway).
This change does not affect the testability plugin (it is not clearly
documented where that plugin is located, and the plugin should never
be used in production code anyway)
Loading OpenSSL libraries
The ssl libraries are handled specially, and searched in this order
(we cannot expect them to always be in the system folder):
1. Application path
2. System libraries path
3. Trying all paths inside the PATH environment variable
Task-number: QT-3825
Reviewed-by: Thiago Macieira
Reviewed-by: Peter Hartmann
|
|\ \
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1: (33 commits)
make shadow builds not write into the source dir
Fix failing tst_qmdisubwindow test after 71a3b1a0d3
Doc: update the documentation of QUrl about HTML Form encoding
search translator comments
Install translators earlier in assistant
Fix compilation on ARM: don't include kernel headers
Fix compilation of qmake with MSVC
Doc: Updating known-issues page.
Recognize @, #, $ and ` as word separators
rebuild configure.exe
Fix dependencies so JOM will handle parallellization properly
Symbian: Disable IPv6 for now
Doc: Fixed namespaced Qt build breakage in the model/view tutorial.
Doc: Fixing links on index page and corecting HTML in the template.
Accept empty authority segments in QUrl as different from not-present
Doc: Adds line about non-support for Webkit on Solaris
protect nil dictionary from release.
Fix potential KERN-EXEC 0 on Symbian.
Ukrainian translation updated
Fix remote crash in delivering D-Bus calls with too few arguments
...
|
| | |
| | |
| | |
| | | |
Reviewed-by: TrustMe
|
| |\ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1: (24 commits)
rebuild configure.exe
Fix dependencies so JOM will handle parallellization properly
Symbian: Disable IPv6 for now
Doc: Fixed namespaced Qt build breakage in the model/view tutorial.
Doc: Fixing links on index page and corecting HTML in the template.
Accept empty authority segments in QUrl as different from not-present
Doc: Adds line about non-support for Webkit on Solaris
protect nil dictionary from release.
Fix potential KERN-EXEC 0 on Symbian.
Ukrainian translation updated
Fix remote crash in delivering D-Bus calls with too few arguments
Ensure that OpenGL contexts are attached to an NSView before first paint
Doc: Updating the 3rd party lib used in Qt - including JQuery
Doc: Changed the default URL to avoid a Flash plugin crash on x86-64.
Reduce memory consumption of QtScript/JSC on Symbian
Doc: Updated Supported Platforms page. Finished
New binary of configure.exe with -mp/-no-mp option
Turn Makefile.win32 into batch-mode, enable /MP and PCH
Add option to enable -MP compile option for MSVC
Rewrite ucstrcmp in terms of ucstrncmp
...
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
JOM wouldn't wait for qmake_pch.obj to finish before starting
the other compiles.
Reviewed-by: trustme
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This should make compiling qmake quite a bit faster, since less
invokations of the compiler is needed, every compile unit starts
of with a already known state, and all the CPU cores are used to
compile the units.
Reviewed-by: bhughes
Reviewed-by: ossi
|
|/ / /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
- No longer require PRE_TARGETDEPS items to be absolute, which was
difficult to achieve sometimes as qmake doesn't provide method
for absolutizing paths.
- Do smart command replacement for commands containing $$QMAKE_*
command variables, such as $$QMAKE_COPY, when generating bld.inf
extensions for QMAKE_EXTRA_* variables.
$$QMAKE_* command variables cannot be passed to sbsv2 toolchain
directly, as it uses cygwin in windows, and they can't simply
be replaced to use sbsv2 equivalents in symbian.conf, because
generated wrapper makefiles need them to be Windows compatible.
Reviewed-by: axis
|
|/ /
| |
| |
| |
| |
| |
| |
| |
| | |
Plenty of filenames got generated with spaces which didn't work
as that was not taken into account. Harmonized TARGET fixing
across prf files and qmake Symbian generator code.
Task-number: QTBUG-13363
Reviewed-by: Thomas Zander
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Removed usage of buggy .release_gcce target from symbian-sbsv2
generated makefiles. Now gcce builds like armv5 as it is supposed to
when using Raptor.
Task-number: QTBUG-13307
Reviewed-by: axis
|
|\ \
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1: (61 commits)
Autotest: don't use Q_FUNC_INFO for testing which method got called
fix generated makefile dependencies
Cocoa: Demo browser can get stuck after closing modal dialog
Restore default if to system default on session close.
tst_qmake doesn't need QtGui
Use the full path to qmake in the qmake unit test
qdoc: Fixed erroneous links to QML basic types.
Fixed item view background color in Gtk style
scope fixes and clutter reduction for sql driver projects
I don't know why some linkers can't call this function, so comment it out.
QNetworkSession::close() method now send closed() signal while faking disconnection.
Add the missing license headers to the QString benchmark data
Fix building of qsimd.cpp on Windows CE
Use QElapsedTimer for the benchlib tests.
Properly implement the CPU feature disabling in qsimd.cpp.
Report the detected CPU features in the corelib boilerplate
Detect CPU features on ARM by reading the ELF auxvec.
Split the CPU-detection code into multiple functions for readability
Fixed delivering gestures to a toplevel widget.
Unroll the SSSE3 code even more to avoid the need to keep an extra variable for inverting the result
...
|
| | |
| | |
| | |
| | |
| | | |
seems like some makes are rather lax about bogus dependencies, so this
went unnoticed in local testing for at least two people.
|
| | |
| | |
| | |
| | |
| | | |
Task-number: QTBUG-13081
Reviewed-by: Martin Petersson
|
| | |
| | |
| | |
| | |
| | | |
Task-number: QTBUG-13080
Reviewed-by: Martin Petersson
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
like in SUBDIRS, the specified strings can now be basenames of
"structures" which specify the actual input and output files:
QMAKE_SUBSTITUTES += test
test.input = infile.txt.in
test.output = foobar.out
Reviewed-by: joerg
|
|\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/qt-s60-public:
Added support for DEPLOYMENT.pkg_build_version
fix for memory leak in QSysInfo::s60Version()
|
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
DEPLOYMENT.pkg_build_version can be used to pad the patch version
number in Symbian pkg files to allow hot-fix deployments to be done
between patch releases.
DEPLOYMENT.pkg_build_version accepts only values 0 - 99, as pkg patch
version number has limits how large numbers it can accept. The specified
number is prefixed with zero if it is single digit and then appended
to patch version number.
Task-number: QTBUG-13147
Reviewed-by: Janne Koskinen
|
|\ \ \
| |/ /
|/| /
| |/
| |
| |
| | |
Conflicts:
bin/syncqt
src/opengl/qgl.cpp
tools/configure/configureapp.cpp
|
| |
| |
| |
| |
| |
| |
| |
| | |
Basically replacing project->values("foo").first() with
project->first("foo") which does a null check.
Task-number: QTBUG-12851
Reviewed-by: ossi
|
| |
| |
| |
| |
| | |
Task-number: QTBUG-12807
Reviewed-by: TrustMe
|
| |
| |
| |
| |
| |
| |
| | |
Don't add the PDB file name to ForcedIncludeFiles.
Task-number: QTBUG-12802
Reviewed-by: TrustMe
|
| |
| |
| |
| |
| |
| | |
it wasn't the fault of fileFixify() after all ...
Reviewed-by: joerg
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When QMake decides to use absolute path for some object file, this
code used to leave the ADDMOD command out from the ar object script,
which resulted in following errors from the issued "ar -M" command:
Syntax error in archive script, line 2
Syntax error in archive script, line 3
etc...
And the build failed in the linking phase, because the generated
static library was empty.
Fix this issue by prepending the file name with ADDMOD command even if
the file path is absolute.
Task-number: QTBUG-12959
Merge-request: 2451
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
still doesn't work for nested directories, but that's because
fileFixify() is plain broken.
on the way, remove a superfluous check: we know that the input file name
ends with .in - three lines up we made sure it does.
Reviewed-by: joerg
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Previously check to remove files from deployment in case the deployment
target location was the same as the source location was done using
QFileInfo equality, which resolved to true when file didn't yet exist
at the time qmake was run. Changed the check to compare absolute paths
instead.
Task-number: QTBUG-12879
Reviewed-by: Shane Kearns
|
| |
| |
| |
| |
| |
| |
| |
| | |
Changes to .pro file weren't previously noticed and therefore
qmake wasn't rerun. Qmake command itself was also incorrect.
Task-number: QTBUG-12716
Reviewed-by: Shane Kearns
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
It is now possible to use "release-gcce" and "debug-gcce" targets
in symbian-sbsv2 mkspec. Exports and cleans should also now work
properly for all target platforms under symbian-sbsv2.
Task-number: QTBUG-12762
Reviewed-by: Jason Barron
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Mmp and mk files previously contained UID3 as unique identifier in
the filename. However, this was not particularly useful for building
subprojects in massive project trees such as Qt, as it was difficult
to remember UID for each subproject. Now with intuitive naming of mmps,
it is possible to build a subproject like this:
sbs -c armv6_urel -p qtcore_dll.mmp
Task-number: QTBUG-12715
Reviewed-by: Jason Barron
|
| |
| |
| |
| |
| |
| |
| |
| | |
Missing minor or patch value from VERSION caused malformed pkg header
to be generated.
Task-number: QTBUG-12617
Reviewed-by: axis
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
the file names are given relative to the build directory, so the command
needs to be run in it as well. this is a more expected (and simpler) fix
than the alternative, which would be giving file names relative to the
source directory.
reasons not to fix:
- due to some other bug, the problem really affects only builds where
the build dir is not at the same level as the source dir - otherwise,
absolute paths would be passed anyway
- it has some breakage potential for the cases where the commands
actually expect being run in the source dir
- it can be worked around by manually injecting the cd statement into
the command
reasons why i still fixed it:
- it doesn't affect in-source builds, and it seems that most complex
build systems (which would define custom compilers with
depend_command) don't support shadow builds anyway
- people who needed things to work probably already used $$OUT_PWD somehow
(either a "cd" at the start, or prepending it to each path), so this
change will be practically a no-op
- "it's just dependencies, and these are known to be broken in qmake
anyway"
Reviewed-by: joerg
Task-number: QTBUG-1918
|
| |
| |
| |
| |
| |
| |
| | |
Pre-built resources didn't get properly copied under \epoc32\data\z.
Task-number: QTBUG-11519
Reviewed-by: Janne Koskinen
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The previous patch caused the pch header to be compiled without exception
handling since the compilertool for the pch compilation do not get
its options filled from the compiler flags.
This patch instead set the value to off before calling parseOptions.
This also reverts commit 73fa311f67b21c9b897de0196d3b8227f27d828f.
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
it is not done for the top-level file, either.
this is minimally behavior-incompatible, but anyone relying on the old
behavior (and thus not using $$basename()) should be shot anyway. :)
Reviewed-by: joerg
|
| | |
|