| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Embedded platforms are not always called exactly
linux-g++ only. A platform may also be called
linux-arm-gnueabi-g++ or similar.
Fail gracefully if the host architecture is not x86_64.
We require 64bit for linking QtWebEngineCore.
Change-Id: I62c32606517bed6ed4307720d3c95e8019ec134b
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Until now we relied on the PATH to pick the right clang version on
OS X which turned out to be a source of issues on build nodes that
have a lot of old cruft lying around which was picked up by gyp.
Set make_clang_dir so that gyp uses the configured clang version
from Qt and remove the env manipulation from our gyp script.
Change-Id: I4ac679ea56fa874eecaa578aad77b462445c0caa
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
|
| |
Enabling rendering into a WGL backbuffer, in addition to the EGL/angle
one.
Change-Id: I8f2e3f5ecf52b6db22712b1129059f462725a256
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
| |
Change-Id: I1edd16e6875198d7480897e7b2e2fbf68b245883
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Remove unneeded codepaths now that we can trust the git shasums
found in the .DEPS.git file.
No need for parsing and verifying remote branches, we can simply
fetch the specified sha1 from origin.
This patch also unifies the 'shasum' and 'ref' members of the
Submodule python class since the 'ref' member can represent
both chromium's version tag and the sha1s of the submodules,
there is no need for a separate codepath for these.
Change-Id: I1300b5b74f4d2e6984943570963b2f813b1b1679
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
| |
Change-Id: If696de316dd7b3b801ba46d1a54c37eafa901bf1
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
| |
Change-Id: Ie9334cc9cdd240b2f5ceea0e31d82258045ea18c
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
This fixes Qt's packaging scripts that use the --to-crlf switch of
zip, which doesn't work when files already contain CRLF.
The script now takes much longer to run and a progress indicator was
added to the console.
Change-Id: I459c0d95f67b40ab57ff3b337210847cc60debc7
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Chromium is now relying on git for its releases and isn't creating
release DEPS files anymore.
This patch makes sure that we ignore any deps not starting with src/
to exclude the script repositories outside of the Chromium sources
and it removes all the svn-related logic.
Change-Id: I24b1839f919821382b8c5993c71251dbd6108d4f
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
This moves init-repository.py to tools/scripts since it should now
mainly be used by Qt WebEngine developers after our integration with
qt5.git.
Also remove the README file since most of the information is no as
relevant for the Qt module audience as it was for a labs project.
Change-Id: Iad9f6582d0ec0834cdaf38d0551d24f2c273badf
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The real problem that this patch fixes is that moc and rcc generated
files are added to the gyp sources variable with an absolute path,
which for some reason triggers gyp to place the output file in the
same directory as the source instead of the Release or Debug
directories. This causes the last compiled file to be shared by both
the release and debug builds when building debug-and-release, and
MSVC will fail linking if the linked files have incompatible CRTs.
This patch fixes the issue by moving both the generated cpp files
and their compiled objects under <(SHARED_INTERMEDIATE_DIR), which
points to (Release|Debug)/gen.
Change-Id: I4143340acf56c3c7ed665aaae8f6221c310aafa9
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
|
|
| |
Upstream Chromium hasn't been building with the version of clang
available in Xcode 5.0 for a while and we have to disable the build
until we see a benefit to maintain this build environment.
Change-Id: Ia91844dc7f3be1b1623610be417409d0d065c623
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
| |
Change-Id: If391781604861961387019ea15e9c29b6a1e3e1a
Reviewed-by: Zeno Albisser <zeno.albisser@digia.com>
|
|
|
|
|
|
|
|
| |
Chromium has enabled this in the 37 based branch and it worked
with Qt WebEngine as well.
Change-Id: Idb1f9a2133be645cd044c9f0163cc87f6ae8e474
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
|
|
|
| |
This is necessary for finding gperf, bison, yacc, etc. on Windows.
Works only if the gnuwin32 directory is on the same level as the
qtwebengine root directory, which is the case with a qt checkout.
Change-Id: I09fe1523d4e4baca6612f1039bbba2e2ff9d87c0
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
| |
Change-Id: If1665b5bb7a6db07ca737aa00d0ddba964e418aa
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Chromium requires relative paths for output-dir and
for generator-output. Also output-dir is expected to
be a subdirectory of generator-output.
This now allows building on windows without running
qmake in recursive mode.
Change-Id: Ie0f9965c1fbfd63eb3cb9c360b29f210128a5c3e
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
| |
Ninja needs to have an _x64 appended to the configuration in order
to build the 64 bit targets on Windows, as opposed to Linux and OS X.
Change-Id: Idc9888c9f80ccb27d26fb86588ed39fba92a2c71
Reviewed-by: Zeno Albisser <zeno.albisser@digia.com>
|
|
|
|
|
|
|
|
| |
Those configurations should be re-enabled once we support them
properly.
Change-Id: Id8019df8e0b34a2fb1a411780689f28def50546f
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Chromium itself requires XCode 5 to build and would require
a few modifications to build with the version of libc++ that
ships with XCode 4.
Disable the build on that environment to at least get passed
the Qt5 CI.
Change-Id: I296055ac1880a9e09e2c0e8343058ff5add837c0
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@digia.com>
|
|
|
|
|
|
|
|
| |
This changed with chrome 37. We only need to define it unconditionally
in gyp_qtwebengine as tcmalloc is never an option for us.
Change-Id: I8b21eb1114582c1d9a8aa61c5f15fc8469721395
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
|
| |
We currently don't support this configuration.
Change-Id: Ibd6ad9e1ffc993a2224fb814109467bd7b65f0fc
Reviewed-by: Zeno Albisser <zeno.albisser@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We don't want to fetch the compiler path from the mkspec on Windows
the same way that we do on linux/embedded. Qt usually let the build
environment decide the MSVC toolchain, as it is usually done on
Windows.
This patch also passes the ninja_use_custom_environment_files ninja
generator option to make sure that gyp doesn't try to construct the
build environment by running the env scripts itself.
Change-Id: I6e04471eb994b81a9f1b8b5f149b4dbddbb320a8
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A gyp variable containing backslashes would be evaluated twice when
passed through grit_action.gypi in src/core/chrome_qt.gyp.
For some reason, gyp would end up reducing the escaping of
backslashes once for each variable evaluation, which would completely
remove backslashes from the value.
Fix the issue by making sure that we pass paths to gyp with slashes
instead of backslashes. Python will know what to do with them
regardless.
Change-Id: Id38866fd63a2ba1524a8c998fe577f7f50fca059
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
qtwe_chromium_obj_dir wasn't falling back to src/3rdparty if the
config couldn't be found in git.
Move the fallback logic to a getChromiumSrcDir qmake function, and
use it in both places.
The build system will now also print that the Chromium sources will
follow the git config if found instead of the other way around,
since this should now be the normal case as part of Qt5.
Change-Id: I5ab972ab53156c1bd033518e2eb51013559491ac
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Most of the patch is about upstream classes/methods that changed.
Other important details:
- icu data files are now used by default
- cygwin is no longer required to build on Windows
- RenderFrameHost has been replacing RenderViewHost in a few places,
following the separate process iframes support in Chromium
- The user agent is accessed through ContentClient::GetUserAgent
instead of from the command line switches
Change-Id: I86cc93aff7ce31176a80b0b4a5d54025674a451c
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- The svn repo string would be something like:
...branches/2062/path_suffix
and I want only the branch number in this case.
Updated git_submodule.py to support that case and use regexp
instead to keep thing a bit simpler.
- .isolate are gyp files, moved the exception at the top
- Changed some inclusions/exclusions
- tools/ is now all excluded by default with exception instead
of all included. We don't need most of it.
Change-Id: I6d0666d9af7fc6a0f3c4fe866adf2744f13f7c35
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
To allow the qt5.git integration, allow the build to succeed if the
repository is checked out on a platform that we don't maintain.
This is the initial list of platforms that we want the integration
to succeed and we can add new ones once we start maintaining them.
Change-Id: Idce2a84eb8d5b1ef7f14d399fe013f666b479108
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
| |
This allows to resolve visited links in the render process.
The newly introduced API at the Core layer should allow exposing
this functionality to our widgets and Qt quick API layers.
Change-Id: I256376afcfe79014dc274e2dddbac1986a884a93
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Use the chromium localized error strings for that purpose, otherwise the
error description is always empty.
While we're at it, let's tap into the chromium error pages, which should
hopefully make sense for most errors, and add some static asserts to
check that the qt quick enum and the core one are in sync.
Change-Id: Icf8fa7c3bf4a674c60a10950422135fb6930447a
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
This prevents python from escaping the special characters that
resulted in git not finding the proper shasum for the git-svn-id
regular expression line.
Change-Id: Ic5449db6319b638cf692d2b4427ce5de7cbac66e
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
Reviewed-by: Adam Kallai <kadam@inf.u-szeged.hu>
|
|
|
|
|
|
|
|
|
|
| |
Since some projects are maintained in git repositories only even the
'svn' DEPS file has some git shasums instead of svn revisions.
Check the current shasum and tag outside of the ref block so we actually
end up checking out the parsed shasum for non-git-svn repositories.
Change-Id: Iabc4a5b15e50a0f0df84179a65929352ee2addb7
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
Instead of grepping for the VERSION file update in the git log
check out the sha1 that has been tagged with the requested version.
Also harden the regexp we use to translate svn revisions to git
sha1's and further clean up the findShaAndCheckout function.
Change-Id: Iaeb7ebee558b7ebadaf3428e06626736601198cf
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
|
| |
This patch makes it possible to reset the upstream chromium repository
to its baseline state.
Change-Id: Idff94a0f873e994716f77c83f4c5b79bcd48aec5
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This script should be ran whenever adding new patches to our chromium
snapshot and makes sure we keep the upstream build functional.
Having clean patches is beneficial to keep it possible to upstream
patches to the Chromium project, and to keep an overview of our changes.
An annotation is not needed if the patch is adding files to the chromium
snapshot that were removed by take_snapshot.py, thus are already present
in the upstream repository.
In these cases the non-annotated patch will be ignored by the scripts.
Change-Id: I46605c559825d9da2653036e9a12a2a5730330a0
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
| |
This makes it possible to use them in other scripts as well.
Change-Id: I8aa76c1a9be91c56fd232a592ac41de5079eb3c7
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Preventing failure while applying patches on submodules that are
skipped on Windows.
Non Windows related submodules are not checked out from upstream
repository. This behavior is a workaround for git issues.
Change-Id: If34d9d2916829f87c7397a07803f79c156c05c57
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We maintain them on top of the Chromium snapshot.
A new script tools/scripts/patch_upstream.py replaced
patches/patch-chromium.sh with the recent snapshot update.
The patches that we have on top of the baseline version
are prepared from the chromium snapshot and applied on
the upstream chromium checkout.
A consequence of this is that init-repository.py -u
from now on requires a snapshot checkout as well.
Change-Id: I639c45904b72952804e5227ebc948a8859cf74dd
Reviewed-by: Zeno Albisser <zeno.albisser@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch also changes the way we handle our patches.
We'll always take a new snapshot without patching Chromium
and rebase our patches on top of the snapshot. This removes
the need for separate patch files for the ustream repository
since we can use the patches of the snapshot and apply them
on the upstream checkout if we want to use an upstream build.
This makes it easier for us to have branches for the release
and maintain and minimize the number of patches we need on
top of upstream Chromium.
For now a snapshot checkout is needed to be able to prepare
the patches for an upstream build but in future we cold improve
this to fetch the patches from the remote snapshot repository
if needed.
Change-Id: I6280ffbe2d50d25d252734bc76d19bfaaa081637
Reviewed-by: Zeno Albisser <zeno.albisser@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
With Qt 5.3 mac_sdk_path ends up being empty, so make sure to set
the proper flags for chromium to pick the right sdk.
Also add clang_use_chrome_plugins=0 so we can remove it as well from
the Mac-Use-libc-instead-of-stdlibc patch.
Change-Id: Ia132f079f905761644ad26c728f7116f538ea62d
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
|
|
| |
This will be useful when updating the snapshot and rebasing patches.
These patches need updated IDs to keep the gerrit database happy.
Change-Id: I62e9fc11a902f57497a2693cc59b873f13cfce12
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
|
|
| |
We had the logic in place to detect mocable cpp files all along,
but only used it for headers.
Change-Id: I00ea4e3ba081c5c7d37fe35ec78233e4961481e3
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
Cygwin is still required for build on Windows. It was accidentally
blacklisted. Utils in cygwin used by third_party/WebKit build and
gyp uses cygwin for wrapping linux style commands on Windows.
Change-Id: Ifac069c67ebd1bef2f41618e7688d3b64028517b
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
| |
The Windows git version is not able to check out some extra_os
dependencies but since they are not relevant on Windows
we can just skip them.
Change-Id: I552e0c3a3bf2f26a8e44901ee700d6e19c138b45
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
| |
We want to include those in the snapshot in the future.
Change-Id: Iee49f64781dba9fb6b7a16eebdf6b2013d410feb
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
| |
Running setup_mount will ensure that every build uses it's own
cygwin and prevents mixup with other instance.
Update 3rdparty's shasum.
Change-Id: Ic8f25ce7c6077318bac36b15cfe23a5a27ed51b9
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
| |
Change-Id: I49415512d9eb152a2c9a20bcc5adf9065717b124
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
SurfaceFactoryQt::loadEGLGLES2Bindings is needed to find
and load EGL and GLES2 libraries on embedded linux.
It also provides a function to query the native display.
We are using qmake variables at compile time to detect
where we should look for the libraries at runtime.
For EGLFS we query the native display through PlatformNativeInterface.
If the resource is not exposed by Qt (pre 5.4),
then we assume EGL_DEFAULT_DISPLAY, which is the right
parameter for close to 100% of the cases anyway.
OzonePlatformEglfs is just boiler plate code.
Change-Id: I3ad08a56c50d6160be653ffb168b3f1f791e5245
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
|
|
|
| |
Extend extra_os with Windows in git_submodule script.
Update blacklist in take_snapshot script.
This also updates take_snapshot.py for branch 1750.
Change-Id: Ie8a1e58534e9f5c2635ed4103d0ae62f804dd5d0
Reviewed-by: Andras Becsi <andras.becsi@digia.com>
|
|
|
|
|
|
| |
Change-Id: I85149bf4405420813d07bee5259f11a192e86a73
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
Reviewed-by: Zoltan Arvai <zarvai@inf.u-szeged.hu>
|