GitLab

BUG=107070
TBR=cmp@chromium.org

Review URL: http://codereview.chromium.org/8895021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113934 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/8872038

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113930 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113929 0039d316-1c4b-4281-b951-d872f2087c98

BUG=chromium-os:23770
TEST=None


Review URL: http://codereview.chromium.org/8890017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113928 0039d316-1c4b-4281-b951-d872f2087c98

TBR=bashi


Review URL: http://codereview.chromium.org/8893005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113927 0039d316-1c4b-4281-b951-d872f2087c98

In crrev.com/113111 I checked in code that reads a app state boolean from an 
ExtensionSyncData object which is uninitialized. This fixes the initialization
for the relevant constructor, and also limits reading that value to only when
the extension is an app.

BUG=106549
TEST=valgrind run of ExtensionService sync-related unit_tests should come back
clean

Review URL: http://codereview.chromium.org/8899010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113926 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=trybots


Review URL: http://codereview.chromium.org/8887001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113924 0039d316-1c4b-4281-b951-d872f2087c98

BUG=106013,106228
TEST=Verify zoom-in/zoom-out work correctly on PDF documents.
Review URL: http://codereview.chromium.org/8896009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113923 0039d316-1c4b-4281-b951-d872f2087c98

Add Pass(), which implements move semantics, to scoped_ptr, scoped_array, and scoped_ptr_malloc.

This modification to the scopers implements the "moveable but not copyable" semantics that were introduced in C++11's unique_ptr<>.

With this, is now possible to use scopers as an argument type or a return type.  This signifies, in the type system, transfer of ownership into a function or out of a function respectively.  Calling, or returning such a function MUST use the temporary resultin

This distinction makes it possible to avoid the implicit ownership transfer issues of auto_ptr, but still allow us to have compiler enforced ownership transfer.

Also adds a Passed() helper that allows using a scoper with Bind().

BUG=96118
TEST=new unittests

Original review URL: http://codereview.chromium.org/8774032

Review URL: http://codereview.chromium.org/8897005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113922 0039d316-1c4b-4281-b951-d872f2087c98

Split --aura-windows into --aura-workspace-manager and --aura-translucent-frames.
Also fixed a bug where the window frame was still opaque despite the latter flag.

BUG=106890
TEST=manual, run with all combinations of above flags


Review URL: http://codereview.chromium.org/8890036

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113921 0039d316-1c4b-4281-b951-d872f2087c98

base/task.h is going away and being replaced by base/bind.h and base/callback.h.
This CL was automatically generated by sed.

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/8879033

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113920 0039d316-1c4b-4281-b951-d872f2087c98

turns out the bot was just out of disk

BUG=107050
TEST=none

Review URL: http://codereview.chromium.org/8899020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113918 0039d316-1c4b-4281-b951-d872f2087c98

since closures don't work the way I thought they did.


BUG=106421
TEST=Manual


Review URL: http://codereview.chromium.org/8830001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113917 0039d316-1c4b-4281-b951-d872f2087c98

BUG=98345
TEST=none
R=ben@chromium.org


Review URL: http://codereview.chromium.org/8892013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113916 0039d316-1c4b-4281-b951-d872f2087c98

BUG=106575
TEST=about:histograms


Review URL: http://codereview.chromium.org/8895016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113915 0039d316-1c4b-4281-b951-d872f2087c98

Allow the users of the utility process to specify whether the zygote
should be used on Linux. This will allow sandboxing the uses of the
utility process that don't do FS access.

This is the first step to get the utility process sandboxed
on Linux. Since most of the uses of the utility process
don't do file access, launching all of those from the zygote
will simplify sandboxing the one that does: extension
unpacking.

Subsequent CLs will port the uses of the utility process
that don't do file access to launch from the zygote.

BUG=93109
TEST=No functional changes yet, but test that nothing broke by
running browser_tests::ExtensionWebstorePrivateApiTest.* and also
logging into Chromium and installing an extension from the Web Store.


Review URL: http://codereview.chromium.org/8849007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113914 0039d316-1c4b-4281-b951-d872f2087c98

We currently have several copies of the code that draws the non-client frame area of windows, including OpaqueBrowserFrameView, ConstrainedWindowFrameViews, CustomFrameView, and others.  We need yet another version for Aura.  Rather than make another copy I'm trying to refactor out the common drawing code.  This is step 1, creating a views::FrameBackground class and using it in two cases.

BUG=102797
TEST=visual, try normal frame, install a theme, open a constrained window like cookie > Show cookies and other site data, open a popup from popuptest.com


Review URL: http://codereview.chromium.org/8772060

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113913 0039d316-1c4b-4281-b951-d872f2087c98

BUG=106979
TEST=none


Review URL: http://codereview.chromium.org/8896016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113912 0039d316-1c4b-4281-b951-d872f2087c98

replace with add/remove

BUG=102610
TEST=manual (for those strings I actually know how to trigger)

with fix for aura


Review URL: http://codereview.chromium.org/8879022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113911 0039d316-1c4b-4281-b951-d872f2087c98

This uses the GpuVideoDecodeAccelerator machinery (already written to enable
ppapi to take advantage of OpenMAX HW where available) to decode <video> data.
This increases idle CPU from 20% to 45% on one particularly large (internal)
test video (red0.mp4), on an ARM crosbook.

HW decode is done on a best-effort basis; if the GPU code doesn't know how to
deal with a codec/profile we still fall back to ffmpeg for decode.  Because the
vast majority of chrome installs will be on HW with no video decode support
(yet) we only attempt HW video decode on platforms we know have a shot at it.

BUG=104579
TEST=manual testing w/ video test matrix, trybots.

Review URL: http://codereview.chromium.org/8686010

TBR=fischman@chromium.org
Review URL: http://codereview.chromium.org/8897022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113908 0039d316-1c4b-4281-b951-d872f2087c98

This CL is mostly motivated to allow users of the web request API to filter network traffic to the extension. The URLParser was configured to broadly before in that is allowed parsing URLPattern::SCHEME_ALL.

BUG=105656
TEST=no


Review URL: http://codereview.chromium.org/8800006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113907 0039d316-1c4b-4281-b951-d872f2087c98

Also change some functions to return ModelEnumSets directly instead of
taking a pointer value.

BUG=79970
TEST=


Review URL: http://codereview.chromium.org/8851006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113906 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/8893020

TBR=fischman@chromium.org
Review URL: http://codereview.chromium.org/8893022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113905 0039d316-1c4b-4281-b951-d872f2087c98

Since GTK initialization loads shared libraries and accesses
the file system, having this code present in every code path
that uses the utility process would prevent us from sandboxing
the uses of the utility process that don't access the filesystem
themselves.

Moreover, since GTK initialization is only used for plugin loading,
which cannot be sandboxed, there is no harm in moving it to the
plugin loading path. We do check that the GTK thread system has not
already been initialized (only needed for glib < 2.24):

http://developer.gnome.org/glib/2.28/glib-Threads.html#g-thread-init

BUG=93109
TEST=Similar to issue 8774048, having libtotem plugins installed,
go to about:plugins and check that they are present.


Review URL: http://codereview.chromium.org/8870005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113904 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/8879028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113903 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113902 0039d316-1c4b-4281-b951-d872f2087c98

BUG=104046
TEST=run webrequest extension with chrome.webRequest.onBeforeSendHeaders.addListener(function(details) {return {cancel: true};}, {urls: ["<all_urls>"]}, ["blocking"]);


Review URL: http://codereview.chromium.org/8894003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113901 0039d316-1c4b-4281-b951-d872f2087c98

This picks up:

r1104 - Fix CHROMIUM_STRIP_SAVE_FILE

BUG=106859
TEST=none


Review URL: http://codereview.chromium.org/8895007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113900 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/8893020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113899 0039d316-1c4b-4281-b951-d872f2087c98

InProcessBrowserTest::CreateIncognitoBrowser() runs its own message loop
as it waits for navigation to complete, which resulted in a race between
the navigation to about:blank completing and trying to close the
Incognito browser.

BUG=106707

Review URL: http://codereview.chromium.org/8892006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113898 0039d316-1c4b-4281-b951-d872f2087c98

Regressions:
xp-release-dual-core/intl2/times/t: http://build.chromium.org/f/chromium/perf/xp-release-dual-core/intl2/report.html?history=150&rev=113640

BUG=106888
TBR=nsylvain@chromium.org


Review URL: http://codereview.chromium.org/8892008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113897 0039d316-1c4b-4281-b951-d872f2087c98

base/task.h is going away and being replaced by base/bind.h and base/callback.h.
This CL was automatically generated by sed.

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/8873032

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113896 0039d316-1c4b-4281-b951-d872f2087c98

This uses the GpuVideoDecodeAccelerator machinery (already written to enable
ppapi to take advantage of OpenMAX HW where available) to decode <video> data.
This increases idle CPU from 20% to 45% on one particularly large (internal)
test video (red0.mp4), on an ARM crosbook.

HW decode is done on a best-effort basis; if the GPU code doesn't know how to
deal with a codec/profile we still fall back to ffmpeg for decode.  Because the
vast majority of chrome installs will be on HW with no video decode support
(yet) we only attempt HW video decode on platforms we know have a shot at it.

BUG=104579
TEST=manual testing w/ video test matrix, trybots.

Review URL: http://codereview.chromium.org/8686010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113895 0039d316-1c4b-4281-b951-d872f2087c98

base/task.h is going away and being replaced by base/bind.h and base/callback.h.
This CL was automatically generated by sed.

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/8872030

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113894 0039d316-1c4b-4281-b951-d872f2087c98

BUG=79722
TEST=none


Review URL: http://codereview.chromium.org/8897006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113893 0039d316-1c4b-4281-b951-d872f2087c98

ExtensionDownloadsEventRouter's ctor pushes a Callback to this queue instead of calling GetDownloadManager(), which would create the manager. EDER doesn't really need the manager in its ctor, it just needs the manager before the manager does something interesting.

BUG=106502

Review URL: http://codereview.chromium.org/8804018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113892 0039d316-1c4b-4281-b951-d872f2087c98

Pasteboard corruption or bad third-party code can cause
+[NSSpellChecker sharedSpellChecker] to throw NSExceptions.  Chrome
treats NSExceptions as fatal, because they generally mean things are
awry.  Since this call can happen in response to calls from the
renderer, and the exceptions are not thrown from Chrome code, this
change enables them while retrieving the shared instance.

BUG=92399,95728


Review URL: http://codereview.chromium.org/8883040

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113891 0039d316-1c4b-4281-b951-d872f2087c98

The comments on both methods I'm checking point out that they need to
be NULL checked. Hopefully, adding the NULL checks fixes the crashes.

BUG=107015
TEST=push to M17, see if it fixes crash in wild.


Review URL: http://codereview.chromium.org/8893010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113890 0039d316-1c4b-4281-b951-d872f2087c98

Properly handle dependencies for installing the ppd
Register for uninstall


BUG=106094
TEST=Copy port monitor dlls, virtual_driver_setup.exe, and gcp-driver.ppd into a single folder and launch the setup as admin.  The driver should be installed.  Run it again with --uninstall (or choose the cloud print driver in add/remove programs) and the driver should be uninstalled and the containing folder removed.


Review URL: http://codereview.chromium.org/8770019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113889 0039d316-1c4b-4281-b951-d872f2087c98

If the PasswordStore is asked to add an entry that already exists it
will overwrite the existing entry.  When this happens, it will report a
PasswordStoreChange::ADD event, much like it would had the original
entry not existed.  This confuses the PasswordChangeProcessor which
assumes that an ADD event implies that the added item did not previously
exist.

This change works around the issue by allowing the password change
processor to try to update an existing node in the event that the
creation of a new node fails.  This allows it to handle the case where a
it receives an ADD event for an existing node.

A better solution would be to have the PasswordStore to always return an
UPDATE event in cases where the node being updated already exists.  If
that solution were implemented the PasswordChangeProcessor would be
right to report an UnrecoverableError if it is unable to create a new
sync node in response to an ADD event.

BUG=87855
TEST=See repro steps in crbug.com/87855, crbug.com/103455


Review URL: http://codereview.chromium.org/8862005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113888 0039d316-1c4b-4281-b951-d872f2087c98