GitLab

Revert of Try backing SkPicture with SkRecord in Chromium. (patchset #4 id:60001 of https://codereview.chromium.org/504823003/)

Reason for revert:
https://code.google.com/p/chromium/issues/detail?id=411330

Original issue's description:
> Try backing SkPicture with SkRecord in Chromium.
>
> This is our first attempt, mostly to feel this out.  Performance and
> correctness problems are both possible.  Please revert this if a benchmark even
> smells funny.
>
> That said, we've been working on this a while, are pretty confident it's good
> stuff, and have had Skia's internal testing entirely switched over for a week.
> Parts of Chromium that serialize and deserialize SkPictures are already
> switched over.  This CL finishes things off by switching the default
> recorded-from-scratch SkPicture to use SkRecord too.
>
> BUG=408985,409110,409138,411330
>
> Committed: https://chromium.googlesource.com/chromium/src/+/f7450daaf38f4951b23fabaaaf659c6af33a705c
>
> Committed: https://chromium.googlesource.com/chromium/src/+/5f2a6ab31313eb2fc6e456f65c3863a77d2d0c30
>
> CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg;tryserver.chromium.linux:linux_browser_asan
>
> Committed: https://chromium.googlesource.com/chromium/src/+/130033d6c732af90ed117f05517e4efffd23ae58
>
> Committed: https://chromium.googlesource.com/chromium/src/+/96730ee0419e528d8d3eac360e9319a871d20e43

TBR=reed@google.com,mtklein@google.com,senorblanco@chromium.org,eroman@chromium.org,fmalita@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=408985,409110,409138,411330

Review URL: https://codereview.chromium.org/544263002

Cr-Commit-Position: refs/heads/master@{#293544}

BUG=306341
NOTRY=True

Review URL: https://codereview.chromium.org/532903002

Cr-Commit-Position: refs/heads/master@{#293543}

This reverts commit 5dde0685.

Broke a bunch of builds:

FAILED: /mnt/data/b/build/goma/gomacc ../../third_party/llvm-build/Release+Asserts/bin/clang++ -MMD -MF obj/sync/internal_api/attachments/sync_core.attachment_service_proxy.o.d -DSYNC_IMPLEMENTATION -DCHROMIUM_BUILD -DENABLE_NOTIFICATIONS -DENABLE_EGLIMAGE=1 -DENABLE_BACKGROUND=1 -DV8_DEPRECATION_WARNINGS -DCLD_VERSION=2 -DENABLE_MDNS=1 -DENABLE_PEPPER_CDMS -DENABLE_PLUGINS=1 -DENABLE_PRINTING=1 -DENABLE_FULL_PRINTING=1 -DENABLE_SPELLCHECK=1 -DUSE_UDEV -DTOOLKIT_VIEWS=1 -DUI_COMPOSITOR_IMAGE_TRANSPORT -DUSE_ASH=1 -DUSE_AURA=1 -DUSE_CAIRO=1 -DUSE_CLIPBOARD_AURAX11=1 -DUSE_DEFAULT_RENDER_THEME=1 -DUSE_GLIB=1 -DUSE_NSS=1 -DUSE_X11=1 -DUSE_XI2_MT=2 -DDISABLE_NACL -DENABLE_EXTENSIONS=1 -DENABLE_CONFIGURATION_POLICY -DENABLE_TASK_MANAGER=1 -DENABLE_THEMES=1 -DENABLE_CAPTIVE_PORTAL_DETECTION=1 -DENABLE_SESSION_SERVICE=1 -DENABLE_APP_LIST=1 -DENABLE_SETTINGS_APP=1 -DENABLE_MANAGED_USERS=1 -DENABLE_SERVICE_DISCOVERY=1 -DENABLE_AUTOFILL_DIALOG=1 -DENABLE_REMOTING=1 -DENABLE_GOOGLE_NOW=1 -DENABLE_ONE_CLICK_SIGNIN -D_FILE_OFFSET_BITS=64 -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -DPROTOBUF_USE_DLLS -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -I../.. -Igen -I/usr/include/glib-2.0 -I/usr/lib/x86_64-linux-gnu/glib-2.0/include -I../../net/third_party/nss/ssl -I/usr/include/nss -I/usr/include/nspr -I../../third_party/zlib -Igen/protoc_out -I../../third_party/protobuf/src -I../../third_party/protobuf -fno-strict-aliasing -fstack-protector --param=ssp-buffer-size=4 -m64 -funwind-tables -fPIC -pipe -pthread -fcolor-diagnostics -Wall -Wsign-compare -Wendif-labels -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-char-subscripts -Wno-covered-switch-default -Wno-deprecated-register -Wno-unused-function -fvisibility=hidden -Xclang -load -Xclang ../../third_party/llvm-build/Release+Asserts/lib/libFindBadConstructs.so -Xclang -add-plugin -Xclang find-bad-constructs -Wheader-hygiene -Wstring-conversion -fno-ident -fdata-sections -ffunction-sections -O2 -g2 -Wexit-time-destructors -fno-threadsafe-statics -fvisibility-inlines-hidden -std=gnu++11 -Wno-reserved-user-defined-literal -fno-rtti -fno-exceptions -c ../../sync/internal_api/attachments/attachment_service_proxy.cc -o obj/sync/internal_api/attachments/sync_core.attachment_service_proxy.o
../../sync/internal_api/attachments/attachment_service_proxy.cc:96:3: error: invalid argument type 'scoped_refptr<base::SequencedTaskRunner>' to unary expression
  DCHECK(wrapped_task_runner_);
  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
../../base/logging.h:622:51: note: expanded from macro 'DCHECK'
  LAZY_STREAM(LOG_STREAM(DCHECK), DCHECK_IS_ON && !(condition))   \
  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~
../../base/logging.h:355:5: note: expanded from macro 'LAZY_STREAM'
  !(condition) ? (void) 0 : ::logging::LogMessageVoidify() & (stream)
    ^
1 error generated.
ninja: build stopped: subcommand failed.

BUG=none
TBR=maniscalco

Review URL: https://codereview.chromium.org/546873003

Cr-Commit-Position: refs/heads/master@{#293542}

It lives in src/extensions now, but has src/chrome dependencies and tests
code implemented in src/chrome. Attempting to refactor it to split out the
src/chrome dependencies just made a mess of the code. Since it tests Chrome
concepts, put it there.

BUG=397164
TEST=unit_tests

Review URL: https://codereview.chromium.org/546633003

Cr-Commit-Position: refs/heads/master@{#293541}

Fixes several clang/win errors like
..\..\sandbox\win\sandbox_poc\pocdll\registry.cc(24,10) :  error(clang): value of type 'HKEY' (aka 'HKEY__ *') is not implicitly convertible to 'LONG_PTR' (aka 'long')
    case HKEY_USERS:
         ^~~~~~~~~~

Also fix a format string warning while here.

No intended behavior change.
BUG=82385

Review URL: https://codereview.chromium.org/542863002

Cr-Commit-Position: refs/heads/master@{#293540}

Deprecation messages are added in:
* getChannelId function - displayed to console
* documentation of chrome.pushMessaging/GCM for Chrome V1

BUG=403952

Review URL: https://codereview.chromium.org/476883003

Cr-Commit-Position: refs/heads/master@{#293539}

https://skia.googlesource.com/skia/+log/7675fb23a0448448662567bf1d100e39bf7b5e65..a1ae66d252edf6da932caed1fe43d11216e56c0e

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=jvanverth@google.com

Review URL: https://codereview.chromium.org/543083002

Cr-Commit-Position: refs/heads/master@{#293538}

Since '\0' delimited concatenation of header field values is not allowed any
more in HTTP/2 draft-14, but servers might prepare such headers, the HPACK
encoder should split them into multiple headers with identical names.

This lands server change 74384948 by bnc.

BUG=400336

Review URL: https://codereview.chromium.org/533073004

Cr-Commit-Position: refs/heads/master@{#293537}

This change is lays groundwork for making attachment upload operations
persistent.

Make GenericChangeProcessor responsible for writing attachments to the
AttachmentStore and calling UploadAttachments.  In a future CL, datatype
code will be responsible for writing attachments to the store.

Queue up attachments for upload inside of AttachmentService.  In a
future CL we'll add rate limiting, retry, and back-off logic.

Expose AttachmentService's AttachmentStore via GetStore method.

BUG=

Review URL: https://codereview.chromium.org/512413003

Cr-Commit-Position: refs/heads/master@{#293536}

The Lock and Guest functions currently try to close all browser windows for their profiles, and any OnBeforeUnload dialogs can cause that closing to fail. This CL takes inspiration from the BrowserCloseManager that is used when an entire browser window is closed.

Methods are added to BrowserList that will trigger all OnBeforeUpload handlers, wait for the user's response, and then close the profile's browsers and perform an appropriate success callback. A abort callback is permitted and will be used when ProfileDeletion also uses this flow.

Guest and Lock will now only open the User Manager (and lock the profile, for Lock) if no OnBeforeUnload event is canceled.

BUG=368497, 289390
TEST=Open a guest profile. Navigate to http://www.4guysfromrolla.com/demos/OnBeforeUnloadDemo1.htm. Click "Exit Guest" in the User Menu. Select "Stay on this page". The window should stay open and the User Manager should not be shown.
Open a signed-in and lockable profile. Navigate to the above URL. Lock the profile from the User Menu. Select "Stay on this page". The User Manager should not be shown. Manually opening the User Manager should show that the Profile has not been locked.

Review URL: https://codereview.chromium.org/471763008

Cr-Commit-Position: refs/heads/master@{#293535}

Treat a header block with a pseudo-header field following a regular one as
malformed, as required by HTTP/2 draft-14 Section 8.1.2.1.

This lands server change 74067938 by bnc.

BUG=400336

Review URL: https://codereview.chromium.org/531253004

Cr-Commit-Position: refs/heads/master@{#293534}

BUG=352654

Review URL: https://codereview.chromium.org/547653002

Cr-Commit-Position: refs/heads/master@{#293533}

It should be refreshed but it was not by bug. This patch fixes this.

BUG=none
TEST=manual
TBR=fukino@chromium.org

Review URL: https://codereview.chromium.org/543103002

Cr-Commit-Position: refs/heads/master@{#293532}

depends on blink CL https://codereview.chromium.org/540033002/

TBR=jochen

Review URL: https://codereview.chromium.org/545513002

Cr-Commit-Position: refs/heads/master@{#293531}

The parentheses were wrong previously in this if-statement:
if ((wcsstr(str_ptr, L"(VFW)") == NULL) &&
     lstrlenW(str_ptr) < name_length ||
    (!(LowerCaseEqualsASCII(str_ptr, str_ptr + name_length,
                                   kGoogleCameraAdapter))))
The condition lstrlenW(str_ptr) < name_length should be grouped with the condition LowerCaseEqualsASCII(...), not wcsstr(str_ptr, L"(VFW)") == NULL.

The check lstrlenW(str_ptr) < name_length is redundant, so I removed it.

Review URL: https://codereview.chromium.org/541603002

Cr-Commit-Position: refs/heads/master@{#293530}

BUG=405449
TEST=Forced WebRtcAudioCapturer::Initialize() to return false and verified log.

Review URL: https://codereview.chromium.org/545203002

Cr-Commit-Position: refs/heads/master@{#293529}

1. Add retry to http_client.
2. Make Findit more robust for unrecognized chromium revision.
3. Workaround a harmless bug in python 2.7
4. Give a better error message for check failure.

NOTRY=true

Review URL: https://codereview.chromium.org/538383002

Cr-Commit-Position: refs/heads/master@{#293528}

BUG=410446
TEST=manual. launch webstore app from home card.

Review URL: https://codereview.chromium.org/544813002

Cr-Commit-Position: refs/heads/master@{#293527}

TBR=schenney@chromium.org

Review URL: https://codereview.chromium.org/544173002

Cr-Commit-Position: refs/heads/master@{#293526}

Chromium is only using a Service Provider, and not an Application. Moreover
chromium should only depends on mojo_base.gyp, not mojo.gyp.

R=darin@chromium.org,davemoore@chromium.org,blundel@chromium.org

Review URL: https://codereview.chromium.org/531513003

Cr-Commit-Position: refs/heads/master@{#293525}

The code that copies cookies set by a SAML IdP on subsequent logins is
wrong as it uses a method which modifies the cookie being copied by
prepending a dot to the domain name. This CL fixes the problem by
switching to the method that is used during initial login. The method
is also renamed to make it clearer that it may be used on subsequent
logins too.

BUG=410416
TEST=unit_tests updated to detect this problem

TBR=dbeam (wallet_signin_helper_unittest.cc)

Review URL: https://codereview.chromium.org/538543003

Cr-Commit-Position: refs/heads/master@{#293524}

BUG=410891
TBR=henrika
NOTRY=true

Review URL: https://codereview.chromium.org/542173002

Cr-Commit-Position: refs/heads/master@{#293523}

NWVM was using the ScopedCaptureClient to attach a capture controller. The problem with this is that the window manager already made one, and there is only one capture controller when you use SCC and this was screwing with event dispatch as coordinate systems between the different root windows were different (not sure this is a condition we've ever really designed for in aura). Anyway. Made NWVM use DefaultCaptureClient directly.

R=sky@chromium.org
BUG=

Review URL: https://codereview.chromium.org/541013002

Cr-Commit-Position: refs/heads/master@{#293522}

Review URL: https://codereview.chromium.org/536533004

Cr-Commit-Position: refs/heads/master@{#293521}

BUG=380349

Committed: https://chromium.googlesource.com/chromium/src/+/bbb7ea7c66f18298079eea4a6b24834986f4a822

Review URL: https://codereview.chromium.org/479873002

Cr-Commit-Position: refs/heads/master@{#293520}

It used to work after my previous fix 37accc2c, but then it broke again.
This CL changes the way how "-I" arguments are transformed.

BUG=NONE
TEST=manually
NOTRY=true

Review URL: https://codereview.chromium.org/528293002

Cr-Commit-Position: refs/heads/master@{#293519}

Per RFC 4122 section 4.4, the format of GUID version 4 must be
xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx, where y is one of [8, 9, A, B].
This CL updates base::GenerateGUID() on POSIX to be compliant with the
GUID v4 format.

This fix is proposed by Alex Vakulenko <avakulenko@chromium.org>

BUG=None
TEST=base_unittests

Review URL: https://codereview.chromium.org/541633005

Cr-Commit-Position: refs/heads/master@{#293518}

Extract ResultCatcher from ExtensionApiTest. Use it in ShellApiTest, which is separated out from AppShellTest.

Move the remaining DnsApiTest to app_shell_browsertests.

BUG=388893

Review URL: https://codereview.chromium.org/529293003

Cr-Commit-Position: refs/heads/master@{#293517}

Currently interactive UI tests typically call `ClickOnView` to transfer
focus. On Mac, the behaviour of a click changes when an application
loses focus, and re-activating the application in a test is not
reliable. The test application shouldn't be losing focus in
interactive_ui_tests, but tracing suggests this is ocurring anyway.

This CL introduces ui_test_utils::FocusView which is a more robust way
to transfer focus in a test.

OmniboxViewTest.BeginningShownAfterBlur is changed to use
ui_test_utils::FocusView and re-enabled.

BUG=406012

Review URL: https://codereview.chromium.org/534993002

Cr-Commit-Position: refs/heads/master@{#293516}

https://chromium.googlesource.com/chromium/blink/+log/e76e0d3c27477b816754225fcdbcae5c79f6940f..dcd4f892fb78a6f808982633fbc7398714f157be
TBR=arv@chromium.org,keishi@chromium.org

Review URL: https://codereview.chromium.org/544073002

Cr-Commit-Position: refs/heads/master@{#293515}

Make ExtensionManagement listening to preference changes.

Move all functionality of admin_policy into StandardManagementPolicyProvider, the latter was essentially a wrapper over admin_policy before.

Merge admin_policy_unittest into extension_management_unittests with minor changes.

BUG=177351
TEST=ExtensionServiceTest,ExtensionManagementTest,ExtensionAdminPolicyTest

Review URL: https://codereview.chromium.org/500043003

Cr-Commit-Position: refs/heads/master@{#293514}

Instead of posting input and compositor tasks directly to the
main thread message loop, forward them to the Blink scheduler
so they can be prioritized over other tasks.

BUG=391005

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=290845

Review URL: https://codereview.chromium.org/363383002

Cr-Commit-Position: refs/heads/master@{#293513}

As side effect, this APIs will also be available for the app_shell.
The patch also removes the dependence this API has on the g_browser_process
and thus, no longer will depend on chrome/.

BUG=392842

Review URL: https://codereview.chromium.org/389633002

Cr-Commit-Position: refs/heads/master@{#293512}

Cr-Commit-Position: refs/heads/master@{#293511}

To support unbundled updates we need to stop depending on system
libraries. Disable using the system version of ICU and instead link
Chromium's copy.

This increases binary size by ~7MB.

BUG=409851

Review URL: https://codereview.chromium.org/541753002

Cr-Commit-Position: refs/heads/master@{#293510}

Previously, Android's enhanced bookmarks feature depended on
--manual-enhanced-bookmarks flag and nothing else. But the
correct way is to use UpdateBookmarksExperimentState function's logic,
which desktop has been using.

BUG=386785

Review URL: https://codereview.chromium.org/497563002

Cr-Commit-Position: refs/heads/master@{#293509}

Folloup to https://codereview.chromium.org/537993002/

BUG=410824

Review URL: https://codereview.chromium.org/545173002

Cr-Commit-Position: refs/heads/master@{#293508}

Re-introduce screen manager

BUG=231331
R=ygorshenin@chromium.org

Review URL: https://codereview.chromium.org/539483002

Cr-Commit-Position: refs/heads/master@{#293507}

TBR=jarin@chromium.org

Review URL: https://codereview.chromium.org/546743002

Cr-Commit-Position: refs/heads/master@{#293506}

While working on issue 410276, in particular in
https://codereview.chromium.org/542733002/ , I had to update the
tracing cache backend. Why? Does anyone use it?

Let's remove it instead.

R=pasko@chromium.org,clamy@chromium.org
BUG=None

Review URL: https://codereview.chromium.org/544923002

Cr-Commit-Position: refs/heads/master@{#293505}