GitLab

When a new version of Chrome has been downloaded but the old version is
still running, any app shims that start will load the new framework version.
This is problematic as the IPC codes (See ipc_message_start.h) may not match.

With this CL, app shims will check a version file written by the Chrome
process and load the corresponding framework version.

This also increments kCurrentAppShortcutsVersion, which causes all app
shims to be rebuilt.

This also moves AppModeChromeLocatorTest to browser_tests. It's not really
necessary for them to have their own target (app_mode_app_tests), and this
way they actually get run.

BUG=405347

Review URL: https://codereview.chromium.org/501303002

Cr-Commit-Position: refs/heads/master@{#292563}

BUG=324071,374440,408737

Review URL: https://codereview.chromium.org/522463002

Cr-Commit-Position: refs/heads/master@{#292562}

BUG=170859
TBR=sgurun, yzshen

Review URL: https://codereview.chromium.org/516833002

Cr-Commit-Position: refs/heads/master@{#292561}

With the switch to pure frame-based calculation of time in 5b6ce11b,
these methods are no longer used.

BUG=370634

Review URL: https://codereview.chromium.org/516113002

Cr-Commit-Position: refs/heads/master@{#292560}

Enabling /Zc:inline on static release build results in

FAILED: E:\b\depot_tools\python276_bin\python.exe gyp-win-tool link-with-manifests environment.x64 True chrome.dll "E:\b\depot_tools\python276_bin\python.exe gyp-win-tool link-wrapper environment.x64 False link.exe /nologo /IMPLIB:chrome.dll.lib /DLL /OUT:chrome.dll @chrome.dll.rsp" 2 mt.exe rc.exe "obj\chrome\chrome_main_dll.chrome.dll.intermediate.manifest" obj\chrome\chrome_main_dll.chrome.dll.generated.manifest ..\..\chrome\app\chrome.dll.manifest
browser_2.lib(browser_2.character_encoding.obj) : error LNK2019: unresolved external symbol "public: bool __cdecl l10n_util::StringComparator<class std::basic_string<wchar_t,struct std::char_traits<wchar_t>,class std::allocator<wchar_t> > >::operator()(class std::basic_string<wchar_t,struct std::char_traits<wchar_t>,class std::allocator<wchar_t> > const &,class std::basic_string<wchar_t,struct std::char_traits<wchar_t>,class std::allocator<wchar_t> > const &)" (??R?$StringComparator@V?$basic_string@_WU?$char_traits@_W@std@@V?$allocator@_W@2@@std@@@l10n_util@@QEAA_NAEBV?$basic_string@_WU?$char_traits@_W@std@@V?$allocator@_W@2@@std@@0@Z) referenced in function "void __cdecl std::_Adjust_heap<struct CharacterEncoding::EncodingInfo *,__int64,struct CharacterEncoding::EncodingInfo,class l10n_util::StringComparator<struct CharacterEncoding::EncodingInfo> >(struct CharacterEncoding::EncodingInfo *,__int64,__int64,struct CharacterEncoding::EncodingInfo &&,class l10n_util::StringComparator<struct CharacterEncoding::EncodingInfo>)" (??$_Adjust_heap@PEAUEncodingInfo@CharacterEncoding@@_JU12@V?$StringComparator@UEncodingInfo@CharacterEncoding@@@l10n_util@@@std@@YAXPEAUEncodingInfo@CharacterEncoding@@_J1$$QEAU12@V?$StringComparator@UEncodingInfo@CharacterEncoding@@@l10n_util@@@Z)

browser_1.lib(browser_1.background_application_list_model.obj) : error LNK2001: unresolved external symbol "public: bool __cdecl l10n_util::StringComparator<class std::basic_string<wchar_t,struct std::char_traits<wchar_t>,class std::allocator<wchar_t> > >::operator()(class std::basic_string<wchar_t,struct std::char_traits<wchar_t>,class std::allocator<wchar_t> > const &,class std::basic_string<wchar_t,struct std::char_traits<wchar_t>,class std::allocator<wchar_t> > const &)" (??R?$StringComparator@V?$basic_string@_WU?$char_traits@_W@std@@V?$allocator@_W@2@@std@@@l10n_util@@QEAA_NAEBV?$basic_string@_WU?$char_traits@_W@std@@V?$allocator@_W@2@@std@@0@Z)

So, move StringComparator<base::string16>::operator() into header and
mark it inline, so that it's available in all translation units where it's
used. The previous code was correct, but this makes all compilers happy.

/Zc:inline ref: http://msdn.microsoft.com/en-us/library/dn642448.aspx

R=rnk@chromium.org
TBR=jshin@chromium.org
BUG=350018

Review URL: https://codereview.chromium.org/518533002

Cr-Commit-Position: refs/heads/master@{#292559}

BUG=377978

Review URL: https://codereview.chromium.org/492133002

Cr-Commit-Position: refs/heads/master@{#292558}

This replaces usage of NPObject in pepper with gin-backed V8 objects. It is unfortunate that this CL is so large, but there isn't a nice way to have the old implementation and the new one side-by-side.

There are 4 major parts to this CL:
1) Changing the HostVarTracker to track V8ObjectVars rather than NPObjectVars (host_var_tracker.cc).
2) Changing plugin elements (in plugin_object.cc) to be gin-backed objects.
3) Changing postMessage bindings (message_channel.cc) be gin-backed objects.
4) Changing the implementation of PPB_Var_Deprecated (ppb_var_deprecated_impl.cc) to call directly into V8.

BUG=351636

Committed: https://chromium.googlesource.com/chromium/src/+/21f446ae855d60cc896b40cb9a3249ed07f150b3

Review URL: https://codereview.chromium.org/459553003

Cr-Commit-Position: refs/heads/master@{#292557}

needed for this CL
https://codereview.chromium.org/517123002/

TBR=

Review URL: https://codereview.chromium.org/517803002

Cr-Commit-Position: refs/heads/master@{#292556}

BUG=407974

Review URL: https://codereview.chromium.org/510353003

Cr-Commit-Position: refs/heads/master@{#292555}

We don't support msvs any more anyway, and clobbering all of build/ is
always a bad idea. (I'm not sure this'll fix anything, but maybe we'll
get a better error.)

R=kbr@chromium.org,iannucci@chromium.org
BUG=408839

Review URL: https://codereview.chromium.org/519503002

Cr-Commit-Position: refs/heads/master@{#292554}

MIDI port id must be unique, as specified in the spec.

BUG=341423

Review URL: https://codereview.chromium.org/516893002

Cr-Commit-Position: refs/heads/master@{#292553}

The new icon is a better fit in terms of size compared to the previous one.

Review URL: https://codereview.chromium.org/518493002

Cr-Commit-Position: refs/heads/master@{#292552}

Chromium shares a message loop with Android on the browser UI thread.
This can cause problems when the associated Looper has a sync barrier,
preventing posted Chromium tasks from being dispatched until the
barrier is removed. Make this sharing more fair by marking all Chromium
Message tasks as asynchronous, avoiding stalls when there is a sync
barrier.

Note: This change is for gathering data about the perf impact and we'll
revert it before cutting a release branch.

BUG=407149,380781,407133

Review URL: https://codereview.chromium.org/512333002

Cr-Commit-Position: refs/heads/master@{#292551}

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=

Committed: https://chromium.googlesource.com/chromium/src/+/f7450daaf38f4951b23fabaaaf659c6af33a705c

Review URL: https://codereview.chromium.org/504823003

Cr-Commit-Position: refs/heads/master@{#292550}

NOTRY=true

Review URL: https://codereview.chromium.org/524433003

Cr-Commit-Position: refs/heads/master@{#292549}

This teaches the native viewport service to accept a surface id from its
client and draw it into the native viewport. This won't work if the
client also tries to bind an onscreen context to the native viewport's
view id, so the surface initialization is lazy and only happens when
the client specifies a surface id. If a client does both it'll blow up.

R=sky@chromium.org

Review URL: https://codereview.chromium.org/510553002

Cr-Commit-Position: refs/heads/master@{#292548}

This test should now be working after the fixes to the screenshotting code
that initialy had made it fail.

Also adding an OWNERS file for the Feedback API directory.

R=kalman@chromium.org
BUG=369886

Review URL: https://codereview.chromium.org/516083002

Cr-Commit-Position: refs/heads/master@{#292547}

Instead of explicitly depending on specific browser threads the USB
service can assume that it is instantiated on BrowserThread::FILE (or
equivalent) and save a TaskRunner reference from this instantiation for
later use when called from other threads.

To reach BrowserThread::UI (required for DBus on Chrome OS) a reference
to the appropriate TaskRunner must be provided when calling
UsbService::GetInstance().

BUG=

Review URL: https://codereview.chromium.org/507503002

Cr-Commit-Position: refs/heads/master@{#292546}

The current warning is the generic "Read your browsing history"
warning. The new warning for topSites will be suppressed by
read only and read/write history warnings, and all hosts warning.

BUG=404334

Review URL: https://codereview.chromium.org/482953003

Cr-Commit-Position: refs/heads/master@{#292545}

In preparation for changing how we compute media time, add some tests to
verify the current behaviour.

BUG=370634

Review URL: https://codereview.chromium.org/517843003

Cr-Commit-Position: refs/heads/master@{#292544}

This is in preparation for implementing the Trepn profiler.

BUG=

Review URL: https://codereview.chromium.org/510943002

Cr-Commit-Position: refs/heads/master@{#292543}

Now we pass Demuxer and Renderer directly into Pipeline, and create TextRenderer
directly in Pipeline. As a result, we don't need FilterCollection anymore.

BUG=408602
TEST=All existing tests pass.

Review URL: https://codereview.chromium.org/511323003

Cr-Commit-Position: refs/heads/master@{#292542}

These CLs implement a simple frame recording and fetch "extension" for the Chromoting protocol, to allow sample sessions to be captured for performance testing.

The three main classes introduced are:
- HostExtensionSessionManager
  This pulls logic for handling Chromoting extensions out of
  ClientSession instances into a dedicated manager class.
  It also introduces hooks through which extensions can
  wrap or replace the Chromoting video encoder or capturer.
- VideoFrameRecorder
  This allows a VideoEncoder to be wrapped to return a new
  encoder that will optionally copy & deliver frames to the
  VideoFrameRecorder before supplying them to the real
  encoder.
- VideoFrameRecorderHostExtension
  This extension uses a VideoFrameRecorder to allow a
  connected client to start/stop recording, and to retrieve
  the resulting frame data.

Original CLs:
crrev.com/402233003
crrev.com/339073002
crrev.com/372943002
crrev.com/462503002

BUG=260879

Review URL: https://codereview.chromium.org/468613002

Cr-Commit-Position: refs/heads/master@{#292541}

Background color issue should already be fixed by
crbug.com/391462

BUG=384559

Review URL: https://codereview.chromium.org/517153002

Cr-Commit-Position: refs/heads/master@{#292540}

An analysis was performed for sources of flakiness in this test.  A
number of implementation problems have been resolved in recent changes,
with this change focusing on addressing problems with the test code
itself:

1. Do not require the color to match the tone at the same time.  A/V
   sync belongs in a perf test.
2. Do not require a color or tone to be held for a full half-second.
   Instead, any momentary observation that matches expectations is
   sufficient.
3. Only examine the content part of each video frame (i.e., exclude the
   surrounding black borders caused by letterboxing).
4. Generate the source content using a canvas and
   requestAnimationFrame().  The source content has a little "ball"
   animating within it to ensure changing video frames are being sent
   through end-to-end.

Also added AES keys to test end-to-end with encryption enabled.

This change re-enables the test for Release builds only.  Once the test
has proven to be stable again, it will be re-enabled for Debug builds as
well.  Baby steps...

BUG=396413

Review URL: https://codereview.chromium.org/515473004

Cr-Commit-Position: refs/heads/master@{#292539}

Don't show "Access Address Book" prompt if Autofill is disabled, the form has
the attribute autocomplete="off", or if Chrome doesn't think the form is
Autofillable.

BUG=398476

Review URL: https://codereview.chromium.org/512933004

Cr-Commit-Position: refs/heads/master@{#292538}

Adjust underline positions so the lines won't get clipped. I
think that the inaccuracy may be occurring due to our
conversion from PangoFontDescriptions (usually containing a
point-based size) to a rounded pixel-based size and then
back to a PangoFontDescription. Luckily, RenderTextPango
will be replaced by RenderTextHarfBuzz soon.

Also delete some dead code in PlatformFontPango for
computing underline metrics.

BUG=393117
TEST=manual: checked that the "Import bookmarks now..."
     underline is visible after setting my UI font to
     "Ubuntu 11"

Review URL: https://codereview.chromium.org/509093004

Cr-Commit-Position: refs/heads/master@{#292537}

BUG=408251
TBR=piman@chromium.org

Review URL: https://codereview.chromium.org/519523002

Cr-Commit-Position: refs/heads/master@{#292536}

In [[Install]] algo. of the latest spec, register() resolves a promise
with a registration object after the worker enters the installation
process. This means the registration needs to set its .installing field
before the register promise is resolved.

But, before this patch, the registration cannot do so before resolving
the register promise because its proxy is set when the registration is
passed to the blink.

This change makes the registration queue notifications of worker version
and 'updatefound'. Once the proxy becomes ready, the registration handles
them.

Spec: https://slightlyoff.github.io/ServiceWorker/spec/service_worker/index.html#installation-algorithm

BUG=399533,406240
TEST=content_unittests --gtest_filter=ServiceWorker*
TEST=run_webkit_tests.py http/tests/serviceworker

Review URL: https://codereview.chromium.org/516873002

Cr-Commit-Position: refs/heads/master@{#292535}

Any thoughts?

BUG=

Review URL: https://codereview.chromium.org/510203002

Cr-Commit-Position: refs/heads/master@{#292534}

BUG=110610

Review URL: https://codereview.chromium.org/511183002

Cr-Commit-Position: refs/heads/master@{#292533}

Revert of Fix the Copresence permission. (patchset #1 of https://codereview.chromium.org/523433002/)

Reason for revert:
This change causes a test failure:

http://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%20Tests%20%281%29/builds/48008/steps/unit_tests/logs/stdio

Error:

[ RUN      ] PermissionsTest.PermissionMessages
../../chrome/common/extensions/permissions/permission_set_unittest.cc:781: Failure
Value of: permission_info->message_id()
Actual: 65
Expected: PermissionMessage::kNone
Which is: 1
unexpected message_id for copresence

Looks like this change is the cause.

Original issue's description:
> Fix the Copresence permission.
>
> The initializer for the Copresence permission is malformed. It specifies the
> message id instead of the flag.
>
> R=kalman@chromium.org
> BUG=408843
>
> Committed to pending queue: https://chromium.googlesource.com/chromium/src/+/c2a41e8

TBR=kalman@chromium.org,rkc@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=408843

Review URL: https://codereview.chromium.org/520493002

Cr-Commit-Position: refs/heads/master@{#292532}

This value was added to the code in commit position 292033.

R=isherman@chromium.org
CC=robertshield@chromium.org
BUG=394352
NOTRY=true

Review URL: https://codereview.chromium.org/515313004

Cr-Commit-Position: refs/heads/master@{#292531}

This CL fixes two issues.
. It brings up a scroll bar if the text in the feedback window doesn't fit in
  the space available on screen (this happens when the user sets the page zoom
  too 120% or above)
. Sets the focus to the description field when the feedback app starts.

R=xiyuan@chromium.org
BUG=295999,330139

Review URL: https://codereview.chromium.org/511393002

Cr-Commit-Position: refs/heads/master@{#292530}

If we're running a test image or have the --disable-demo-mode flag specified
we should not be launching demo mode. The channel check doesn't exist for the
HID screen. This CL moves the check into the main demo mode detector so if it
is called from anywhere, the check is always performed.

R=xiyuan@chromium.org
BUG=398271

Review URL: https://codereview.chromium.org/514273003

Cr-Commit-Position: refs/heads/master@{#292529}

Test is flaky. See bug for details.

NOTRY=true
TBR=nkostylev@chromium.org
BUG=408867

Review URL: https://codereview.chromium.org/522453004

Cr-Commit-Position: refs/heads/master@{#292528}

Nest a QUIC SerializedPacket inside a QUIC QueuedPacket.

Merge internal change: 74239145
https://codereview.chromium.org/509203003/

Remove PacketType from QUIC because the QUEUED type is not used and the
other types can be reduced to a bool.

Merge internal change: 74148481
https://codereview.chromium.org/515003003/

Don't send a QUIC SCUP message until after handshake confirmed.

A server config update was occasionally being sent before the crypto
handshake was complete, causing the client to close the connection.

Merge internal change: 74132773
https://codereview.chromium.org/516713002/

Change TransmissionInfo's all_transmissions SequenceNumberSet* to
default to NULL, and only be present if there are multiple transmissions.

Estimated to save ~2% of CPU.

Merge internal change: 74076012
https://codereview.chromium.org/509073004/

Log the quic version for internal server tracing.
Not used in production.

Merge internal change: 74069715
https://codereview.chromium.org/514043002/

Optimize QuicUnackedPacketMap by changing from a LinkedHashMap to a
deque.

Estimated to save ~3% of CPU based on pprof profiling of 100 large gets
on tools/quic/end_to_end_test.cc.

Merge internal change: 74054196
https://codereview.chromium.org/514033002/

Unit test for empty packet closes QUIC connection bug.

Merge internal change: 74041239
https://codereview.chromium.org/495423011/

Log the QUIC transmission type in QUIC internal server trace visitor.
Not used in production.

Merge internal change: 73895739
https://codereview.chromium.org/514023002/

Not used in production. Log whether a QUIC packet sent was a crypto
packet or not for internal server side tracing.

Merge internal change: 73894063
https://codereview.chromium.org/515003002/

Fix a QUIC bug in which PING frames were not being ACK'd.

Merge internal change: 73837644
https://codereview.chromium.org/512933005/

R=rch@chromium.org

Review URL: https://codereview.chromium.org/515303003

Cr-Commit-Position: refs/heads/master@{#292527}

https://skia.googlesource.com/skia/+log/60cc7d353b2adc8fcc1712e1f7a965e70121ebcc..088f4b8eb31b2af26e66dcbaae062f40b3955059

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=djsollen@google.com

Review URL: https://codereview.chromium.org/516973003

Cr-Commit-Position: refs/heads/master@{#292526}

BUG=407994
TEST=as in bug

Review URL: https://codereview.chromium.org/518443003

Cr-Commit-Position: refs/heads/master@{#292525}

unittest produces a suite._ErrorHolder object in these cases. We can parse its description to figure out what class/module it is and what tests we have in that class/module.

BUG=405766
TEST=Make a TestCase fail setUpClass, run it with --retry-limit=3 and check for retries.
R=dpranke, tonyg

Review URL: https://codereview.chromium.org/515263003

Cr-Commit-Position: refs/heads/master@{#292524}