GitLab

AKA "lazy logging".

This slightly improves readability because it decreases
the number of parentheses. It also saves a tiny amount of time
constructing the interpolated strings for debug statements
when the logging level would make it so that debug logs
are not printed.

This fixes the logging-not-lazy warnings which were
just enabled in http://crrev.com/2120083002.

This CL also changes the few instances of _log.warn to
_log.warning for consistency.

Note, most of the changes were done with sed:
s/\.\(debug\|warning\|warn\|info\|error\|critical\)(\(.*\) % (\(.*\)))/\.\1(\2, \3)/
s/\.\(debug\|warning\|warn\|info\|error\|critical\)(\(.*\) % \(.*\))/\.\1(\2, \3)/

BUG=598897

Review-Url: https://codereview.chromium.org/2188623002
Cr-Commit-Position: refs/heads/master@{#408745}

This change adds a PageLoadMetricsProvider, which implements the new
OnAppEnterBackground hook, and informs all MetricsWebContentsObservers
and in turn all PageLoadTrackers when the app enters the background.

On Android (and iOS), when an app enters the background, it may be
killed without further notifications. Thus, having this callback
will allow us to persist metrics before the app may be killed.
Our page load metrics subsystem isn't used on iOS, so we only target
Android in this change.

Unfortunately, we can't use the existing WebContents hidden callback,
as this gets invoked after the UMA subsystem has already flushed
metrics to disk. Thus, we need to add this explicit callback which
allows us to persist metrics as part of the UMA subsystem handling
the backgrounding callback.

For now, we log a debugging histogram in PageLoadTracker to help
understand how often trackers enter the background, and how often
those trackers are destroyed without running their destructors. In a
subsequent change, we'll add support for informing PLMObservers that
the app entered the background. This will allow any PLMObsevers that
implement OnComplete or OnFailedProvisionalLoad to also get called back
in cases where the tracker may be destroyed before its destructor runs,
due to the application being killed.

BUG=608360

Review-Url: https://codereview.chromium.org/2189543002
Cr-Commit-Position: refs/heads/master@{#408744}

BUG=631162
TBR=pfeldman

Review-Url: https://codereview.chromium.org/2194683003
Cr-Commit-Position: refs/heads/master@{#408743}

We need to clear ScopedPersistents so that V8 doesn't call phantom callbacks
after Oilpan starts lazy sweeping.

BUG=626893

Review-Url: https://codereview.chromium.org/2196543003
Cr-Commit-Position: refs/heads/master@{#408742}

BUG=590369
R=yosin,esprehn

Review-Url: https://codereview.chromium.org/2142903003
Cr-Commit-Position: refs/heads/master@{#408741}

two that seem unused as iOS only.

BUG=none

Review-Url: https://codereview.chromium.org/2197713002
Cr-Commit-Position: refs/heads/master@{#408740}

BUG=none

Review-Url: https://codereview.chromium.org/2071463002
Cr-Commit-Position: refs/heads/master@{#408739}

BUG=632379

Review-Url: https://codereview.chromium.org/2195563002
Cr-Commit-Position: refs/heads/master@{#408738}

In particular, the code was being run on non-data URLs, which caused it
to do weird and exciting things, and it had an incorrect DCHECK.

BUG=632605

Review-Url: https://codereview.chromium.org/2191323005
Cr-Commit-Position: refs/heads/master@{#408737}

This mostly cleans up some routing code in Quick Unlock. It also has a few small fixes in People.

BUG=608115
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2154213008
Cr-Commit-Position: refs/heads/master@{#408736}

The similar CL for IPC https://codereview.chromium.org/2148303005/

BUG=582434

Review-Url: https://codereview.chromium.org/2189543004
Cr-Commit-Position: refs/heads/master@{#408735}

BUG=616536

Review-Url: https://codereview.chromium.org/2129253002
Cr-Commit-Position: refs/heads/master@{#408734}

Revert of Linux: Support the --class argument (patchset #4 id:60001 of https://codereview.chromium.org/2186813002/ )

Reason for revert:
Change breaks compile on Linux x64: https://build.chromium.org/p/chromium.chrome/builders/Google%20Chrome%20Linux%20x64/builds/12770

Original issue's description:
> Linux: Support the --class argument
>
> It appears we lost support for --class when we stopped using gtk for our
> windowing.  This CL adds that feature back.
>
> BUG=118613
>
> Committed: https://crrev.com/b4a38edc75149162fcbdf01e42d5411cf8e72520
> Cr-Commit-Position: refs/heads/master@{#408709}

TBR=thestig@chromium.org,benwells@chromium.org,thomasanderson@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=118613

Review-Url: https://codereview.chromium.org/2193393002
Cr-Commit-Position: refs/heads/master@{#408733}

The existing doc (https://sites.google.com/a/chromium.org/dev/developers/design-documents/idl-compiler)
looks old, but it's probably better to keep it next to the code.

R=haraken,yukishiino
BUG=

Review-Url: https://codereview.chromium.org/2195743003
Cr-Commit-Position: refs/heads/master@{#408732}

TBR=sullivan@chromium.org
BUG=630854
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.perf:mac_retina_perf_cq;

Review-Url: https://codereview.chromium.org/2189383002
Cr-Commit-Position: refs/heads/master@{#408731}

Add expired as a logged event in offline-internal page.

BUG=NONE

Review-Url: https://codereview.chromium.org/2191183002
Cr-Commit-Position: refs/heads/master@{#408730}

First of three patches to reject promises if the frame is not connected
or the frame disconnected while the promise was not fulfilled.

Implements:

1. The ActiveAlgorithm set[1] as m_activeAlgorithms in BluetoothRemoteGATTServer
2. gattServer-connection-checking wrapper[2] as GetPrimaryServicesCallback in
   BluetoothRemoteGattServer.cpp
3. Connection check of query bluetooth cache[3] in getPrimaryServicesImpl.

[1] https://webbluetoothcg.github.io/web-bluetooth/#dom-bluetoothremotegattserver-activealgorithms-slot
[2] https://webbluetoothcg.github.io/web-bluetooth/#connection-checking-wrapper
[3] https://webbluetoothcg.github.io/web-bluetooth/#query-the-bluetooth-cache

BUG=608538

Review-Url: https://codereview.chromium.org/2183903002
Cr-Commit-Position: refs/heads/master@{#408729}

This patch updates the code to prevent auto deducing to a raw pointer.

R=mattm@chromium.org
BUG=554600

Review-Url: https://codereview.chromium.org/2194613002
Cr-Commit-Position: refs/heads/master@{#408728}

This patch updates the code to prevent auto deducing to a raw pointer.

R=avi@chromium.org
BUG=554600

Review-Url: https://codereview.chromium.org/2195483004
Cr-Commit-Position: refs/heads/master@{#408727}

A helper function that determines whether an IP address is loopback was
checking for a "::1" substring in the IP address. Non-loopback IPv6
addresses can have that substring - e.g. "1234::1234:1234". If the test
runs on a machine with one of these addresses, it would fail
erroneously.

The fix simply adds spaces on both sides of the pattern, so that it only
matches complete IP addresses (always surrounded by spaces in an ICE
candidate line).

BUG=629705

Review-Url: https://codereview.chromium.org/2185533007
Cr-Commit-Position: refs/heads/master@{#408726}

BUG=628831

Review-Url: https://codereview.chromium.org/2186643002
Cr-Commit-Position: refs/heads/master@{#408725}

This patch updates the code to prevent auto deducing to a raw pointer.

R=creis@chromium.org
BUG=554600

Review-Url: https://codereview.chromium.org/2188393002
Cr-Commit-Position: refs/heads/master@{#408724}

In ComputeTransformToTarget and ComputeTransformFromTarget. This is
turned on only when verify_transform_tree_calculations is true (only
cc_unittests as of now) and when target is an ancestor (as that's
the only case dynamically calculated target space transforms support
right now)
This CL also sets verify_transform_tree_calculations and
verify_clip_tree_calculations in some tests.

BUG=622372
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2191983002
Cr-Commit-Position: refs/heads/master@{#408723}

BUG=632468

Review-Url: https://codereview.chromium.org/2193293003
Cr-Commit-Position: refs/heads/master@{#408722}

This patch updates the code to prevent auto deducing to a raw pointer.

R=anthonyvd@chromium.org
BUG=554600

Review-Url: https://codereview.chromium.org/2192033002
Cr-Commit-Position: refs/heads/master@{#408721}

This patch updates the code to prevent auto deducing to a raw pointer.

R=mattm@chromium.org
BUG=554600

Review-Url: https://codereview.chromium.org/2194653002
Cr-Commit-Position: refs/heads/master@{#408720}

shows up in chrome://gpu

makes use of build/util/lastchange.py

BUG=235455

Review-Url: https://codereview.chromium.org/2187533002
Cr-Commit-Position: refs/heads/master@{#408719}

This patch updates the code to prevent auto deducing to a raw pointer.

R=dsinclair, jochen
BUG=554600

Review-Url: https://codereview.chromium.org/2184913006
Cr-Commit-Position: refs/heads/master@{#408718}

When running "./chrome --mash" most child processes no longer have the
mojo application name as part of the command string. This makes it
difficult to figure out what part of mus+ash LOG statements or CHECK
failures come from. Add a DVLOG(0) that prints out the name and pid when
a child process is successfully launched.

Using verbosity level 0 since it is important information for debugging
and should occur rarely.

BUG=none

Review-Url: https://codereview.chromium.org/2194943002
Cr-Commit-Position: refs/heads/master@{#408717}

R=tommycli@chromium.org
BUG=629604
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2178303003
Cr-Commit-Position: refs/heads/master@{#408716}

This patch updates the code to prevent auto deducing to a raw pointer.

R=sergeyu@chromium.org
BUG=554600

Review-Url: https://codereview.chromium.org/2194673002
Cr-Commit-Position: refs/heads/master@{#408715}

This reverts commit 7d314cf3 as a follow up to the decision about delayed capturing
https://github.com/w3c/pointerevents/issues/32

This also gets rid of the need to make PointerEventManager a GC object.

BUG=627192, 629935

Review-Url: https://codereview.chromium.org/2193473002
Cr-Commit-Position: refs/heads/master@{#408714}

This patch updates the code to prevent auto deducing to a raw pointer.

R=dewittj@chromium.org
BUG=554600

Review-Url: https://codereview.chromium.org/2193653002
Cr-Commit-Position: refs/heads/master@{#408713}

This patch updates the code to prevent auto deducing to a raw pointer.

R=alekseys@chromium.org
BUG=554600

Review-Url: https://codereview.chromium.org/2194643003
Cr-Commit-Position: refs/heads/master@{#408712}

RequestFinishedListener originally depended on NetworkQualityEstimator's
executor. These are separate features which should not depend on each
other, and as of recently the NQE listeners also provide their own
executors.

Also move RequestFinishedListener out of CronetEngine into its own file,
and move its tests into their own file.

BUG=618034

Review-Url: https://codereview.chromium.org/2178053002
Cr-Commit-Position: refs/heads/master@{#408711}

Revert of services/ui: Revert CLs that broke Windows. (patchset #2 id:20001 of https://codereview.chromium.org/2194893002/ )

Reason for revert:
crrev.com/408695 fixes the issue. So we can reland
these CLs.

Original issue's description:
> services/ui: Revert CLs that broke Windows.
>
> The new GPU code path does not seem to work correctly on Windows yet
> (everything is black). So restore the deprecated path.
>
> Revert "services/ui: Remove deprecated gpu code" crrev.com/408207
> Revert "services/ui: Remove unused connector argument" crrev.com/408230
>
> BUG=623157, 630893
> TBR=ben@, piman@, penghuang@
>
> Committed: https://crrev.com/f99035fbb67293836944993268f0357ad3d65a4d
> Cr-Commit-Position: refs/heads/master@{#408648}

TBR=ben@chromium.org,piman@chromium.org,penghuang@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=623157, 630893

Review-Url: https://codereview.chromium.org/2196503003
Cr-Commit-Position: refs/heads/master@{#408710}

It appears we lost support for --class when we stopped using gtk for our
windowing.  This CL adds that feature back.

BUG=118613

Review-Url: https://codereview.chromium.org/2186813002
Cr-Commit-Position: refs/heads/master@{#408709}

They inherit from XCTestCase, so TestCase suffix makes
it a bit more clear.

BUG=

Review-Url: https://codereview.chromium.org/2195773002
Cr-Commit-Position: refs/heads/master@{#408708}

This has been done in the C++ layer already, and I'd like to do this on the Java side as well since a single object doesn't always want to hold both the registry and the provider objects.

R=sievers@chromium.org
BUG=
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2191033002
Cr-Commit-Position: refs/heads/master@{#408707}

BUG=631162

Review-Url: https://codereview.chromium.org/2197723002
Cr-Commit-Position: refs/heads/master@{#408706}