GitLab

If there is an untriggered idle task when the document is gone, it
introduces memory leak because IdleRequestCallbackWrapper keeps a
reference to the ScriptedIdleTaskController, making it live longer
than the document.

This patch fixes the issue by letting IdleRequestCallbackWrapper
store a WeakPersistent instead.

BUG=595155

Review-Url: https://codereview.chromium.org/2618893002
Cr-Commit-Position: refs/heads/master@{#441929}

The Reading List Collection View does not need to be reloaded for every change.
This CL reloads it only if some changes occurred.

BUG=677924

Review-Url: https://codereview.chromium.org/2609173002
Cr-Commit-Position: refs/heads/master@{#441928}

Sync Encryption screen now listens to sync changes and reloads itself
when necessary. This ensures modifications to the passphrase (done by
the Sync Encryption Passphrase screen) are reflected in the UI.

BUG=678609

Review-Url: https://codereview.chromium.org/2610403003
Cr-Commit-Position: refs/heads/master@{#441927}

Reload will delete Pending Item so altering it and reload will reload
the lastCommittedItem.
If there is a pending item, stop loading and trigger a new navigation.

BUG=678293

Review-Url: https://codereview.chromium.org/2610953002
Cr-Commit-Position: refs/heads/master@{#441926}

Browser actions will be reordered in a future change.

BUG=648558, 648563, 673362

Review-Url: https://codereview.chromium.org/2607533004
Cr-Commit-Position: refs/heads/master@{#441925}

It's not needed. ContextClient is used to detect if a frame is detached,
but we can check it with scriptState->contextIsValid().

BUG=610176

Review-Url: https://codereview.chromium.org/2614943003
Cr-Commit-Position: refs/heads/master@{#441924}

This class is only used for the RecentTabsPage.

BUG=659631

Review-Url: https://codereview.chromium.org/2612243002
Cr-Commit-Position: refs/heads/master@{#441923}

We're deprecating the default constructor of Supplement.

BUG=610176

Review-Url: https://codereview.chromium.org/2619543002
Cr-Commit-Position: refs/heads/master@{#441922}

Moves the test code out of RangeCheck. Reorders range comparisons and
removes bitfields to avoid a GCC constexpr bug. Removes some custom
optimizations that are no longer necessary after the restructuring.

TBR=tsepez

Review-Url: https://codereview.chromium.org/2614073002
Cr-Commit-Position: refs/heads/master@{#441921}

We're deprecating the default constructor of Supplement.

BUG=610176

Review-Url: https://codereview.chromium.org/2615713006
Cr-Commit-Position: refs/heads/master@{#441920}

There are crash reports that |m_resource| is broken in
ResourceLoader::didReceiveResponse. It is strange as ResourceLoader::m_resource
is constant after its construction. To investigate the issue, this CL adds some
CHECKs in ResourceLoader.

BUG=675947
R=tyoshino@chromium.org

Review-Url: https://codereview.chromium.org/2619513002
Cr-Commit-Position: refs/heads/master@{#441919}

If we pass in ScriptState to DOMWindowSpeechSynthesis::speechSynthesis,
DOMWindowSpeechSynthesis doesn't need to observe ContextClient.

BUG=610176

Review-Url: https://codereview.chromium.org/2618613006
Cr-Commit-Position: refs/heads/master@{#441918}

When dismissing a suggestion through the context menu, it is possible
that by the time the animation ends, we navigated away from the NTP
and it got destroyed, resulting in an NPE. We now exit early when
that is the case.

This CL also reverts the ineffective tentative fix from
https://codereview.chromium.org/2553163002

BUG=668945

Review-Url: https://codereview.chromium.org/2614753002
Cr-Commit-Position: refs/heads/master@{#441917}

BUG=673904
R=sdefresne@chromium.org

Review-Url: https://codereview.chromium.org/2610423002
Cr-Commit-Position: refs/heads/master@{#441916}

BUG=677908

Review-Url: https://codereview.chromium.org/2609403006
Cr-Commit-Position: refs/heads/master@{#441915}

This allows the compiler to compute the length of literal strings that
are compared and then we can early abort comparing if the length is
different.

This does introduce a strlen call for non-literal strings, which we
have very few of today.

This patch fixes the web idl enum checking loop to use WTF::equal to
avoid the strlen since that seemed like an easy way to remove lots of strlen
calls. We can make other callers use a similar approach in the future if
if we see strlen on profiles.

By code auditing (with code search method finding) I noticed a few other
places will now call strlen (ex. iframe sandbox and permissions attribute
parsing) but this patch opts not to convert all places we now call strlen and
instead go for the simple path of fixing those if we run into problems. This
is a similar trade off to code elsewhere in Chromium that uses
base::StringPiece which also puts a strlen call for non-literal strings.

I also took this as an opportunity to remove the operator== methods which aren't
used for things like LChar* (strangely UChar* was missing) and Vector<char>.

BUG=678266

Review-Url: https://codereview.chromium.org/2614123002
Cr-Commit-Position: refs/heads/master@{#441914}

BUG=678393

Review-Url: https://codereview.chromium.org/2611973003
Cr-Commit-Position: refs/heads/master@{#441913}

The Objective-C frameworks must be imported and not included.
Command to find Objective-C frameworks
dir="`xcode-select -p`/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS.sdk/System/Library/Frameworks";
for f in `ls $dir`; do
  name=${f%%.framework};
  grep "^\s*#import " $dir/$f/Headers/$name.h > /dev/null 2>/dev/null;
  if [ "$?" = "0" ]; then
    echo $name;
  fi;
done

BUG=677965

Review-Url: https://codereview.chromium.org/2606753004
Cr-Commit-Position: refs/heads/master@{#441912}

We're deprecating a default constructor of Supplement<>.
This CL moves Supplement<Document> to use a constructor that takes a Document parameter.

This enables to remove a couple of unnecessary m_document's.

BUG=610176
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2617103002
Cr-Commit-Position: refs/heads/master@{#441911}

BUG=none

Review-Url: https://codereview.chromium.org/2610923010
Cr-Commit-Position: refs/heads/master@{#441910}

Rebase protobuf to current HEAD to pull the 3 following changes:
* 4c5d3ed Fix integer overflow in FastUInt32ToBufferLeft
* 4587a3f [arm/gcc] Don't rely on KUSER_HELPERS feature for atomics
* a5a2c1d generic atomicops: Use strong compare_exchange

This change has been made following steps described in
third_party/protobuf/README.chromium file, please refer to it
for further information.

This is a reland of https://codereview.chromium.org/2599263002/

BUG=599051,673488

Review-Url: https://codereview.chromium.org/2598373002
Cr-Commit-Position: refs/heads/master@{#441909}

WindowProxy::initializeIfNeeded() is no-op when isContextInitialized() returns true.
So we can remove the check from ScriptController::windowProxy() and unconditionally
call WindowProxy::initializeIfNeeded().

This CL will cause a subtle change: Before this CL, dispatchDidClearWindowObjectInMainWorld
was not called if WindowProxy gets initialized via WindowProxy::setGlobal. After this CL,
dispatchDidClearWindowObjectInMainWorld will be called always when WindowProxy gets initialized.
I don't think the difference matters (and I guess the new behavior will be more correct).

BUG=677253

Review-Url: https://codereview.chromium.org/2613893002
Cr-Commit-Position: refs/heads/master@{#441908}

NoExceptionStateAssertionChecker is equivalent to NonThrowableExceptionState.
This CL replaces the former with the latter.

BUG=
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2616023004
Cr-Commit-Position: refs/heads/master@{#441907}

Now it's unused. We can remove it.

BUG=677253

Review-Url: https://codereview.chromium.org/2619523003
Cr-Commit-Position: refs/heads/master@{#441906}

ArcContentFileSystemService is renamed to ArcFileSystemService, and it
will register two ARC file systems.

Also constants in arc_content_file_system_url_util.h are renamed
now that we have two ARC file systems.

BUG=chromium:671511
TEST=unit_tests --gtest_filter='Arc*'
TEST=trybot

Review-Url: https://codereview.chromium.org/2589643002
Cr-Commit-Position: refs/heads/master@{#441905}

Returning nullptr caused crash in fuzz testing when the call sites
attempts to create a string object out of it. Previously it was
returning an empty string(std::string()). This change makes it work
as it used to by changing the return type back to std::string.

BUG=678359

Review-Url: https://codereview.chromium.org/2619493002
Cr-Commit-Position: refs/heads/master@{#441904}

BUG=661105
TEST=Manual. Check that focus can move on a close button of an ARC
     notification by TAB.

Review-Url: https://codereview.chromium.org/2610853011
Cr-Commit-Position: refs/heads/master@{#441903}

Now that DOMWindowPerformance can access DOMWindow with host(),
DOMWindowPerformance doesn't need to observe ContextClient.

This CL removes DOMWindowPerformance::m_window.

BUG=610176

Review-Url: https://codereview.chromium.org/2615633011
Cr-Commit-Position: refs/heads/master@{#441902}

Now that DOMWindowStorage can access DOMWindow with host(),
DOMWindowStorage doesn't need to observe ContextClient.

BUG=610176

Review-Url: https://codereview.chromium.org/2617703003
Cr-Commit-Position: refs/heads/master@{#441901}

It's not needed.

BUG=610176

Review-Url: https://codereview.chromium.org/2611163003
Cr-Commit-Position: refs/heads/master@{#441900}

The StyleEngine for html import documents does not have a Frame which
means size media queries will always evaluate to true. We incorrectly
replaced a passed-in master document StyleEngine with the StyleEngine
from the DocumentStyleSheetCollection in https://crrev.com/90d4ea3d
That was wrong for evaluating media queries.

R=meade@chromium.org
BUG=677963

Review-Url: https://codereview.chromium.org/2618803002
Cr-Commit-Position: refs/heads/master@{#441899}

Now that Navigator's supplement can get a Navigator object via host().
Then Navigator's supplement doesn't need to observe ContextClient.

BUG=610176

Review-Url: https://codereview.chromium.org/2617583005
Cr-Commit-Position: refs/heads/master@{#441898}

It's not needed.

BUG=610176

Review-Url: https://codereview.chromium.org/2615163002
Cr-Commit-Position: refs/heads/master@{#441897}

ScriptController::updateDocument is unnecessarily complicated.

Assuming that ScriptController::updateDocument is never called after
the context is detached (I added a dcheck to verify the fact in WindowProxy::updateDocument()),
we can simplify ScriptController::updateDocument.

It also enables us to remove ScriptController::initializeMainWorld.

BUG=677253

Review-Url: https://codereview.chromium.org/2608163009
Cr-Commit-Position: refs/heads/master@{#441896}

This is a followup CL for [1] and does following things:

- documents lifetime of SharedWorkerHost.
- removes SharedWorkerHost::WorkerContextDestroyed that clears member fields.
  This is not meaningful because the host is destructed immediately after this
  function.
- removes |instance_| checks because it's always valid.

[1] https://codereview.chromium.org/2601893002/#msg39

BUG=612308

Review-Url: https://codereview.chromium.org/2614013004
Cr-Commit-Position: refs/heads/master@{#441895}

This allows CSSProperties.in to have unordered entries.

Generated code is now:

namespace blink {

bool CSSPropertyMetadata::isDescriptorOnly(CSSPropertyID property) {
  switch(property) {
    case CSSPropertyInvalid:
      ASSERT_NOT_REACHED();
      return false;
    case CSSPropertyFontDisplay:
    case CSSPropertyMinZoom:
    case CSSPropertyOrientation:
    case CSSPropertyUnicodeRange:
    case CSSPropertyMaxZoom:
    case CSSPropertyUserZoom:
    case CSSPropertySrc:
      return true;
    default:
      return false;
  }
}

BUG=677884

Review-Url: https://codereview.chromium.org/2614773007
Cr-Commit-Position: refs/heads/master@{#441894}

Now that Navigator's supplement can get a Navigator object via host().
Then Navigator's supplement doesn't need to observe ContextClient.

BUG=610176

Review-Url: https://codereview.chromium.org/2618773004
Cr-Commit-Position: refs/heads/master@{#441893}

It's unused.

BUG=610176

Review-Url: https://codereview.chromium.org/2616013002
Cr-Commit-Position: refs/heads/master@{#441892}

BUG=chromium:671511
TEST=unit_tests --gtest_filter='Arc*'
TEST=trybot

Review-Url: https://codereview.chromium.org/2580713004
Cr-Commit-Position: refs/heads/master@{#441891}

Now that Navigator's suppelemnt can access the Navigator object via host(),
it doesn't need to observe ContextClient.

BUG=610176

Review-Url: https://codereview.chromium.org/2615153002
Cr-Commit-Position: refs/heads/master@{#441890}