GitLab

Revert of DatabaseBackend objects should be destructed in the originator thread (https://codereview.chromium.org/183883005/)

Reason for revert:
Unnecessary change


Original issue's description:
> DatabaseBackend objects should be destructed in the originator thread.
> 
> Before this CL, it was possible that DatabaseThread held the last
> references of DatabaseBackend objects and it was released in the
> database thread.
> We must destruct objects in the originator thread in Oilpan.
> 
> Implementation:
> DatabaseThread::recordDatabaseClosed doesn't remove a RefPtr<
> DatabaseBackend>, and we release RefPtr<DatabaseBackend> objects
> in ~DatabaseThread, which is executed in the main/worker thread.
> Note that there are no ways to close a database explicitly, and
> DatabaseBackend objects are never destructed before their
> DatabaseContext is stopped. DatabaseThread::recordDatabaseClosed
> is actually unnecessary.
> 
> DatabaseThread doesn't know DatabaseBackend open status any longer.
> We remove DatabaseThread::isDatabaseOpen, and its existing
> callsites use DatabaseBackend::open() instead.
> 
> 
> BUG=347902
> TEST=none; no visible behavior changes hopefully.
> R=michaeln@chromium.org
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=168362

TBR=michaeln@chromium.org,oilpan-reviews@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=347902

Review URL: https://codereview.chromium.org/178103007

git-svn-id: svn://svn.chromium.org/blink/trunk@168431 bbb929c8-8fbe-4397-9dbb-9b2b20218538

All text selection colour again.

TBR=dpranke@chromium.org
BUG=348800

Review URL: https://codereview.chromium.org/187023006

git-svn-id: svn://svn.chromium.org/blink/trunk@168430 bbb929c8-8fbe-4397-9dbb-9b2b20218538

All text selection colour.

TBR=pdr@chromium.org
BUG=348800

Review URL: https://codereview.chromium.org/178103006

git-svn-id: svn://svn.chromium.org/blink/trunk@168429 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=168335

BUG=114658
TBR=cbiesinger@chromium.org

Review URL: https://codereview.chromium.org/183013012

git-svn-id: svn://svn.chromium.org/blink/trunk@168428 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=168335

BUG=114658
TBR=cbiesinger@chromium.org

Review URL: https://codereview.chromium.org/187393003

git-svn-id: svn://svn.chromium.org/blink/trunk@168426 bbb929c8-8fbe-4397-9dbb-9b2b20218538

More selection colour.

TBR=dpranke@chromium.org
BUG=348800

Review URL: https://codereview.chromium.org/181063028

git-svn-id: svn://svn.chromium.org/blink/trunk@168425 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=dpranke@chromium.org
BUG=348800

Review URL: https://codereview.chromium.org/177473005

git-svn-id: svn://svn.chromium.org/blink/trunk@168424 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Avoid calling hasAttribute() and then getAttribute() for performance.
This causes the attribute to be looked up (linear search) twice instead
of once if we call getAttribute() directly and then check if the returned
AtomicString is null.

R=eseidel@chromium.org, eseidel, rwlbuis

Review URL: https://codereview.chromium.org/183763033

git-svn-id: svn://svn.chromium.org/blink/trunk@168423 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Have DOMWindow deal with references instead of pointers when possible. This
makes the code look safer and gets rid of unnecessary null checks.

This CL also updates the generated bindings so that they allow for the
implementation to return references instead of pointers in more cases.

R=adamk@chromium.org, nbarth@chromium.org, adamk

Review URL: https://codereview.chromium.org/176763009

git-svn-id: svn://svn.chromium.org/blink/trunk@168422 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Move Document code out of Node.cpp and into Document.cpp for clarity. There is
no reason (e.g. inlining) requiring that this code be in Node.cpp.

R=adamk@chromium.org, adamk, eseidel

Review URL: https://codereview.chromium.org/178543006

git-svn-id: svn://svn.chromium.org/blink/trunk@168421 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Have positionInParentBeforeNode() / positionInParentAfterNode() take a
reference as the implementation expected the pointer to be non-null. This makes
the code look safer.

R=adamk@chromium.org, adamk

Review URL: https://codereview.chromium.org/182383012

git-svn-id: svn://svn.chromium.org/blink/trunk@168420 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Store IdTargetObserver::m_registry as a reference instead of a pointer.
This makes the code look safer and gets rid of an unnecessary null check.

Also make IdTargetObserverRegistry non copyable to protect against bad
assignments to references.

R=adamk@chromium.org, adamk, rwlbuis

Review URL: https://codereview.chromium.org/183463004

git-svn-id: svn://svn.chromium.org/blink/trunk@168419 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Have NodeWithIndex deal with references instead of pointers. This makes the
code look safer.

R=adamk@chromium.org, adamk, rwlbuis

Review URL: https://codereview.chromium.org/176863009

git-svn-id: svn://svn.chromium.org/blink/trunk@168418 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Pass rootNode by reference when constructing node lists / collections to make
it clear that the rootNode cannot be null. Their implementation already relies
on the fact that the rootNode cannot be null.

R=adamk@chromium.org, adamk

Review URL: https://codereview.chromium.org/176933013

git-svn-id: svn://svn.chromium.org/blink/trunk@168417 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Have ElementRareData::ensure*() methods return references instead of pointers
for consistency and clarity.

R=adamk@chromium.org, adamk

Review URL: https://codereview.chromium.org/183763027

git-svn-id: svn://svn.chromium.org/blink/trunk@168416 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Have MutationObserverRegistration deal with references instead of pointers when
possible. This makes the code look safer.

R=adamk@chromium.org, adamk

Review URL: https://codereview.chromium.org/178473026

git-svn-id: svn://svn.chromium.org/blink/trunk@168415 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Investigation on bug into why colour of selection changed by one value
from 0x32,0x32,0xfd to 0x32,0x32,0xfe.

R=dpranke@chromium.org
TBR=dpranke@chromium.org
BUG=348800

Review URL: https://codereview.chromium.org/185713014

git-svn-id: svn://svn.chromium.org/blink/trunk@168413 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Prior to this CL, we were setting the style of the RenderView on every style
recalc because we were comparing the RenderView's overflow style with the wrong
value. This CL corrects the bug.

This CL doesn't have much effect yet, but it will become important in a future
CL when we use style modification on RenderLayers to scope compositing updates.

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=168254

Review URL: https://codereview.chromium.org/185263002

git-svn-id: svn://svn.chromium.org/blink/trunk@168412 bbb929c8-8fbe-4397-9dbb-9b2b20218538

See linked bug for investigation on source of minor diff.

R=pdr@chromium.org
TBR=pdr@chromium.org
BUG=348800

Review URL: https://codereview.chromium.org/187143002

git-svn-id: svn://svn.chromium.org/blink/trunk@168411 bbb929c8-8fbe-4397-9dbb-9b2b20218538

On Android, some locale-preferred fonts (e.g. DroidSansFallback.ttf for
Chinese locale) doesn't contain all characters that the default locale
system fonts contain.

For characters that is missing in the font, fallback to the default
locale system font, by calling getLastResortFallbackFont() in
FontCache::platformFallbackForCharacter().

Also modified getLastResortFallbackFont() to prefer Skia default font
instead of the hard-coded default font (sans) and to map also cursive
and fantasy generic families.

Combined similar code in FontCacheSkia.cpp.

BUG=347121
TEST=FontCacheAndroid.platformFallbackForCharacter (will be enabled in another CL)
R=dglazkov@chromium.org

Review URL: https://codereview.chromium.org/185863007

git-svn-id: svn://svn.chromium.org/blink/trunk@168410 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The ShapeInfo hierarchy has grown organically as the spec has changed,
and the naming has become very stale and confusing. The spec is now in
Last Call, so it seems like a good time to take a stab towards better
naming that matches the spec terminology and is also more consistent in
general.

R=leviw@chromium.org

BUG=322165

Review URL: https://codereview.chromium.org/185393015

git-svn-id: svn://svn.chromium.org/blink/trunk@168409 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=343801
R=senorblanco@chromium.org

Review URL: https://codereview.chromium.org/186063003

git-svn-id: svn://svn.chromium.org/blink/trunk@168407 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=348702
R=senorblanco@chromium.org

Review URL: https://codereview.chromium.org/180633010

git-svn-id: svn://svn.chromium.org/blink/trunk@168406 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=348702
R=senorblanco@chromium.org

Review URL: https://codereview.chromium.org/186023003

git-svn-id: svn://svn.chromium.org/blink/trunk@168405 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This changes some FAIL to PASS lines, likely cause was the V8 roll in
https://codereview.chromium.org/184163003, specifically
https://github.com/v8/v8/commit/1e02b1d60095b06ac2f7ef372ca2af6028609119

CC=jkummerow@chromium.org,verwaest@chromium.org
TBR=yangguo@chromium.org
BUG=

Review URL: https://codereview.chromium.org/183583003

git-svn-id: svn://svn.chromium.org/blink/trunk@168404 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=
R=pfeldman@chromium.org

Review URL: https://codereview.chromium.org/184303005

git-svn-id: svn://svn.chromium.org/blink/trunk@168402 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is one noted in the bug as potentially interesting.

It looks like a progression to me as on Windows we go from FAIL->PASS,
and can remove the platform-specific baselines.

R=dpranke@chromium.org
BUG=348800

Review URL: https://codereview.chromium.org/177353008

git-svn-id: svn://svn.chromium.org/blink/trunk@168401 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Confirming that I can infact rebaseline tests.

TBR=dpranke@chromium.org
BUG=348800

Review URL: https://codereview.chromium.org/180633011

git-svn-id: svn://svn.chromium.org/blink/trunk@168400 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/186283003

git-svn-id: svn://svn.chromium.org/blink/trunk@168399 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Also handle zero frame width correctly (happens occasionally per crbug.com/340987), and simplify the LayoutScope interface.

BUG=347384
NOTRY=true

Review URL: https://codereview.chromium.org/180743004

git-svn-id: svn://svn.chromium.org/blink/trunk@168397 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=caseq@chromium.org

Review URL: https://codereview.chromium.org/180273023

git-svn-id: svn://svn.chromium.org/blink/trunk@168396 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The code path resulting in a NULL here currently segfaults, but will return
NULL after the Chromium-side patch https://codereview.chromium.org/185653006/

BUG=347901

Review URL: https://codereview.chromium.org/183833014

git-svn-id: svn://svn.chromium.org/blink/trunk@168395 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Non-renderer-based elements such as optGroup do not have renderers.
BUG=348330
R=esprehn@chromium.org

Review URL: https://codereview.chromium.org/185293011

git-svn-id: svn://svn.chromium.org/blink/trunk@168394 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Since there is always root inline box, InlineBox::root() should return
reference.

BUG=

Review URL: https://codereview.chromium.org/182413005

git-svn-id: svn://svn.chromium.org/blink/trunk@168391 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=304341

Review URL: https://codereview.chromium.org/184043011

git-svn-id: svn://svn.chromium.org/blink/trunk@168390 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=eustas@chromium.org, eustas, pfeldman
BUG=348297

Review URL: https://codereview.chromium.org/183803018

git-svn-id: svn://svn.chromium.org/blink/trunk@168389 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=lushnikov@chromium.org, pfeldman

Review URL: https://codereview.chromium.org/177963004

git-svn-id: svn://svn.chromium.org/blink/trunk@168388 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=
R=pfeldman@chromium.org

Review URL: https://codereview.chromium.org/186553002

git-svn-id: svn://svn.chromium.org/blink/trunk@168387 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The HeapVector was put in a normal Vector of pairs on the stack
which is not traced.

This issue was found with the ASAN build on layout tests.

R=erik.corry@gmail.com, haraken@chromium.org, wibling@chromium.org, zerny@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/180273021

git-svn-id: svn://svn.chromium.org/blink/trunk@168386 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Oilpan: Workaround for crash on quora and geek-and-poke. There is a finalization issue that we need to resolve in CSSCursorImageValue.

R=erik.corry@gmail.com, haraken@chromium.org, wibling@chromium.org
BUG=347943
NOTRY=true

Review URL: https://codereview.chromium.org/185723006

git-svn-id: svn://svn.chromium.org/blink/trunk@168385 bbb929c8-8fbe-4397-9dbb-9b2b20218538