GitLab

WTF::String can rely on implicit move constructor and assignment
operator as WTF::RefPtr class has them defined now (r177066).

Review URL: https://codereview.chromium.org/367343003

git-svn-id: svn://svn.chromium.org/blink/trunk@178541 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL adds new hasTagName() functions that have the efficiency of
hasLocalName(), but are safe. We can always use hasTagName(), and we'll get a
compile time error if we try to use an SVG tag name with an HTML element.

This makes the code safer and no longer requires good understanding of
hasTagName() vs hasLocalName() to write efficient code. This CL drops support
for hasLocalName(QualifiedName) as calling hasTagName() is now as efficient.

This is based on WebKit r165699 by darin@apple.com: 
http://trac.webkit.org/changeset/165699

R=esprehn@chromium.org, adamk@chromium.org, eseidel@chromium.org

Review URL: https://codereview.chromium.org/406843002

git-svn-id: svn://svn.chromium.org/blink/trunk@178540 bbb929c8-8fbe-4397-9dbb-9b2b20218538

-Add back the dividers between failure cards.
The wall of text and buttons was overwhelming.
-Fix the heights of things so that the page contents
isn't all overflowing out of height:0 containers.
-Remove the useless divider in ct-builder-grid.
-Make the buttons look like esprehn's codereview tool.
-Add a new sheriff-friendly favicon.
-Have the results panel consist of two overflow:auto
regions. One for the menu and one for the results.
-Get rid of tabs on top. Instead have a header that
converts to a "go back" button when you are in the
results page.
-Have ct-tree-status be collapsed to 0 height when it's empty.
To make this work, have the default status be the empty
string so that the status attribute isn't removed when the value
is open.
-Fix ellipsizing CSS in ct-tree-status

NOTRY=true
R=abarth@chromium.org

Review URL: https://codereview.chromium.org/400353006

git-svn-id: svn://svn.chromium.org/blink/trunk@178539 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Also, fix the styling of FAIL failure types now that gtests use that
instead of having FAIL masquerade as TEXT.

BUG=395496
R=pdr@chromium.org

Review URL: https://codereview.chromium.org/409523002

git-svn-id: svn://svn.chromium.org/blink/trunk@178538 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is so you can see the non-ellipsized tree status and
also to have a link to the page for opening the tree.

NOTRY=true

Review URL: https://codereview.chromium.org/398653005

git-svn-id: svn://svn.chromium.org/blink/trunk@178537 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This broke when I changed the results comparison to use polymer components.
This isn't 100% the same as the old pixelzoomer code. I took this as an
opportunity to simplify it.

Also, change the 4-space indent to 2-space indent and replace
endOfMicroTask calls with requestAnimationFrame to match the rest of
the codebase.

NOTRY=true

Review URL: https://codereview.chromium.org/403983002

git-svn-id: svn://svn.chromium.org/blink/trunk@178536 bbb929c8-8fbe-4397-9dbb-9b2b20218538

NOTRY=true

Review URL: https://codereview.chromium.org/407903002

git-svn-id: svn://svn.chromium.org/blink/trunk@178535 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The current distribution of files doesn't make sense. I have an idea for a
better split that I'd like to try, but that's speculative, so first we should
reset to a good state.

Review URL: https://codereview.chromium.org/408443004

git-svn-id: svn://svn.chromium.org/blink/trunk@178534 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=391662

Review URL: https://codereview.chromium.org/408543002

git-svn-id: svn://svn.chromium.org/blink/trunk@178532 bbb929c8-8fbe-4397-9dbb-9b2b20218538

As per recent spec changes, the start time should no longer factor in to
AnimationPlayer sort order. I've left in the SortInfo class instead of
simply holding on to an unsigned sequence number since we may in the
future add a mechanism for changing sort order, although this is easy
to remove later if won't be adding such functionality.

Review URL: https://codereview.chromium.org/399033002

git-svn-id: svn://svn.chromium.org/blink/trunk@178531 bbb929c8-8fbe-4397-9dbb-9b2b20218538

-Make NetworkSimulator error when runTest calls are nested.
-Harden NetworkSimulator against exceptions.
-Fix incorrect nested runTest call in treestatus_unittests.js.
-Fix network requests in ct-results-by-builder-tests.html.

This exposes that some of the tests were not overriding net
and thus the tests now do a ton of network requests.
This will go away when https://codereview.chromium.org/402603007
lands, so it doesn't seem worth the effort to fix all the
broken cases.

NOTRY=true
R=abarth@chromium.org

Review URL: https://codereview.chromium.org/401163002

git-svn-id: svn://svn.chromium.org/blink/trunk@178529 bbb929c8-8fbe-4397-9dbb-9b2b20218538

-Replace some methods with sugarjs equivalents (e.g. endsWith and unique).
-For methods that are only used in one place, move them to the module
they're used in. These and almost all the code left in the scripts
directory will all go away with https://codereview.chromium.org/402603007.

NOTRY=true

Review URL: https://codereview.chromium.org/400423002

git-svn-id: svn://svn.chromium.org/blink/trunk@178528 bbb929c8-8fbe-4397-9dbb-9b2b20218538

NOTRY=true

Review URL: https://codereview.chromium.org/405853002

git-svn-id: svn://svn.chromium.org/blink/trunk@178527 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The port to polymer is (basically) complete. Sheriff-o-matic
is replacing garden-o-matic.

Only functional change in this patch is to sort the scripts
and imports in ct-sheriff-o-matic.html. The rest of the patch
is just deleting unused files.

There's more code to delete with unused functions in some of
the remaining files, but that can be done in a followup.

NOTRY=true

Review URL: https://codereview.chromium.org/405843002

git-svn-id: svn://svn.chromium.org/blink/trunk@178526 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Tidy up after r178426 and remove now-unused includes.

R=haraken@chromium.org
BUG=394828

Review URL: https://codereview.chromium.org/409433002

git-svn-id: svn://svn.chromium.org/blink/trunk@178525 bbb929c8-8fbe-4397-9dbb-9b2b20218538

In the spirit of sheriff-o-matic only showing problems that
need solving, there's no point in showing tree open status.
We might want to reconsider this in the future where we
show you the actual status message, but right now the open
state is hard-coded to say "Tree is open", which is not
information I need to see.

Also, cleaned up a couple minor polymer things:
-No need to set the attributes attribute if we're using publish.
-Initialize all the member variables on ct-tree-status.
-Don't use setAttribute. Should basically always set element
values by property.

NOTRY=true
R=abarth@chromium.org

Review URL: https://codereview.chromium.org/403723005

git-svn-id: svn://svn.chromium.org/blink/trunk@178524 bbb929c8-8fbe-4397-9dbb-9b2b20218538

[Blink] XMLTreeViewer should load resources from blink_resources.grd instead of relying upon xxd.py (Part-2)

Blink should make use of blink_resources.grd for the inline resources for displaying XML tree view.
This removes the dependency upon using xxd.py which embeds the resources as strings.

BUG=312586

Review URL: https://codereview.chromium.org/391963002

git-svn-id: svn://svn.chromium.org/blink/trunk@178522 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Oilpan: WebAudio: Apply the weak HashMap pattern to remove an entry from AudioContext::m_dirtyAudioSummingJunctions.

r178238, which made AudioSummingJunction::m_context a weak member, didn't
work. If the AudioContext and an AudioSummingJunction object become unreachable
togeter, AudioSummingJunction::m_context is not cleared and their destuction
order is not deterministic. We must not touch weak members in destructors.

With this CL, we register all of AudioSummingJunction objects to AudioContext,
and applies the weak HashMap pattern to clear m_dirtyAudioSummingJunctions.

BUG=392788

Review URL: https://codereview.chromium.org/393133003

git-svn-id: svn://svn.chromium.org/blink/trunk@178521 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This blink side patch has to be preceded to delete below reduntdant
19 header files in ui/gfx of chromium.
(Please refer to https://codereview.chromium.org/392013005/)
These header files need to be deleted 
since they were already moved from ui/gfx to ui/gfx/geometry.

[Rendunt 19 header files to be deleted on chromium-side]
ui/gfx/box_f.h
ui/gfx/point.h
ui/gfx/point3_f.h
ui/gfx/quad_f.h
ui/gfx/insets.h
ui/gfx/point_f.h
ui/gfx/vector2d.h
ui/gfx/matrix3_f.h
ui/gfx/size.h
ui/gfx/size_f.h
ui/gfx/rect.h
ui/gfx/rect_f.h
ui/gfx/rect_conversions.h
ui/gfx/size_conversions.h
ui/gfx/point_conversions.h
ui/gfx/vector2d_conversions.h
ui/gfx/safe_interget_conversions.h
ui/gfx/vector3d_f.h
ui/gfx/vector2d_f.h

BUG=395370

Review URL: https://codereview.chromium.org/400403002

git-svn-id: svn://svn.chromium.org/blink/trunk@178520 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Text wrapping is a little funny in an absolute-positioned element: the
available width of an off-screen element is zero which causes wrapping.
I created a minimal example of this is at http://pr.gg/wrapgenius.html
This behavior is well-known and there are workarounds [1].

The flakiness dashboard popup is similarly affected (see:
http://pr.gg/flakywrapping.png). Because the flakiness dashboard
revision popup has breaks already, we can just use
white-space: nowrap to fix this without any involved workarounds.

[1] http://stackoverflow.com/questions/14819574/preserve-normal-word-wrapping-inside-absolutely-positioned-container

NOTRY=true

Review URL: https://codereview.chromium.org/403553002

git-svn-id: svn://svn.chromium.org/blink/trunk@178519 bbb929c8-8fbe-4397-9dbb-9b2b20218538

r178351 enabled ScriptForbiddenScope in release builds, then we hit a lot of crashes from real-world plugins (see the below bugs). We should suppress the ScriptForbiddenScopes in Document.cpp and FrameView.cpp in release builds.

BUG=395299,395081,395164

Review URL: https://codereview.chromium.org/403133002

git-svn-id: svn://svn.chromium.org/blink/trunk@178518 bbb929c8-8fbe-4397-9dbb-9b2b20218538

when constructing IDBVersionChangeEvent instances only allowing the
correct enumeration values.

BUG=358067

Review URL: https://codereview.chromium.org/386883008

git-svn-id: svn://svn.chromium.org/blink/trunk@178517 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=393945, 302024

Review URL: https://codereview.chromium.org/390173003

git-svn-id: svn://svn.chromium.org/blink/trunk@178516 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Currently, text editor auto-inserts spaces/tabs in certain cases. For example,
if you hit "enter" when your cursor is located on indented line, the new line
will be auto-indented. If you hit "enter" one more time, a new line will be
added, but the previous empty line with trailing auto-added whitespaces will
persist.

This patch suggests removing auto-added indents on every next "enter" keypress.
Note, that they will be removed only if they have not been edited.

R=vsevik, apavlov

Review URL: https://codereview.chromium.org/402433007

git-svn-id: svn://svn.chromium.org/blink/trunk@178515 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The Image object is not on the heap, adjust transition type usage to
reflect that. Not doing so lead to

  virtual/threaded/compositing/webgl/webgl-copy-image.html

failing in Oilpan builds --

 http://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=virtual%2Fthreaded%2Fcompositing%2Fwebgl%2Fwebgl-copy-image.html

TBR=haraken@chromium.org,oilpan-reviews@chromium.org
BUG=392765
NOTRY=true

Review URL: https://codereview.chromium.org/401013003

git-svn-id: svn://svn.chromium.org/blink/trunk@178514 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=374912

Review URL: https://codereview.chromium.org/401923002

git-svn-id: svn://svn.chromium.org/blink/trunk@178513 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=395076

Review URL: https://codereview.chromium.org/407523002

git-svn-id: svn://svn.chromium.org/blink/trunk@178512 bbb929c8-8fbe-4397-9dbb-9b2b20218538

driver. This is for end to end testing for print preview.
BUG=388517

Review URL: https://codereview.chromium.org/401883007

git-svn-id: svn://svn.chromium.org/blink/trunk@178511 bbb929c8-8fbe-4397-9dbb-9b2b20218538

There's no guarantee that the postMessage will arrive before
the requestAnimationFrame.

NOTRY=true

Review URL: https://codereview.chromium.org/397343006

git-svn-id: svn://svn.chromium.org/blink/trunk@178509 bbb929c8-8fbe-4397-9dbb-9b2b20218538

"Copy image" menu item crashes on WebGL elements because the
copiedImage in HTMLCanvasElement gets cleared before it can
be copied to the clipboard.
Also fixed HTMLCanvasElement::copiedImage() so that the copy
is done on the latest frame composited to the screen.

BUG=392765

Review URL: https://codereview.chromium.org/389273002

git-svn-id: svn://svn.chromium.org/blink/trunk@178508 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=374912
TBR=rsleevi

Review URL: https://codereview.chromium.org/407703002

git-svn-id: svn://svn.chromium.org/blink/trunk@178507 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This ensures that we never GC a MediaQueryList that has listeners.
It looks like that already doesn't happen (perhaps due to how
MediaQueryListListener keeps references to the JS object), but
this seems safer.

Review URL: https://codereview.chromium.org/397403002

git-svn-id: svn://svn.chromium.org/blink/trunk@178506 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Add Traversal<>::firstAncestorOrSelf() overload taking a const reference in
argument, and returning a const pointer. Make use of this new function in
Element::rendererIsFocusable().

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

Review URL: https://codereview.chromium.org/397303005

git-svn-id: svn://svn.chromium.org/blink/trunk@178504 bbb929c8-8fbe-4397-9dbb-9b2b20218538

As currently we only use positionFromPaintInvalidationContainer to
check if it has changed, neither the issue nor the fix visibly affects
any behavior.

The fix is required by https://codereview.chromium.org/399173005.

Also removed roundedIntPoint because now we pass sub-pixel coordinates
to GraphicsLayer for invalidations
(after https://codereview.chromium.org/353403006/).

BUG=395217

Review URL: https://codereview.chromium.org/408483005

git-svn-id: svn://svn.chromium.org/blink/trunk@178503 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Following Chromium r284167 the "ibm866" encoding is supported and
we can enable the TextDecoder test for it.

R=jshin@chromium.org
BUG=277023

Review URL: https://codereview.chromium.org/403793003

git-svn-id: svn://svn.chromium.org/blink/trunk@178502 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Fixes a regression introduced by sorting the result types by name.

Also a minor style tweak so the builder names stay aligned.

NOTRY=true

Review URL: https://codereview.chromium.org/406523004

git-svn-id: svn://svn.chromium.org/blink/trunk@178501 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Rename CSS styles in grid layout tests to make the name more explicit. As the
default mode for auto-placement algorithm is "sparse".

BUG=384099

Review URL: https://codereview.chromium.org/401473003

git-svn-id: svn://svn.chromium.org/blink/trunk@178500 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=abarth@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/408603002

git-svn-id: svn://svn.chromium.org/blink/trunk@178498 bbb929c8-8fbe-4397-9dbb-9b2b20218538

https://codereview.chromium.org/337883003/ accidentally made this test flaky.
Fix this by calling finish from the event listener.

BUG=390445
R=yurys@chromium.org

Review URL: https://codereview.chromium.org/398413002

git-svn-id: svn://svn.chromium.org/blink/trunk@178497 bbb929c8-8fbe-4397-9dbb-9b2b20218538

- less code
- able to use features of text-based-repaint.js

Review URL: https://codereview.chromium.org/403843002

git-svn-id: svn://svn.chromium.org/blink/trunk@178496 bbb929c8-8fbe-4397-9dbb-9b2b20218538