Change CSSImageValue's member variables from String to AtomicString

Change CSSImageValue's member variables from String to AtomicString

Before this patch, CSSImageValue has two String member variables for
URLs (a relative URL and an absolute URL). This means that there should
be two StringImpl objects for one CSSImageValue, even when they are same
URLs. This could be a problem when the URL is a very long data URL and
more memory could be consumed than necessary. Actually, this is the case
on GMail site on mobile platforms and about 136 [KiB] can be saved by this
CL on GMail.

This CL changes them from String to AtomicString so that they can share
a StringImpl object when the two URLs are same to reduce memory.

Note: I created this CL based on the result of memory-usage investigation
in Blink:
https://docs.google.com/document/d/1XnN8RAOJDeoiNR9A4LqaUt4PQ3GXLPUJUzVIY49oRRg/edit#
We found that there were two duplicated big data-url on GMail and
CSSImageValue's implementation was culprit.

BUG=n/a
TEST=n/a; no behavior change

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=202511

[hajimehoshi, 2015-09-14 06:44:44]

hajimehoshi@chromium.org changed reviewers:
+ haraken@chromium.org, tasak@google.com

[hajimehoshi, 2015-09-14 06:44:50]

PTAL

[haraken, 2015-09-14 06:46:00]

haraken@chromium.org changed reviewers:
+ alancutter@chromium.org, timloh@chromium.org

[haraken, 2015-09-14 06:46:01]

+timloh, +alan

[haraken, 2015-09-14 06:46:33]

You can explain how much memory is saved with this CL.

[hajimehoshi, 2015-09-14 06:48:49]

On 2015/09/14 06:46:33, haraken wrote:
> You can explain how much memory is saved with this CL.

Thanks, done.

[hajimehoshi, 2015-09-17 09:57:54]

ping

[haraken, 2015-09-17 10:06:20]

LGTM, but I want to have someone in the style team confirm this.

[Timothy Loh, 2015-09-18 01:52:30]

lgtm

https://codereview.chromium.org/1335183002/diff/40001/Source/core/css/parser/...
File Source/core/css/parser/CSSPropertyParser.cpp (right):

https://codereview.chromium.org/1335183002/diff/40001/Source/core/css/parser/...
Source/core/css/parser/CSSPropertyParser.cpp:594: String uri = value->string;
I think this line does a conversion, might as well make it convert to
AtomicString directly.

[hajimehoshi, 2015-09-18 04:23:08]

Thank you!

https://codereview.chromium.org/1335183002/diff/40001/Source/core/css/parser/...
File Source/core/css/parser/CSSPropertyParser.cpp (right):

https://codereview.chromium.org/1335183002/diff/40001/Source/core/css/parser/...
Source/core/css/parser/CSSPropertyParser.cpp:594: String uri = value->string;
On 2015/09/18 01:52:30, Timothy Loh wrote:
> I think this line does a conversion, might as well make it convert to
> AtomicString directly.

Done.

[hajimehoshi, 2015-09-18 04:23:59]

The CQ bit was checked by hajimehoshi@chromium.org

[hajimehoshi, 2015-09-18 04:23:59]

The patchset sent to the CQ was uploaded after l-g-t-m from
haraken@chromium.org, timloh@chromium.org
Link to the patchset: https://codereview.chromium.org/1335183002/#ps60001
(title: "Timothy's review")

[commit-bot: I haz the power, 2015-09-18 04:24:08]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1335183002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1335183002/60001

[commit-bot: I haz the power, 2015-09-18 04:26:09]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-18 04:26:10]

Try jobs failed on following builders:
  ios_dbg_simulator_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_dbg_simulator...)
  ios_rel_device_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_rel_device_ni...)

[hajimehoshi, 2015-09-18 05:54:16]

The CQ bit was checked by hajimehoshi@chromium.org

[hajimehoshi, 2015-09-18 05:54:17]

The patchset sent to the CQ was uploaded after l-g-t-m from
haraken@chromium.org, timloh@chromium.org
Link to the patchset: https://codereview.chromium.org/1335183002/#ps80001
(title: "(rebase)")

[commit-bot: I haz the power, 2015-09-18 05:54:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1335183002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1335183002/80001

[commit-bot: I haz the power, 2015-09-18 08:28:17]

Committed patchset #5 (id:80001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=202511