Reland: Move the user agent styles sheets to blink_resources.grd (Part 3)

Reland: Move the user agent styles sheets to blink_resources.grd (Part 3)

Blink should make use of blink_resources.grd for the inline resources of user agent stylesheets.
This removes the dependency upon using make-file-arrays.py which embeds these resources as strings.

Also the .rodata section of libblink_web (in component build mode) is reduced by ~33kb.

Part 1: https://codereview.chromium.org/436843004/
Part 2: https://codereview.chromium.org/422023008

The chromium dependent patches [1] and [2] have already landed.

[1] https://codereview.chromium.org/818563003
[2] https://codereview.chromium.org/819573002

BUG=312586
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=188875

[vivekg, 2015-01-15 17:18:39]

vivekg@chromium.org changed reviewers:
+ jochen@chromium.org, vivekg@chromium.org

[vivekg, 2015-01-15 17:18:40]

Hi jochen@,

I have combined both the patches i.e. the migration and the removal of
make-file-arrays.py in this single patch. 

This patch was reverted due to the failures reported on android_dbg_test_recipe
as:
C  832.869s Main  [==========] 522 tests ran.
C  832.869s Main  [  PASSED  ] 519 tests.
C  832.869s Main  [  FAILED  ] 3 tests, listed below:
C  832.869s Main  [  FAILED  ] BatteryMonitorIntegrationTest.DefaultValues
(TIMEOUT)
C  832.869s Main  [  FAILED  ] BatteryMonitorIntegrationTest.EventListener
(TIMEOUT)
C  832.869s Main  [  FAILED  ] BatteryMonitorIntegrationTest.ResolvePromise
(TIMEOUT)

The reason for these failure is that the BatteryMonitorIntegrationTest tries to
use ShellContentClient and it doesn't load the content_shell.pak [1]. When I
made it load the content_shell.pak explicitly just for verification, I got an
assertion failure here [2] when tried to access the path service. 

I am wondering, the code in [1] is used by other tests as well but it does pass
for them and fails for the above 3 cases alone. Do you have any pointers on
this? Or any other solution you may think of? It will be really helpful as this
seems to be the last step of this migration.

Thank you,
Vivek


[1]
https://code.google.com/p/chromium/codesearch#chromium/src/content/test/conte...

[2]
https://code.google.com/p/chromium/codesearch#chromium/src/base/android/jni_a...

[jochen (gone - plz use gerrit), 2015-01-16 10:52:12]

do you know why the test times out? is that a unit test or a browser test?

[vivekg, 2015-01-16 10:59:03]

On 2015/01/16 at 10:52:12, jochen wrote:
> do you know why the test times out? is that a unit test or a browser test?

This is a browsertest here:
content/browser/battery_status/battery_monitor_integration_browsertest.cc

The test times out as the render process asserts on empty string returned [1]. 
This happens due to ShellContentClient::GetDataResource [2] just relies upon the
shared ResourceBundle to do the job which returns an empty string. 
Other places, [3], we made the pak file loading explicit. 

I tried to do the same logic of [3] at the location, [4], but the returned path
is wrong. 

[1]
https://code.google.com/p/chromium/codesearch#chromium/src/third_party/WebKit...
[2]
https://code.google.com/p/chromium/codesearch#chromium/src/content/shell/comm...
[3]
https://code.google.com/p/chromium/codesearch#chromium/src/content/test/test_...
[4]
https://code.google.com/p/chromium/codesearch#chromium/src/content/browser/ba...

[jochen (gone - plz use gerrit), 2015-01-16 11:48:57]

i don't really know a lot about the android testing framework. i'd recommend to
find somebody with more expertise in this area to help you debugging this

[vivekg, 2015-01-16 12:29:19]

On 2015/01/16 at 11:48:57, jochen wrote:
> i don't really know a lot about the android testing framework. i'd recommend
to find somebody with more expertise in this area to help you debugging this

Oh sure. Thank you, I will try to find out that.

[vivekg, 2015-01-19 09:01:09]

On 2015/01/16 at 12:29:19, vivekg_ wrote:
> On 2015/01/16 at 11:48:57, jochen wrote:
> > i don't really know a lot about the android testing framework. i'd recommend
to find somebody with more expertise in this area to help you debugging this
> 
> Oh sure. Thank you, I will try to find out that.

jochen@, this requires the patch https://codereview.chromium.org/845673004 to
make the loading of necessary pak files. 
PTAL at both the patches. Thank you.

[vivekg, 2015-01-21 16:25:54]

jochen@ the dependent chromium patch got landed and the failing bot,
android_dbg_test_recipe, has turned green. Can you pls rubber stamp this so that
we can hit CQ.

[vivekg, 2015-01-22 16:52:59]

Friendly ping. :)

[jochen (gone - plz use gerrit), 2015-01-23 10:02:37]

lgtm

[vivekg, 2015-01-23 10:04:02]

The CQ bit was checked by vivekg@chromium.org

[commit-bot: I haz the power, 2015-01-23 10:04:23]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/849163003/1

[commit-bot: I haz the power, 2015-01-23 10:04:38]

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

[commit-bot: I haz the power, 2015-01-23 10:04:40]

Failed to apply patch for Source/core/rendering/RenderThemeChromiumAndroid.cpp:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file Source/core/rendering/RenderThemeChromiumAndroid.cpp
  Hunk #1 FAILED at 28.
  1 out of 2 hunks FAILED -- saving rejects to file
Source/core/rendering/RenderThemeChromiumAndroid.cpp.rej

Patch:       Source/core/rendering/RenderThemeChromiumAndroid.cpp
Index: Source/core/rendering/RenderThemeChromiumAndroid.cpp
diff --git a/Source/core/rendering/RenderThemeChromiumAndroid.cpp
b/Source/core/rendering/RenderThemeChromiumAndroid.cpp
index
53e4abdaac45b354b29ebd935e6681e721de7417..4b5f7588c5343e44f3d06f386a6a0892e76b5fb2
100644
--- a/Source/core/rendering/RenderThemeChromiumAndroid.cpp
+++ b/Source/core/rendering/RenderThemeChromiumAndroid.cpp
@@ -28,13 +28,13 @@
 
 #include "core/CSSValueKeywords.h"
 #include "core/InputTypeNames.h"
-#include "core/UserAgentStyleSheets.h"
 #include "core/rendering/PaintInfo.h"
 #include "core/rendering/RenderMediaControls.h"
 #include "core/rendering/RenderObject.h"
 #include "core/rendering/RenderProgress.h"
 #include "core/rendering/RenderSlider.h"
 #include "platform/LayoutTestSupport.h"
+#include "platform/PlatformResourceLoader.h"
 #include "platform/graphics/Color.h"
 #include "platform/scroll/ScrollbarTheme.h"
 #include "public/platform/Platform.h"
@@ -60,14 +60,15 @@ RenderThemeChromiumAndroid::~RenderThemeChromiumAndroid()
 
 String RenderThemeChromiumAndroid::extraMediaControlsStyleSheet()
 {
-    return String(mediaControlsAndroidCss, sizeof(mediaControlsAndroidCss));
+    return loadResourceAsASCIIString("mediaControlsAndroid.css");
 }
 
 String RenderThemeChromiumAndroid::extraDefaultStyleSheet()
 {
     return RenderThemeChromiumDefault::extraDefaultStyleSheet() +
-        String(themeChromiumLinuxCss, sizeof(themeChromiumLinuxCss)) +
-        String(themeChromiumAndroidCss, sizeof(themeChromiumAndroidCss));
+        loadResourceAsASCIIString("themeChromiumLinux.css") +
+        loadResourceAsASCIIString("themeChromiumAndroid.css");
+
 }
 
 void RenderThemeChromiumAndroid::adjustInnerSpinButtonStyle(RenderStyle* style,
Element*) const

[vivekg, 2015-01-23 10:26:17]

The CQ bit was checked by vivekg@chromium.org

[commit-bot: I haz the power, 2015-01-23 10:27:25]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/849163003/20001

[commit-bot: I haz the power, 2015-01-23 11:40:57]

Committed patchset #2 (id:20001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=188875