Bug 38924 – Add a guarded window.timing binding.

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Tony Gentilcore]

Add a guarded window.timing binding.

[Tony Gentilcore]

Created attachment 55736 [details]
Patch

[WebKit Review Bot]

Attachment 55736 [details] did not pass style-queue:

Failed to run "['WebKitTools/Scripts/check-webkit-style', '--no-squash']" exit_code: 1
WebCore/ChangeLog:9:  Line contains tab character.  [whitespace/tab] [5]
Total errors found: 1 in 17 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Tony Gentilcore]

I've been working to get the final issues ironed out of the Web Timing spec and am interested in driving the WebCore implementation forward. This patch is intended to install the minimum amount of code to add the window.timing binding while still remaining disabled-by-default. Once this is in place, I can tackle the implementation piece-by-piece.

[Tony Gentilcore]

Created attachment 55762 [details]
Patch

[Dimitri Glazkov (Google)]

Comment on attachment 55762 [details]
Patch

WebCore/ChangeLog:27
 +  
Don't forget FeatureDefines.xcconfig (in four places) -- see http://trac.webkit.org/changeset/59174 :)


WebCore/page/DOMWindow.cpp:448
 +  
Typically, we guard these entirely.

WebCore/page/DOMWindow.cpp:581
 +  
Ditto.

WebCore/page/NavigationTiming.cpp:31
 +  #include "config.h"
And this one, too. See SVG or Inspector files for an example.

WebCore/page/NavigationTiming.h:33
 +  
And that one.

[Tony Gentilcore]

Created attachment 55801 [details]
Patch

[Tony Gentilcore]

(In reply to comment #5)
> (From update of attachment 55762 [details])
> WebCore/ChangeLog:27
>  +  
> Don't forget FeatureDefines.xcconfig (in four places) -- see http://trac.webkit.org/changeset/59174 :)

Good catch. Done.
> 
> 
> WebCore/page/DOMWindow.cpp:448
>  +  
> Typically, we guard these entirely.

Done.
> 
> WebCore/page/DOMWindow.cpp:581
>  +  
> Ditto.

Done.
> 
> WebCore/page/NavigationTiming.cpp:31
>  +  #include "config.h"
> And this one, too. See SVG or Inspector files for an example.

Done.
> 
> WebCore/page/NavigationTiming.h:33
>  +  
> And that one.

Done.

[Early Warning System Bot]

Attachment 55801 [details] did not build on qt:
Build output: http://webkit-commit-queue.appspot.com/results/2207158

[Tony Gentilcore]

Created attachment 55868 [details]
Patch

[Tony Gentilcore]

Created attachment 55887 [details]
Patch

[Tony Gentilcore]

Created attachment 56107 [details]
Patch

[Tony Gentilcore]

The ChangeLogs are updated per our offline discussion.

[Dimitri Glazkov (Google)]

Comment on attachment 56107 [details]
Patch

This does look good to me. Maciej, do you have any problems with adding this?

[Tony Gentilcore]

(In reply to comment #13)
> (From update of attachment 56107 [details])
> This does look good to me. Maciej, do you have any problems with adding this?

Maciej, if you have any questions about Web Timing, I'm happy to answer them.

[Tony Gentilcore]

Created attachment 56938 [details]
Patch

[WebKit Review Bot]

Attachment 56938 [details] did not build on win:
Build output: http://webkit-commit-queue.appspot.com/results/2269565

[Dimitri Glazkov (Google)]

Comment on attachment 56938 [details]
Patch

Given that IE9 implements this, I think we should forge full-steam ahead.

[Tony Gentilcore]

Created attachment 59660 [details]
Patch

[Tony Gentilcore]

I synced this up and added some missing files. Probably worth scanning for review again.

The major differences:
1. window.timing is now window.performance.timing.
2. I was missing a lot of platform-specific build files. Hopefully I have them all now.

Here is the mozilla patch:
https://bugzilla.mozilla.org/show_bug.cgi?id=570341

And here is window.msPerformance.timing in the IE9 PP3:
http://ie.microsoft.com/testdrive/Performance/msPerformance/Default.html

[Eric Seidel (no email)]

Attachment 59660 [details] did not build on mac:
Build output: http://webkit-commit-queue.appspot.com/results/3321724

[Dimitri Glazkov (Google)]

Comment on attachment 59660 [details]
Patch

The mac fail might be pointing to something being wrong with the xcodeproj changes. Can you look?

[Tony Gentilcore]

(In reply to comment #21)
> (From update of attachment 59660 [details])
> The mac fail might be pointing to something being wrong with the xcodeproj changes. Can you look?

Just patched it in on my mac. Will upload a fix shortly.

[David Levin]

Comment on attachment 59660 [details]
Patch

Based on comment about uploading a new patch to fix mac build.

[Tony Gentilcore]

Created attachment 59676 [details]
Fix xcode project

[Dimitri Glazkov (Google)]

Comment on attachment 59676 [details]
Fix xcode project

yay.

[WebKit Commit Bot]

Comment on attachment 59676 [details]
Fix xcode project

Rejecting patch 59676 from commit-queue.

Failed to run "[u'/Users/eseidel/Projects/CommitQueue/WebKitTools/Scripts/svn-apply', u'--reviewer', u'Dimitri Glazkov', u'--force']" exit_code: 1
Last 500 characters of output:
mac/ChangeLog
Hunk #1 succeeded at 1 with fuzz 3.
patching file WebKit/mac/Configurations/FeatureDefines.xcconfig
patching file WebKit2/ChangeLog
Hunk #1 succeeded at 1 with fuzz 3.
patching file WebKit2/Configurations/FeatureDefines.xcconfig
patching file WebKitLibraries/ChangeLog
patching file WebKitLibraries/win/tools/vsprops/FeatureDefines.vsprops
patching file WebKitTools/ChangeLog
Hunk #1 succeeded at 1 with fuzz 3.
patching file WebKitTools/Scripts/build-webkit
patching file configure.ac

Full output: http://webkit-commit-queue.appspot.com/results/3278765

[Tony Gentilcore]

Created attachment 59824 [details]
Patch for landing

[WebKit Commit Bot]

Comment on attachment 59824 [details]
Patch for landing

Clearing flags on attachment: 59824

Committed r61950: <http://trac.webkit.org/changeset/61950>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Eric Seidel (no email)]

I think this may have caused a delayed Windows break:
http://build.webkit.org/builders/Windows%20Release%20(Build)/builds/953/steps/compile-webkit/logs/stdio


C:\cygwin\home\buildbot\slave\win-release\build\WebKitBuild\obj\WebCore\DerivedSources\JSDOMWindow.cpp(1462) : error C2039: 'performance' : is not a member of 'WebCore::DOMWindow'
        c:\cygwin\home\buildbot\slave\win-release\build\WebCore\page\DOMWindow.h(84) : see declaration of 'WebCore::DOMWindow'

[Eric Seidel (no email)]

Looks like this turned on WebTiming support on windows:
http://trac.webkit.org/changeset/61950/trunk/WebKitLibraries/win/tools/vsprops/FeatureDefines.vsprops

Even though it's not implemented for JSC.  That seems like the error.

[Eric Seidel (no email)]

(In reply to comment #31)
> Looks like this turned on WebTiming support on windows:
> http://trac.webkit.org/changeset/61950/trunk/WebKitLibraries/win/tools/vsprops/FeatureDefines.vsprops
> 
> Even though it's not implemented for JSC.  That seems like the error.

I failed to understand how that file works.  Not sure how this is caused.  May need to roll this out?

[Adam Roben (:aroben)]

(In reply to comment #32)
> (In reply to comment #31)
> > Looks like this turned on WebTiming support on windows:
> > http://trac.webkit.org/changeset/61950/trunk/WebKitLibraries/win/tools/vsprops/FeatureDefines.vsprops
> > 
> > Even though it's not implemented for JSC.  That seems like the error.

Steve fixed this in r61980.

> I failed to understand how that file works.

All ENABLE flags are listed in this file, whether turned on or off. Flags that are turned on look like this:

<UserMacro
              Name="ENABLE_CHANNEL_MESSAGING"
              Value="ENABLE_CHANNEL_MESSAGING"
              PerformEnvironmentSet="true"
      />

Flags that are turned off look like this:

<UserMacro
              Name="ENABLE_BLOB_SLICE"
              Value=""
              PerformEnvironmentSet="true"
      />

This is pretty similar to how FeatureDefines.xcconfig works.

[Tony Gentilcore]

(In reply to comment #33)
> (In reply to comment #32)
> > (In reply to comment #31)
> > > Looks like this turned on WebTiming support on windows:
> > > http://trac.webkit.org/changeset/61950/trunk/WebKitLibraries/win/tools/vsprops/FeatureDefines.vsprops
> > > 
> > > Even though it's not implemented for JSC.  That seems like the error.
> 
> Steve fixed this in r61980.

Sorry for the breakage. Thank you for the fix and explanation!

> 
> > I failed to understand how that file works.
> 
> All ENABLE flags are listed in this file, whether turned on or off. Flags that are turned on look like this:
> 
> <UserMacro
>               Name="ENABLE_CHANNEL_MESSAGING"
>               Value="ENABLE_CHANNEL_MESSAGING"
>               PerformEnvironmentSet="true"
>       />
> 
> Flags that are turned off look like this:
> 
> <UserMacro
>               Name="ENABLE_BLOB_SLICE"
>               Value=""
>               PerformEnvironmentSet="true"
>       />
> 
> This is pretty similar to how FeatureDefines.xcconfig works.