Bug 100393 – Generate Settings from a .in file

• Component: WebCore Misc.
• Status: RESOLVED FIXED

[Tony Chang]

We can also use this to generate InternalSettings.  The goal is to make it so adding a new setting is just adding a single line to an in file.  Tests will then be able to use internals.settings to override the pref and we can make sure the generated code properly resets the setting at the end of the test.

This would obsolete testRunner.overridePreference.

[Tony Chang]

This is motivated by the discussion on webkit-dev:
http://lists.webkit.org/pipermail/webkit-dev/2012-September/022323.html

Oh, we could also try to automate or simplify adding exports.

[Alexey Proskuryakov]

> This would obsolete testRunner.overridePreference.

How would it obsolete that? We can't go directly to settings if we want to test how preferences work.

[Tony Chang]

(In reply to comment #2)
> > This would obsolete testRunner.overridePreference.
> 
> How would it obsolete that? We can't go directly to settings if we want to test how preferences work.

That's true, we should keep testRunner.overridePreference so we can test WebKit API methods.  However, we should be able to remove some API methods that were added to overridePreference just for DRT testing.

[Tony Chang]

Created attachment 173410 [details]
Patch

[Tony Chang]

Comment on attachment 173410 [details]
Patch

I expect some ews failures.

[Tony Chang]

Created attachment 173411 [details]
generated SettingsMacros.h

Here's what the generated file looks like.

[Early Warning System Bot]

Comment on attachment 173410 [details]
Patch

Attachment 173410 [details] did not pass qt-wk2-ews (qt):
Output: http://queues.webkit.org/results/14790292

[EFL EWS Bot]

Comment on attachment 173410 [details]
Patch

Attachment 173410 [details] did not pass efl-ews (efl):
Output: http://queues.webkit.org/results/14778610

[Build Bot]

Comment on attachment 173410 [details]
Patch

Attachment 173410 [details] did not pass win-ews (win):
Output: http://queues.webkit.org/results/14785382

[kov's GTK+ EWS bot]

Comment on attachment 173410 [details]
Patch

Attachment 173410 [details] did not pass gtk-ews (gtk):
Output: http://queues.webkit.org/results/14779632

[Build Bot]

Comment on attachment 173410 [details]
Patch

Attachment 173410 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/14781492

[Tony Chang]

Created attachment 173668 [details]
Patch

[Build Bot]

Comment on attachment 173668 [details]
Patch

Attachment 173668 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/14813428

[EFL EWS Bot]

Comment on attachment 173668 [details]
Patch

Attachment 173668 [details] did not pass efl-ews (efl):
Output: http://queues.webkit.org/results/14818248

[Tony Chang]

Created attachment 173686 [details]
Patch

[EFL EWS Bot]

Comment on attachment 173686 [details]
Patch

Attachment 173686 [details] did not pass efl-ews (efl):
Output: http://queues.webkit.org/results/14811526

[Tony Chang]

Created attachment 173701 [details]
Patch

[Tony Chang]

Comment on attachment 173701 [details]
Patch

I'm not sure what's up with the Mac EWS bot, but I think this is ready for review now.

[Adam Barth]

Comment on attachment 173701 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=173701&action=review

> Source/WebCore/page/make_settings.pl:3
> +# Copyright (C) 2011 Adam Barth <[e-mail]>

I wrote this file!  I'm amazing.

> Source/WebCore/page/make_settings.pl:208
> +    substr($setterFunctionName, 3, 1) = uc(substr($setterFunctionName, 3, 1));
> +    if (substr($settingName, 0, 3) eq "css" || substr($settingName, 0, 3) eq "xss" || substr($settingName, 0, 3) eq "ftp") {
> +        substr($setterFunctionName, 3, 3) = uc(substr($setterFunctionName, 3, 3));
> +    }

Yeah, this is probably better than writing a general camel-case converter.

[Adam Barth]

Comment on attachment 173701 [details]
Patch

InternalsSettings is next?

[Tony Chang]

(In reply to comment #20)
> (From update of attachment 173701 [details])
> InternalsSettings is next?

Soon.  There are still lots of simple getters/setters in Settings.h that I'm going to fix up first (e.g., isScriptEnabled/setScriptEnabled, inApplicationChromeMode/setApplicationChromeMode, isCSSCustomFilterEnabled/setCSSCustomFilterEnabled, scrollAnimatorEnabled/setEnableScrollAnimator).

[Build Bot]

Comment on attachment 173701 [details]
Patch

Attachment 173701 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/14826292

[Tony Chang]

Created attachment 174004 [details]
Patch for landing

[Build Bot]

Comment on attachment 174004 [details]
Patch for landing

Attachment 174004 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/14832087

[Tony Chang]

Created attachment 174021 [details]
Patch

[Tony Chang]

Created attachment 174191 [details]
Patch

[Build Bot]

Comment on attachment 174191 [details]
Patch

Attachment 174191 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/14847254

New failing tests:
inspector-protocol/nmi-webaudio.html

[WebKit Review Bot]

Comment on attachment 174191 [details]
Patch

Clearing flags on attachment: 174191

Committed r134828: <http://trac.webkit.org/changeset/134828>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.