Bug 85534 – Add test for subpixel precision in DOM APIs

• Component: Layout and Rendering
• Status: RESOLVED FIXED

[Emil A Eklund]

Add test for increased precision due to subpixel support in the getBoundingClientRect api. Also test that clientWidth returns the pixel snapped value.

[Emil A Eklund]

Created attachment 140074 [details]
Patch

[Emil A Eklund]

This will clearly need to land _after_ we turn on the flag. Just wanted to make sure it didn't get lost and have it ready to commit once the flag is flipped.

[Eric Seidel (no email)]

Comment on attachment 140074 [details]
Patch

OK.  I suspect this would be way cleaner using js-test-pre.js however. :)

[Emil A Eklund]

(In reply to comment #3)
> (From update of attachment 140074 [details])
> OK.  I suspect this would be way cleaner using js-test-pre.js however. :)

I'm surprised that you think so. The only thing js-test-pre would do is provide a log function at the cost of a lot of overhead as the test is not making use of any of the functionality that js-test-pre provides.

[Emil A Eklund]

I should add that I'd happily change it to use js-test-pre if you think it makes sense, I just didn't see the value.

The same applies to the other tests I've added as a part of 60318.

[WebKit Review Bot]

Comment on attachment 140074 [details]
Patch

Attachment 140074 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/12610341

New failing tests:
fast/sub-pixel/client-rect-has-subpixel-precision.html

[WebKit Review Bot]

Created attachment 140081 [details]
Archive of layout-test-results from ec2-cr-linux-01

The attached test failures were seen while running run-webkit-tests on the chromium-ews.
Bot: ec2-cr-linux-01  Port: <class 'webkitpy.common.config.ports.ChromiumXVFBPort'>  Platform: Linux-2.6.35-28-virtual-x86_64-with-Ubuntu-10.10-maverick

[Eric Seidel (no email)]

Comment on attachment 140074 [details]
Patch

Why would you want this to be a ref test instead of a dumpAsText test?

[Emil A Eklund]

(In reply to comment #8)
> (From update of attachment 140074 [details])
> Why would you want this to be a ref test instead of a dumpAsText test?

I want to make sure that the box is painted at the pixel-snapped size and that clientWidth returns the same width as is painted.

I suppose I could split it into two separate tests instead if you think that makes more sense/is easier to understand.

[Eric Seidel (no email)]

Wouldn't you want your reference html to use integer values then if you wish to show the box is painted a specific pixel boundaries?

[Emil A Eklund]

(In reply to comment #10)
> Wouldn't you want your reference html to use integer values then if you wish to show the box is painted a specific pixel boundaries?

I do. I could have sword I changed that. Good catch!

[Emil A Eklund]

Created attachment 140087 [details]
Patch

[Eric Seidel (no email)]

Comment on attachment 140087 [details]
Patch

OK.

[WebKit Review Bot]

Comment on attachment 140087 [details]
Patch

Attachment 140087 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/12620248

New failing tests:
fast/sub-pixel/client-rect-has-subpixel-precision.html

[WebKit Review Bot]

Created attachment 140112 [details]
Archive of layout-test-results from ec2-cr-linux-02

The attached test failures were seen while running run-webkit-tests on the chromium-ews.
Bot: ec2-cr-linux-02  Port: <class 'webkitpy.common.config.ports.ChromiumXVFBPort'>  Platform: Linux-2.6.35-28-virtual-x86_64-with-Ubuntu-10.10-maverick

[Emil A Eklund]

Committed r118184: <http://trac.webkit.org/changeset/118184>