Bug 76143 – setAttributeNS should comply with the obscure rules of DOM2, just like createAttributeNS and createElementNS do

• Component: Tools / Tests
• Status: RESOLVED FIXED

[Eric Seidel (no email)]

setAttributeNS should comply with the obscure rules of DOM2, just like createAttributeNS and createElementNS do

http://samples.msdn.microsoft.com/ietestcenter/domcore/showdomcoretest.htm?Exception_Element_setAttributeNS3
http://samples.msdn.microsoft.com/ietestcenter/domcore/showdomcoretest.htm?Exception_Element_setAttributeNS4
http://samples.msdn.microsoft.com/ietestcenter/domcore/showdomcoretest.htm?Exception_Element_setAttributeNS5

Noticed that we didn't implement this oddity. :)

I implemented the createAttributeNS and createElementNS versions of this code for Acid3.

Don't make me do setAttributeNodeNS or getAttributeNodeNS too. :p

[Eric Seidel (no email)]

Created attachment 122160 [details]
Patch

[WebKit Review Bot]

Comment on attachment 122160 [details]
Patch

Rejecting attachment 122160 [details] from commit-queue.

New failing tests:
css2.1/20110323/abspos-containing-block-initial-004b.htm
css2.1/20110323/background-intrinsic-007.htm
css2.1/20110323/background-intrinsic-004.htm
css2.1/20110323/background-intrinsic-006.htm
css2.1/20110323/replaced-intrinsic-004.htm
css2.1/20110323/background-intrinsic-008.htm
css2.1/20110323/background-intrinsic-005.htm
compositing/images/direct-svg-image.html
css2.1/20110323/background-intrinsic-009.htm
css3/images/cross-fade-overflow-position.html
css2.1/20110323/replaced-intrinsic-002.htm
css2.1/20110323/background-intrinsic-002.htm
css2.1/20110323/replaced-intrinsic-005.htm
css2.1/20110323/abspos-containing-block-initial-004d.htm
css2.1/20110323/replaced-intrinsic-003.htm
css2.1/20110323/background-intrinsic-001.htm
http/tests/misc/SVGFont-delayed-load.html
css2.1/20110323/replaced-intrinsic-001.htm
css2.1/20110323/background-intrinsic-003.htm
Full output: http://queues.webkit.org/results/11170689

[WebKit Review Bot]

Created attachment 122173 [details]
Archive of layout-test-results from ec2-cq-03

The attached test failures were seen while running run-webkit-tests on the commit-queue.
Bot: ec2-cq-03  Port: Chromium  Platform: Linux-2.6.35-28-virtual-x86_64-with-Ubuntu-10.10-maverick

[WebKit Review Bot]

Comment on attachment 122160 [details]
Patch

Attachment 122160 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/11170723

New failing tests:
css2.1/20110323/replaced-intrinsic-005.htm
css2.1/20110323/background-intrinsic-004.htm
css2.1/20110323/background-intrinsic-006.htm
css2.1/20110323/replaced-intrinsic-004.htm
css2.1/20110323/background-intrinsic-008.htm
css2.1/20110323/background-intrinsic-005.htm
compositing/images/direct-svg-image.html
css2.1/20110323/background-intrinsic-009.htm
css3/images/cross-fade-overflow-position.html
css2.1/20110323/replaced-intrinsic-002.htm
css2.1/20110323/background-intrinsic-002.htm
css2.1/20110323/background-intrinsic-007.htm
css2.1/20110323/replaced-intrinsic-003.htm
css2.1/20110323/background-intrinsic-001.htm
http/tests/misc/SVGFont-delayed-load.html
css2.1/20110323/replaced-intrinsic-001.htm
css2.1/20110323/background-intrinsic-003.htm

[WebKit Review Bot]

Created attachment 122180 [details]
Archive of layout-test-results from ec2-cr-linux-03

The attached test failures were seen while running run-webkit-tests on the chromium-ews.
Bot: ec2-cr-linux-03  Port: Chromium  Platform: Linux-2.6.35-28-virtual-x86_64-with-Ubuntu-10.10-maverick

[Adam Barth]

Assigning to investigate test failures.

[Eric Seidel (no email)]

Sorry.  I will also take care of this tomorrow if you don't fix it today.

[Adam Barth]

> Sorry.  I will also take care of this tomorrow if you don't fix it today.

I've got it in the debugger.  I think it's a subtle issue with integrating with the XML parser.

[Eric Seidel (no email)]

I expect that the XML parser is using this DOM API when it really wants to use some parser API.  It may also be using the "wrong" namespaces or prefixes.

[Adam Barth]

The issue seems to be that we're not handling default namespaces correctly.

[Adam Barth]

Created attachment 123006 [details]
Patch

[Eric Seidel (no email)]

Comment on attachment 123006 [details]
Patch

createAttributeNS will have this same trouble too.

[Eric Seidel (no email)]

Comment on attachment 123006 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=123006&action=review

> LayoutTests/fast/dom/Element/script-tests/setAttributeNS-namespace-err.js:127
> +// Moz throws a "Not enough arguments" exception in these, we don't:

We probably should. :)

[Adam Barth]

Eric points out that createAttributeNS has a similar problem: https://bugs.webkit.org/show_bug.cgi?id=76579

[Adam Barth]

Comment on attachment 123006 [details]
Patch

I'm switching this to be reviewed by me since the ChangeLog says it's authored by Eric.

[WebKit Review Bot]

Comment on attachment 123006 [details]
Patch

Clearing flags on attachment: 123006

Committed r105379: <http://trac.webkit.org/changeset/105379>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.