Bug 36279

Summary: CSS3 :nth-child selector and valid whitespace test fails
Product: WebKit Reporter: Simon Fraser (smfr) <simon.fraser>
Component: CSSAssignee: Simon Fraser (smfr) <simon.fraser>
Status: RESOLVED FIXED    
Severity: Normal CC: abarth, commit-queue, darin, eric, jamesr, simon.fraser, vsvabe, webkit.review.bot, yael
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Mac   
OS: OS X 10.5   
URL: http://samples.msdn.microsoft.com/ietestcenter/css3/selectors/ValidWhitespace.htm
Attachments:
Description Flags
Testcase
none
Patch none

Description Simon Fraser (smfr) 2010-03-17 20:24:46 PDT 
Created attachment 51005 [details]
Testcase

This test fails:
http://samples.msdn.microsoft.com/ietestcenter/css3/selectors/ValidWhitespace.htm
Comment 1 Yael 2011-01-03 18:09:04 PST 
Created attachment 77859 [details]
Patch

Teach the tokenizer to allow white spaces around [+/-] signs in nth constructs.
Comment 2 Darin Adler 2011-01-03 21:39:25 PST 
Comment on attachment 77859 [details]
Patch

Changing the tokenizer but not updating CSSSelector::RareData::parseNth is not going to work correctly. It seems the test case here was insufficient to detect that. Please work on a patch that changes both the tokenizer and the parser to allow spaces!
Comment 3 Yael 2011-01-04 07:19:38 PST 
(In reply to comment #2)
> (From update of attachment 77859 [details])
> Changing the tokenizer but not updating CSSSelector::RareData::parseNth is not going to work correctly. It seems the test case here was insufficient to detect that. Please work on a patch that changes both the tokenizer and the parser to allow spaces!

Thank you for your review, Darin.
I was under the impression that CSSSelector::RareData::parseNth already allows the spaces, but I could be wrong.

CSSSelector::RareData::parseNth uses String::find() to find the sign + or -, and does not assume that the sign follows immediately after n.

CSSSelector::RareData::parseNth uses String::toInt() to find the value of b, and that already knows how to skip leading spaces.

Do you have a specific case in mind that could fail ?
Comment 4 Simon Fraser (smfr) 2011-01-04 12:13:59 PST 
RareData::parseNth() looks like it deals with whitespace OK. Darin, are you OK with the patch?
Comment 5 Darin Adler 2011-01-04 13:07:14 PST 
(In reply to comment #3)
> I was under the impression that CSSSelector::RareData::parseNth already allows the spaces

You are right. Sorry, I was wrong!
Comment 6 WebKit Commit Bot 2011-01-04 18:19:24 PST 
The commit-queue encountered the following flaky tests while processing attachment 77859 [details]:

http/tests/appcache/progress-counter.html bug 51913 (author: michaeln@google.com)
The commit-queue is continuing to process your patch.
Comment 7 WebKit Commit Bot 2011-01-04 18:21:25 PST 
Comment on attachment 77859 [details]
Patch

Clearing flags on attachment: 77859

Committed r75039: <http://trac.webkit.org/changeset/75039>
Comment 8 WebKit Commit Bot 2011-01-04 18:21:32 PST 
All reviewed patches have been landed.  Closing bug.
Comment 9 WebKit Review Bot 2011-01-04 19:51:55 PST 
http://trac.webkit.org/changeset/75039 might have broken Leopard Intel Debug (Tests)