Bug 37787 - nth-* pseudo-class parsing is incorrect
Summary: nth-* pseudo-class parsing is incorrect
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: CSS (show other bugs)
Version: 528+ (Nightly build)
Hardware: PC OS X 10.5
: P2 Normal
Assignee: Yuzo Fujishima
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-04-18 19:38 PDT by Yuzo Fujishima
Modified: 2010-05-17 21:53 PDT (History)
0 users

See Also:

Attachments
Fix nth-* pseudo-class parsing. (1.44 KB, patch)
2010-04-18 19:41 PDT, Yuzo Fujishima 		hamaji: review+
Details | Formatted Diff | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Yuzo Fujishima 2010-04-18 19:38:33 PDT 
WebCore::CSSSelector::RareData::parseNth doesn't check whether the search for '-' in the argument was successful.
This results in incorrect parsing.
For example, '3n' is parsed as if it were '3n-3'. The code has been working OK just accidentally.
Comment 1 Yuzo Fujishima 2010-04-18 19:41:06 PDT 
Created attachment 53650 [details]
Fix nth-* pseudo-class parsing.
Comment 2 Shinichiro Hamaji 2010-04-20 19:35:46 PDT 
Comment on attachment 53650 [details]
Fix nth-* pseudo-class parsing.

This change looks good.

How -n-3 will be parsed, by the way? I'm not sure, but I guess it will be parsed as -n-0, right?
Comment 3 Yuzo Fujishima 2010-04-27 02:13:05 PDT 
-n-3 is parsed correctly.

In determining m_b, '-' is searched by argument.find('-', 1), which finds the second '-' before '3'.
Comment 4 Yuzo Fujishima 2010-04-27 02:53:44 PDT 
Committed r58300: <http://trac.webkit.org/changeset/58300>
Comment 5 Yuzo Fujishima 2010-05-17 21:53:04 PDT 
The original behavior was actually incorrect and the patch did fix it.
Please see: https://bugs.webkit.org/show_bug.cgi?id=31267 .
I added tests there.