Bug 18066

Summary: REGRESSION: createAttribute throws NAMESPACE_ERR exception
Product: WebKit Reporter: Romain Bohdanowicz <romain.bohdanowicz>
Component: DOMAssignee: Julien Chaffraix <jchaffraix>
Status: RESOLVED FIXED    
Severity: Major CC: ap, cdumez, eric
Priority: P1 Keywords: EasyFix, Regression
Version: 528+ (Nightly build)   
Hardware: Mac   
OS: OS X 10.5   
Attachments:
Description Flags
Test Case
none
Test Case
none
New Test Case
none
Add a boolean parameter to createAttributeNS to bypass namespace checks darin: review+

Romain Bohdanowicz
Reported 2008-03-25 07:20:15 PDT
Since Nightly r31232 webkit throws an exception if a namespace is used with the createAttribute method. Example : document.createAttribute("ba:profile");
Attachments
Test Case (519 bytes, patch)
2008-03-25 07:21 PDT, Romain Bohdanowicz 		no flags
DetailsFormatted DiffDiff
Test Case (576 bytes, text/html)
2008-03-28 03:38 PDT, Romain Bohdanowicz 		no flags
Details
New Test Case (576 bytes, text/html)
2008-03-28 03:39 PDT, Romain Bohdanowicz 		no flags
Details
Add a boolean parameter to createAttributeNS to bypass namespace checks (5.42 KB, patch)
2008-05-29 17:12 PDT, Julien Chaffraix 		darin: review+
DetailsFormatted DiffDiff
Show Obsolete (2) View All Add attachment proposed patch, testcase, etc.
Romain Bohdanowicz
Comment 1 2008-03-25 07:21:19 PDT
Created attachment 20025 [details] Test Case
Mark Rowe (bdash)
Comment 2 2008-03-25 11:58:29 PDT
Sounds like a side-effect of <http://webkit.org/blog/167/webkit-gets-an-a-on-acid3/>.
Eric Seidel (no email)
Comment 3 2008-03-25 13:26:31 PDT
I *knew* I should have created a createAttribute("foo:bar") test case... Should be easy to fix.
Romain Bohdanowicz
Comment 4 2008-03-28 03:34:23 PDT
Comment on attachment 20025 [details] Test Case ><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html><head> > <meta http-equiv="Content-type" content="text/html; charset=utf-8"> > <title>Test</title> > <script type="text/javascript" charset="utf-8"> > function test() > { > try > { > var attribute = document.createAttribute("ba:test"); window.alert("No bug with this WebKit"); > } > catch(e) > { > window.alert(e.message); > } > } > </script> > </head><body id="regression" onload="javascript:test();"> > ></body></html>
Romain Bohdanowicz
Comment 5 2008-03-28 03:35:37 PDT
Comment on attachment 20025 [details] Test Case ><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html><body><pre style="word-wrap: break-word; white-space: pre-wrap;"><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" > "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> ><html> > <head> > <meta http-equiv="Content-type" content="text/html; charset=utf-8"> > <title>Test</title> > <script type="text/javascript" charset="utf-8"> > function test() > { > try > { > var attribute = document.createAttribute("ba:test"); > window.alert("Bug not happening with this WebKit"); > } > catch(e) > { > window.alert(e.message); > } > } > </script> > </head> > <body id="regression" onload="javascript:test();"> > </body> ></html>
Romain Bohdanowicz
Comment 6 2008-03-28 03:38:46 PDT
Created attachment 20153 [details] Test Case
Romain Bohdanowicz
Comment 7 2008-03-28 03:39:39 PDT
Created attachment 20154 [details] New Test Case
Eric Seidel (no email)
Comment 8 2008-04-08 02:05:59 PDT
All regressions are P1.
Romain Bohdanowicz
Comment 9 2008-05-14 07:05:07 PDT
Still happens in r33431.
Julien Chaffraix
Comment 10 2008-05-29 17:12:08 PDT
Created attachment 21424 [details] Add a boolean parameter to createAttributeNS to bypass namespace checks
Darin Adler
Comment 11 2008-05-29 17:27:48 PDT
Comment on attachment 21424 [details] Add a boolean parameter to createAttributeNS to bypass namespace checks Looks good, r=me.
Julien Chaffraix
Comment 12 2008-06-02 12:55:20 PDT
Committed in r34315.
Lucas Forschler
Comment 13 2019-02-06 09:03:26 PST
Mass moving XML DOM bugs to the "DOM" Component.
Note You need to log in before you can comment on or make changes to this bug.