WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
6520
Add support for insertAdjacentElement
https://bugs.webkit.org/show_bug.cgi?id=6520
Summary
Add support for insertAdjacentElement
Geoffrey Garen
Reported
2006-01-12 20:39:05 PST
http://msdn.microsoft.com/workshop/author/dhtml/reference/methods/insertadjacentelement.asp
Attachments
Darin's rough cut patch
(2.19 KB, patch)
2006-01-12 20:40 PST
,
Geoffrey Garen
no flags
Details
Formatted Diff
Diff
simple testcase
(770 bytes, text/html)
2006-03-09 11:23 PST
,
Alexander Kellett
no flags
Details
patch taking into account move to Element, and implementing bindings, no exceptions thrown yet
(3.70 KB, patch)
2006-03-09 11:37 PST
,
Alexander Kellett
darin
: review-
Details
Formatted Diff
Diff
latest testcase
(1.59 KB, text/html)
2006-03-19 09:57 PST
,
Alexander Kellett
no flags
Details
latest testcase
(1.87 KB, text/html)
2006-03-19 10:51 PST
,
Alexander Kellett
no flags
Details
testcases
(2.66 KB, patch)
2006-03-26 21:06 PST
,
Alexander Kellett
darin
: review+
Details
Formatted Diff
Diff
patch taking into account move to Element, and implementing bindings, no exceptions thrown yet
(4.13 KB, patch)
2006-03-26 21:06 PST
,
Alexander Kellett
darin
: review+
Details
Formatted Diff
Diff
Show Obsolete
(5)
View All
Add attachment
proposed patch, testcase, etc.
Geoffrey Garen
Comment 1
2006-01-12 20:40:26 PST
Created
attachment 5638
[details]
Darin's rough cut patch "A rough first cut of the method only, not the JavaScript binding."
Anders Carlsson
Comment 2
2006-01-13 08:32:57 PST
I did some investigation on how insertAdjacentElement works in WinIE: * This function only exists on elements, so it should be moved to ElementImpl * Passing a non-element object as newChild throws an E_INVALIDARG COM exception * If the element has no parent and where is "beforeBegin" or "afterEnd", a document fragment is created and the elements appended in the correct order. This document fragment isn't returned anywhere * Passing a bogus string as where also throws an E_INVALIDARG exception.
Alexander Kellett
Comment 3
2006-03-08 14:08:35 PST
does this need to be bound to HTMLElement or DOMElement?
Alexander Kellett
Comment 4
2006-03-09 11:23:01 PST
Created
attachment 6960
[details]
simple testcase
Alexander Kellett
Comment 5
2006-03-09 11:37:06 PST
Created
attachment 6961
[details]
patch taking into account move to Element, and implementing bindings, no exceptions thrown yet
Alexander Kellett
Comment 6
2006-03-11 13:41:57 PST
andersca, not sure i understand your 3rd point? there is a side effect? does dom tree walking after insertion result in the added element having been inserted into the fragment? what would be the equiv of a "E_INVALIDARG COM exception" for webkit? can't really go any further at this point as i don't have ie. going to mark for review and hope for some feedback.
Darin Adler
Comment 7
2006-03-11 22:10:10 PST
Comment on
attachment 6961
[details]
patch taking into account move to Element, and implementing bindings, no exceptions thrown yet This should be done with an new style IDL-file-based binding, not an old style hand-written binding, because it has no special requirements. So it should be added to Element.idl and kjs_dom.cpp and kjs_dom.h should be left untouched.
Alexander Kellett
Comment 8
2006-03-12 02:21:58 PST
andersca had already said not to do so.. tho i forget the reasoning.
Darin Adler
Comment 9
2006-03-13 20:49:28 PST
(In reply to
comment #8
)
> andersca had already said not to do so.. tho i forget the reasoning.
Anders, why?
Alexander Kellett
Comment 10
2006-03-13 21:50:24 PST
spoke to anders since my last comment and realized that his comment was in response to something else :P this works just fine - + // IE only extension + Node insertAdjacentElement(in DOMString position, + in Node newAttr) + raises(DOMException); + do i need to fill out the exception throwing before committing (after another review) or is this enough for the moment?
Darin Adler
Comment 11
2006-03-14 10:56:02 PST
I'd like to see the test cases and exceptions in before we land this. Shouldn't be too difficult to research how it works and make a nice thorough case. I don't want to have to revisit this later.
Alexander Kellett
Comment 12
2006-03-14 12:20:46 PST
still not sure what andersca's reference to "E_INVALIDARG" implies, is this just a case of adding that to the DOMException class?
Anders Carlsson
Comment 13
2006-03-14 12:55:32 PST
Since the JS wrappers in WinIE wrap COM objects, there are some native COM exceptions that are being used in WinIE, for example the E_INVALIDARG exception.
Alexander Kellett
Comment 14
2006-03-19 09:57:40 PST
Created
attachment 7175
[details]
latest testcase
Alexander Kellett
Comment 15
2006-03-19 10:51:11 PST
Created
attachment 7176
[details]
latest testcase
Alexander Kellett
Comment 16
2006-03-26 21:06:18 PST
Created
attachment 7320
[details]
testcases
Alexander Kellett
Comment 17
2006-03-26 21:06:54 PST
Created
attachment 7321
[details]
patch taking into account move to Element, and implementing bindings, no exceptions thrown yet
Darin Adler
Comment 18
2006-03-26 22:46:45 PST
Comment on
attachment 7321
[details]
patch taking into account move to Element, and implementing bindings, no exceptions thrown yet Looks fine. r=me
Eric Seidel (no email)
Comment 19
2006-03-27 17:07:02 PST
Lypie commited this.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug