Bug 37818

Summary: Auto-generate WebGLRenderingContext overloads in V8
Product: WebKit Reporter: Yaar Schnitman <yaar>
Component: New BugsAssignee: Nobody <webkit-unassigned>
Status: RESOLVED FIXED    
Severity: Normal CC: commit-queue, kbr
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Other   
OS: OS X 10.5   
Attachments:
Description Flags
Patch
none
Why fails?
none
Why fail too?
none
Patch
none
Patch none

Yaar Schnitman
Reported 2010-04-19 13:14:45 PDT
Auto-generate WebGLRenderingContext overloads in V8
Attachments
Patch (40.11 KB, patch)
2010-04-19 13:18 PDT, Yaar Schnitman 		no flags
DetailsFormatted DiffDiff
Why fails? (3.75 KB, text/plain)
2010-04-19 15:01 PDT, Yaar Schnitman 		no flags
Details
Why fail too? (832 bytes, text/plain)
2010-04-19 15:02 PDT, Yaar Schnitman 		no flags
Details
Patch (41.34 KB, patch)
2010-04-22 17:15 PDT, Yaar Schnitman 		no flags
DetailsFormatted DiffDiff
Patch (41.43 KB, patch)
2010-04-22 19:43 PDT, Yaar Schnitman 		no flags
DetailsFormatted DiffDiff
Show Obsolete (2) View All Add attachment proposed patch, testcase, etc.
Yaar Schnitman
Comment 1 2010-04-19 13:18:40 PDT
Created attachment 53712 [details] Patch
Yaar Schnitman
Comment 2 2010-04-19 13:19:32 PDT
Comment on attachment 53712 [details] Patch Work in progress
Yaar Schnitman
Comment 3 2010-04-19 13:24:11 PDT
Ken, this is still work in progress, but I have some questions: 1. Two WebGL tests fail on seemingly unrelated reasons (see 2 attachments). Any clues why would that happen? 2. The ExceptionCode argument seems not to be used really in texImage2D and texSubImage2D. Am I right? If so, this patch could be made much smaller. 3. Any other overloads I'm missing?
Yaar Schnitman
Comment 4 2010-04-19 15:01:11 PDT
Created attachment 53724 [details] Why fails? null-object-behaviour.html
Yaar Schnitman
Comment 5 2010-04-19 15:02:04 PDT
Created attachment 53725 [details] Why fail too? uniform-location.html
Kenneth Russell
Comment 6 2010-04-20 12:03:48 PDT
Comment on attachment 53712 [details] Patch The update to the texImageTest looks okay but since the JSC custom bindings aren't changed in this patch I suspect they will start failing when run inside the WebKit build. Have you verified this? If this is the case, then the JSC custom bindings need to be temporarily modified so that the new test passes. The failures look to me like your LayoutTests directory is out of date with respect to WebCore.
Yaar Schnitman
Comment 7 2010-04-22 17:15:02 PDT
Created attachment 54111 [details] Patch
Yaar Schnitman
Comment 8 2010-04-22 17:16:16 PDT
Comment on attachment 54111 [details] Patch Feedback incorporated. All tests succeed on chromium and vanilla webkit. Please cq+ if good.
Kenneth Russell
Comment 9 2010-04-22 17:30:51 PDT
There is a failure baked in to texImageTest-expected.txt. Either the JSC custom bindings should be modified so that this does not fail, or a FIXME should be added to the test (and another bug filed) to correct this failure in the future.
Yaar Schnitman
Comment 10 2010-04-22 19:43:16 PDT
Created attachment 54121 [details] Patch
Yaar Schnitman
Comment 11 2010-04-22 19:44:09 PDT
Comment on attachment 54121 [details] Patch Added a fixme with link to new bug: https://bugs.webkit.org/show_bug.cgi?id=38024
Adam Barth
Comment 12 2010-04-23 00:37:50 PDT
Comment on attachment 54121 [details] Patch I don't 100% understand the test change, but I trust you to follow up in the bug you filed. I love the massive removal of custom code.
WebKit Commit Bot
Comment 13 2010-04-23 02:42:11 PDT
Comment on attachment 54121 [details] Patch Clearing flags on attachment: 54121 Committed r58161: <http://trac.webkit.org/changeset/58161>
WebKit Commit Bot
Comment 14 2010-04-23 02:42:16 PDT
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.