Bug 35088
| Summary: | Setting src on <source> shouldn't restart the resource selection algorithm | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Simon Pieters (:zcorpan) <zcorpan> |
| Component: | Media | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED WORKSFORME | ||
| Severity: | Normal | CC: | arun.patole, eric.carlson |
| Priority: | P2 | ||
| Version: | 528+ (Nightly build) | ||
| Hardware: | PC | ||
| OS: | OS X 10.5 | ||
Simon Pieters (:zcorpan)
URLs (if applicable) :
Other browsers tested:
Add OK or FAIL after other browsers where you have tested this issue:
Chrome: FAIL
Firefox 3.x: has different bugs
IE 7: N/A
IE 8: N/A
Opera 10.50 beta: OK
What steps will reproduce the problem?
1. Consider:
<video><source></video>
<script>
onload = function(){
document.body.firstChild.firstChild.src = 'test.mp4';
}
</script>
What is the expected result?
This should not load anything, because the resource selection algorithm
should have been started during parsing and have discarded the <source>
element (so that "pointer" is between the <source> and the end of the list).
What happens instead?
Safari happily loads the video. The author should be required to
explicitly call load() per spec.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Simon Pieters (:zcorpan)
http://code.google.com/p/chromium/issues/detail?id=36120
Simon Pieters (:zcorpan)
If I run the script when clicking a button instead of onload, Safari still fails but Chrome passes (though I might have an older webkit for Safari).
Arun Patole
I couldn't reproduce this issue with the latest webkit code. Also the related chromium issue mentioned in Comment #1 is already closed.