[Webkit-unassigned] [Bug 51760] Changeset #67245 was incompletely implemented

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=51760





--- Comment #4 from Kyle Simpson <getify at gmail.com>  2011-02-01 14:51:12 PST ---
The reason I filed this bug is because some Webkit developers were (in bug #50115) working on implementing my "async=false" proposal, and there were questions as to why it was even necessary for Webkit. Upon investigating those questions, I discovered, contradicting my previous assumptions, that changeset #67245 was only applied to parser-inserted scripts, which is why the breakage I expected to see in Webkit nightlies was in fact not happening. 

I brought this to those devs' attention, asking if the changeset was intended for only parser-inserted scripts, and in comment-49 of bug #50115, Adam said: 

"It's not intentional. Please file a new bug (and ideally attach a patch)."

...which is why *this* bug was filed. His assertion is that changeset #67245 was *intended* (as clearly implied from the spec wording from which it came) to apply to both parser-inserted AND script-inserted script elements.

The whole "reason" for needing the "async=false" proposal from bug #50115 implemented in Webkit was based on my assumption that changeset #67245 was fully implemented for both types of script elements, which would have lead to breakage in script loaders.

You may be tempted to say "well, that's easy then, just leave things as-is and don't implement the breaking change." But as I said, I agree with Adam and I think the spec wording clearly indicates that the change *should* apply to both types of script elements. In my opinion, the proper thing is to do both this bug (which will break script loaders) and then bug #50115 (which will fix them again).

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.