[Webkit-unassigned] [Bug 84893] [UserTiming] UserTiming metabug

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Jul 9 17:12:50 PDT 2012


https://bugs.webkit.org/show_bug.cgi?id=84893





--- Comment #10 from James Simonsen <simonjam at chromium.org>  2012-07-09 17:12:48 PST ---
Thanks for the patch. Just a couple of high level comments at this point:

- I probably shouldn't have said "ready for implementation" without taking another pass through the spec. Sorry. The optional return types are kinda icky. I'd rather we just used the Performance Timeline getEntries*() functions. I've started a thread on public-webperf. You might want to hold off until that's settled.

- Everything needs to be vendor-prefixed until the spec moves forward. So all of the functions need to start with "webkit".

- I'd prefer to use the W3C style test suites instead of adding more WebKit specific ones. It'd be better to just write one set of tests and have them work for both WebKit and the W3C. I have bug 84887 that imports the W3C tests for Navigation Timing. Maybe we can build off that.

- User Timing should integrate with Performance Timeline. Maybe the spec isn't very clear on that, but the returned objects (PerformanceMeasure and PerformanceMark) should derive from PerformanceEntry. They should also be returned when you call the Performance Timeline getEntries*() functions. If the spec isn't clear, please post comments on public-webperf.

- This is a metabug. It shouldn't have any patches on it. We should create additional bugs to implement aspects of the spec and upload patches to those. Those bugs would block this metabug. For instance, you could have one bug for the main implementation and another to integrate it with Performance Timeline and one to remove the vendor prefixes when ready. Then you wouldn't have to implement everything all at once either. I suspect there'll be some other tweaks here and there as we go too. That's how the other timing specs have been.

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



More information about the webkit-unassigned mailing list