[Webkit-unassigned] [Bug 37602] AppCache Progress Events - need to include additional info per the spec

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=37602





--- Comment #8 from Patrick Mueller <pmuellr at yahoo.com>  2010-05-26 13:42:46 PST ---
I've had some other things hit my radar, don't think I'll be able to continue with this for a while.  If someone gets to this before I get back to it, have at it.

Some responses to Michael:

- this code isn't ready for review :-)

- TPAC appears to be a w3 thing; are there links to "it had been decided to move away ..." discussions?  I completely concur, but it would be nice to have this info available easily.  It looks like there is some discussion here - anywhere else?

   http://www.w3.org/2009/11/02-webapps-minutes.html#item02

- re: computing "total".  I hadn't really studied this code all that closely; my assumption was that it would be possible to compute a total.  I was focusing more on getting something basically working first.  Given the mention that Chrome couldn't support total without changes to chromium (outside of webkit), I suspect it may be easier for the first actual implementation to set lengthComputable to false, set total to 0 always (except maybe the last one), and increment loaded each time the message is sent.

Other general notes:

- the existing test case is not sufficient; will need more.  As you can also note from the last patch, if we send a final progress event at the end, this will end up breaking some existing test cases, but they were easy to fix.

- really no clue if I did the PassRefPtr/etal stuff correctly.  Seemed like a best guess.  The reason why changes were needed for this was to be able to send events with data, instead of the existing bare events we have today.  To do that, the data needs to be kept in the heap; the existing code maintains a vector of eventtype ints, for queued events, which is all you need if you're not sending any data with your events ...

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