XHR: reset error flag on aborting request that restarts.

XHR: reset error flag on aborting request that restarts.

Close a small window in internal abort logic that takes care of
repeated uses of the XHR object upon cancellation.

The possibility existed that such a repeated use might reset the error
flag, so that upon return to the outer request (the one that is
aborting), it was left as 'false'. Spot this and (re) set the error
flag to 'true'.

Not a common use of XHR; added test coverage.

R=
BUG=315488
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=162346

[sof, 2013-11-19 09:14:10]

At your own leisure, please take a look.

The final piece from crrev.com/61793010, taking care of a problem in the
handling of the request object's error flag, as uncovered by tyoshino during the
review.

[tyoshino (SeeGerritForStatus), 2013-11-19 17:07:41]

lgtm.

thanks. the reentrant test result is much more readable

[sof, 2013-11-19 19:01:00]

On 2013/11/19 17:07:41, tyoshino wrote:
> lgtm.
> 
> thanks. the reentrant test result is much more readable

Verified locally that the crashers are reproducible with the re-entrancy
handling disabled from internalAbort().

Needs a four letter acronym starting with an L from a core/ owner? :)

[eseidel, 2013-11-19 19:05:39]

lgtm

[commit-bot: I haz the power, 2013-11-19 19:06:20]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sigbjornf@opera.com/76133002/1

[commit-bot: I haz the power, 2013-11-19 21:25:39]

Retried try job too often on win_blink_rel for step(s) webkit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_blink_...

[commit-bot: I haz the power, 2013-11-20 07:51:26]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sigbjornf@opera.com/76133002/1

[commit-bot: I haz the power, 2013-11-20 08:38:25]

Change committed as 162346