424212 - [FIX]xmlhttprequest abort method should set status to 0 when it changes readystate to 4

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Foteos Macrides]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9b5pre) Gecko/2008032005 Minefield/3.0b5pre
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9b5pre) Gecko/2008032005 Minefield/3.0b5pre

The fix for https://bugzilla.mozilla.org/show_bug.cgi?id=379824 does not set .status when Fx3b4 and Fx3b5pre (and presumably the earlier betas) handle the abort method by changing readystate to 4, so that the .status value ends up as 2147746065 (NS_ERROR_NOT_AVAILABLE).  IE's ActiveX implementation and IE7's XMLHttpRequest object sets .status to 0 in that case, and so it is typical to check for both 200 and 0 in the .onreadystatechange handler as an indication that an error has not occurred.  The 0 value similarly could be used for Fx.

The demonstration URL is equivalent to the minimal test case from Bug 379824.


Reproducible: Always

Steps to Reproduce:
1.Access the demonstration URL with Fx3 versus Fx2 or Fx1.5 versions (and other browsers).
2.
3.



This matter is discussed in the m.d.a.f thread starting with:

http://groups.google.com/group/mozilla.dev.apps.firefox/browse_thread/thread/40e0918f5161e7e6/e790d64c5409f05b?hl=en#e790d64c5409f05b

Comment 1

[Boris Zbarsky [:bzbarsky]]

restoring flags

Comment 2

[Boris Zbarsky [:bzbarsky]]

Attachment: Could do this

Jonas, what do you think?  Should we be doing this?  Should we do this for 1.9?

Comment 3

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Does the spec say anything here?

Comment 4

[Damon Sicore (:damons)]

-'ing.  Will take a patch.  Not a regression.

Comment 5

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 310843 [details] [diff] [review]
Could do this

Looking for approval.  Simple change that gives us IE compat.  Affects web compat, so probably best to get into the beta.

Comment 6

[Mike Beltzner [:beltzner, not reading bugmail]]

Comment on attachment 310843 [details] [diff] [review]
Could do this

a=beltzner

Comment 7

[Boris Zbarsky [:bzbarsky]]

Checked in, with test.

Comment 8

[Foteos Macrides]

Changing file name in the demonstration URL to indicate the current bug number.