[webkit-reviews] review denied: [Bug 77854] [Gtk] XMLHttpRequest modified during onreadystatechange handler : [Attachment 169167] Patch

[bugzilla-daemon at webkit.org]

Alexey Proskuryakov <ap at webkit.org> has denied Mark Toller
<mark.toller at samsung.com>'s request for review:
Bug 77854: [Gtk] XMLHttpRequest modified during onreadystatechange handler
https://bugs.webkit.org/show_bug.cgi?id=77854

Attachment 169167: Patch
https://bugs.webkit.org/attachment.cgi?id=169167&action=review

------- Additional Comments from Alexey Proskuryakov <ap at webkit.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=169167&action=review


r- for lack of ChangeLog, test cases, and for broken regression test.

I do not understand how this Gtk-only bug has come to modify cross-platform
behavior. Could you please explain that in more detail?

> Source/WebCore/xml/XMLHttpRequest.cpp:1024
> +    // From the http://www.w3.org/TR/XMLHttpRequest/#the-status-attribute:
> +    // If the error flag is set, return 0 and terminate these steps.

This comment is not very helpful. It just describes what the code below does,
and provides a link to XHR spec.

First, anyone working on this code would know where the spec is anyway. But
also, matching an ancient version of the spec will not be relevant five years
from now. We make such behavior decisions based on multiple factors, including
spec compliance, and our own compatibility testing. The desired behavior is
captured in regression tests, not in code comments, unless there is something
exceptionally interesting to say.