[webkit-reviews] review requested: [Bug 8210] Conditional XMLHttpRequest gets should pass 304 responses unchanged : [Attachment 9186] proposed fix

[bugzilla-request-daemon at opendarwin.org]

Alexey Proskuryakov <ap at nypop.com> has asked  for review:
Bug 8210: Conditional XMLHttpRequest gets should pass 304 responses unchanged
http://bugzilla.opendarwin.org/show_bug.cgi?id=8210

Attachment 9186: proposed fix
http://bugzilla.opendarwin.org/attachment.cgi?id=9186&action=edit

------- Additional Comments from Alexey Proskuryakov <ap at nypop.com>
(In reply to comment #9)
The WebKit side doesn't have a separate code path for XHR-initiated requests
(which is probably good), but other requests we are making do not carry
manually set conditional headers, so they are not affected. I am not sure if
bypassing the cache for all conditional requests is necessarily the expected
behavior for NSURLConnection.

(In reply to comment #10)
> (From update of attachment 9171 [edit])
> I think that _web_isConditionalRequest would be more efficient if it called
> valueForHTTPHeaderField: instead of keyEnumerator. It's almost going out of
its
> way to be slow!

  I am not sure if this change makes it much faster (probably depends on the
internal implementation of valueForHTTPHeaderField:), but the resulting code
definitely looks better, thank you!

> Also, why not return BOOL rather than bool since it's a Objective-C method?

Oops, fixed.