I've committed the changes that I mentioned, here: http://dev.jquery.com/changeset/6432
The test suite is now passing in all browsers (save Opera 10b1, which has issues with :enabled/:disabled). --John On Tue, Jul 14, 2009 at 5:00 PM, John Resig <jere...@gmail.com> wrote: > So I refused to believe that such a massive bug actually existed. I did > some digging and built a test case: > http://ejohn.org/files/bugs/modified/ > http://ejohn.org/files/bugs/modified/headers.phps > > Sure enough, there's a complete failure. > > Opera <= 9.6 are incapable of doing .setRequestHeader("If-Modified-Since", > ...) or .setRequestHeader("If-None-Match", ...) (nothing is sent to the > server). > > All versions of Opera (including 10b1) return a status of 0 instead of 304 > (for when the content hasn't been modified). > > Thus, the test suite is failing in Opera 9.6 because the headers are never > even sent (and, thus, act like a normal web page) - in Opera 10 it hangs > because the tests are waiting for a success callback and are never receiving > one. > > This deserves a blog post, what a complete fail. > > So, as far as fixing this goes, I'm really not sure. There doesn't appear > to be any reasonable way to force Opera to send the headers. And even if we > could it's not clear if it's possible to write a feature detect to determine > if this will work properly (short of performing a full extra request to the > server, which is unacceptable). I'm glad things are starting to get better > in Opera 10, at least. > > I think we can change the xhr.status == 304 line to xhr.status == 304 || > xhr.status == 0 to catch this painful case in the interim - although this > will still cause the tests to fail in Opera 9.6. Ugh. I'll dig in to it a > bit more, what a mess. > > --John > > > > On Mon, Jul 13, 2009 at 7:42 PM, lawrence.pit <lawrence....@gmail.com>wrote: > >> >> Hi John >> >> > So it appears as if your patch doesn't work in Opera (the tests you >> > committed fail in Opera 9.6 and hang in Opera 10). >> >> Upon first inspection it appears Opera is not sending the If-Modified- >> Since header to the server even though it's definitely set on the xhr >> object (I've tested this with Opera 9.5). >> >> The link below suggests that setting the If-Modified-Since and If-None- >> Match headers didn't work with previous version of opera either, it's >> not a jquery specific issue. >> >> >> http://my.opera.com/qdb/blog/2008/09/28/an-jquery-bug-problem-feature-with-opera-9-27-ubuntu-with-etag-caching >> >> I don't think anything can be done about it except patching Opera. >> >> Using Opera 10 it hangs when you set the If-Modified-Since header on >> the xhr object. I have no insight in whether Opera is working on >> improving their xhr support for this version or not; clearly the Opera >> community needs to step up if they want proper if-modified-since and >> etag support. >> >> >> Lawrence >> >> >> > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "jQuery Development" group. To post to this group, send email to jquery-dev@googlegroups.com To unsubscribe from this group, send email to jquery-dev+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en -~----------~----~----~----~------~----~------~--~---
