Jeremy Quinn wrote:
On 22 Jun 2005, at 10:01, Sylvain Wallez wrote:
Gregor J. Rothfuss wrote:
Sylvain Wallez wrote:
Don't be! The issue with Safari is that its XMLHttpRequest doesn't
seem to consider responses of length zero as valid, which Firefox
and IE do.
The problem is that fixing this will require to add an additional
ajax-specific pipeline in each cforms sitemap, which I would like
to avoid.
http://bugzilla.opendarwin.org/enter_bug.cgi?product=WebKit
Sure. Funny thing is that existing entries for WebKit are mostly
about XHR!
I have seen other 'Ajax' type frameworks that allegedly work on
Safari, so there must be some solution (?)
Yes, but they certainly don't use responses with zero-length content
(i.e. just headers) which is what is used in CForms to indicate that
interaction on the form is finished and that the full page should be
reloaded. And that's this zero-length response that seem to have
problems with Safari.
It is possible to circumvent this by always sending a response, but I'd
like to avoid this as far as possible, as it would require an additional
pipeline in the sitemap just for this purpose. That's why I want to
investigate further this Safari problem before going to that
unsatisfying solution.
Sylvain
--
Sylvain Wallez Anyware Technologies
http://apache.org/~sylvain http://anyware-tech.com
Apache Software Foundation Member Research & Technology Director
