All the Perl that's Practical to Extract and Report
use Perl Log In
perigrin (3495)
{chris} {at} {prather.org}
http://chris.prather.org/
AOL IM: marceusx (Add Buddy, Send Message)
After Middle School Chris bounced through various high schools around the state of Florida for a bit. He ended up getting a BA (in English) from the University of Central Florida at about the time others his age were getting their MAs. His (now) wife, their child, and he went to Europe for a bit (there was a programming job in Scotland) and came back unemployed where upon he got offered a job in St. Paul. He moved on from that job after a few years to one where he was paid to do the things he previously did as a hobby. That sounds much more exciting than it really was. Really his life is based on two kids, two cats, and a bunny.
WebKit-- # breaks Catalyst::Controller::REST
WebKit is has a broken Accept header, it puts text/xml and application/xml
first which breaks Catalyst::Action::REST’s default configuration and makes
the idea of being able to dispatch html/xhtml different from XML difficult at
best.
The reason it turns out is that the webkit developers cargo-culted from
Firefox, and someone in 2007 provided a patch. The
response to the patch is “buh buh Firefox is doing it!”. When I tried to
add the comment below I discovered that their bugzilla required me to log in
and didn’t appear to have an option for OpenID. Because I’m loath to create
yet another account to file a single comment on a bug I’ve included the text
here:
Firefox (3.1 beta 2 at least) is no longer sending this Accept headers.
Currently writing a REST-ful interface that renders XML different from
XHTML/HTML at the same URI is difficult (requires us to browser sniff and do
we really want to go back to that?) because Webkit based browsers will all
send then wrong Accept headers first. I actually ran into this while writing
an app that targets Android/iPhone and I had to disable my XML rendering code
(luckily it was a stub and not a required feature).
Last the logic “Firefox does this” is a fallacy that I thought most people
were taught better by their mothers at an early age … if Firefox were to
jump off a bridge should WebKit too?
help them think they understand.
--Larry Wall
it would solve a lot of pain (Score:1)
Catalyst::Request::REST::ForBrowsers (Score:2)
You should take a look at this plugin. If it doesn't handle WebKit properly, patches would be welcome.
I've found it very handy when trying to write a RESTful app where the browser-side and API-side have the same URIs.