Bug #46974 for Catalyst-Action-REST: 'Accept' header should have priority for the deserialization format vs. 'Content-Type'

Mon Jun 15 19:05:48 2009 ddascalescu+perl [...] gmail.com - Ticket created

Subject:

'Accept' header should have priority for the deserialization format vs. 'Content-Type'

Conversation in #catalyst: dandv: I'm expecting the response serialization to be based on the Accept header, then lacking that, to fall back on the Content-Type header. For example, a client could send a JSON request but expect a text/plain response. I know C::C::REST doesn't work that way, but is this assumption flawed? confound: Accept should be primary for response serialization and Content-Type for request deserialization, but right now Content-Type rules in both directions

Fri Apr 02 05:44:02 2010 icestar [...] inbox.ru - Correspondence added

From:

icestar [...] inbox.ru

On Mon Jun 15 19:05:48 2009, dandv wrote: Show quoted text

> Conversation in #catalyst: > > dandv: I'm expecting the response serialization to be based on the > Accept header, then lacking that, to fall back on the Content-Type > header. For example, a client could send a JSON request but expect a > text/plain response. I know C::C::REST doesn't work that way, but is > this assumption flawed? > > confound: Accept should be primary for response serialization and > Content-Type for request deserialization, but right now Content-Type > rules in both directions

I've found the same omission and agree with previous post. Sometimes it's useful to obtain data in one format and return in another.

Fri Apr 02 05:44:03 2010 The RT System itself - Status changed from 'new' to 'open'

Thu May 13 09:10:18 2010 http://www.gerv.net/ - Correspondence added

From:

Gerv

The bug is in Catalyst/TraitFor/Request: # First, we use the content type in the HTTP Request. It wins all. $types{ $self->content_type } = 3 if $self->content_type; The 3 needs to be a 0 or a -1 (and the comment changed). That would give this type a q value lower than a URL parameter or the Accept header. Gerv

Thu May 13 09:21:51 2010 bobtfish [...] bobtfish.net - Correspondence added

I'd love someone to work up some failing tests and a patch to fix them for this issue, and I'd happily apply it. However I don't have time to work on this issue myself at the moment :(

Fri Apr 08 08:17:36 2016 abraxxa [...] cpan.org - Correspondence added

Just hit the same problem, subscribing in case someone fixes it.