You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@cxf.apache.org by "Sergey Beryozkin (JIRA)" <ji...@apache.org> on 2011/09/25 23:01:26 UTC

[jira] [Commented] (CXF-3822) JAX-RS: A 406 status could should be returned for an improperly formatted Accept header

    [ https://issues.apache.org/jira/browse/CXF-3822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13114365#comment-13114365 ] 

Sergey Beryozkin commented on CXF-3822:
---------------------------------------

I agree 500 is a wrong status to return, but I'm thinking that 400 is more appropriate, given that it's a client bug ?

If we have Accept: foo and return 406, then what should happen if we have

Accept: foo
and 

@Produces("foo") ?

Technically, it's a match ? 
May be we should get some clarifications at the spec level
 

> JAX-RS: A 406 status could should be returned for an improperly formatted Accept header
> ---------------------------------------------------------------------------------------
>
>                 Key: CXF-3822
>                 URL: https://issues.apache.org/jira/browse/CXF-3822
>             Project: CXF
>          Issue Type: Bug
>          Components: JAX-RS
>    Affects Versions: 2.3.6, 2.4.2
>         Environment: Java 1.6
>            Reporter: Ka-Lok Fung
>            Priority: Minor
>             Fix For: 2.3.7, 2.4.3, 2.5
>
>         Attachments: cxf3822.trunk.patch.diff
>
>   Original Estimate: 0.25h
>  Remaining Estimate: 0.25h
>
> If you send an Accept header that doesn't contain a slash (i.e., /), CXF's JAX-RS implementation returns a HTTP status of 500 (Internal Server Error). Instead of this status, it would be preferred if a HTTP status of 406 (Not Acceptable) is returned.
> Here's the test header:
> bq. {{Accept: foo}}
> Here's the current response:
> bq. {{Status Code: 500 org.apache.cxf.interceptor.Fault: Media type separator is missing}}
> Here's the preferred response:
> bq. {{Status Code: 406 Not Acceptable}}
> I've attached a potential patch against trunk for this issue:

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira