You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@camel.apache.org by "Claus Ibsen (Jira)" <ji...@apache.org> on 2022/09/02 09:02:00 UTC

[jira] [Resolved] (CAMEL-8625) camel-restlet - Form fields overwrite headers in InMessage

     [ https://issues.apache.org/jira/browse/CAMEL-8625?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Claus Ibsen resolved CAMEL-8625.
--------------------------------
    Resolution: Won't Fix

restlet is removed in camel 3

> camel-restlet - Form fields overwrite headers in InMessage
> ----------------------------------------------------------
>
>                 Key: CAMEL-8625
>                 URL: https://issues.apache.org/jira/browse/CAMEL-8625
>             Project: Camel
>          Issue Type: Improvement
>          Components: camel-restlet
>    Affects Versions: 2.13.4, 2.14.2, 2.15.1
>         Environment: Linux
>            Reporter: Tomasz Ptak
>            Priority: Minor
>             Fix For: Future
>
>         Attachments: DefaultRestletBindingFixed.java, FormParameter.java
>
>
> DefaultRestletBinding adds values from submitted form as headers in the inMessage of the exchange. As a result it is close to impossible to rely on any of the values provided.
> Suggested solution:
> a) form values should be stored separately, or
> b) populating headers from the form should fail if a header with this key already exists
> c) prefixing keys coming from the form with something, for instance "form."
> Solution a is a safer solution which will break compatibility with previous versions of Apache Camel (it can also be spread over a number of releases, first introducing new field for form values and logging a warning if form data is being extracted, next making it configurable and disabled by default, then removing this functionality).
> Solution b is backwards compatible if this behaviour is not used as a feature  (if it is, it should not). That said someone could still use it to populate RestletRequest-like headers in the Message.
> Solution c would be rather ok as well, but I like a more, personally.
> I can provide a patch if any solution is accepted.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)