No, it's more like

* may not be null

* is to long 

etc

Not providing IDs, credentials etc. Makes sense ?

On Wed, Sep 18, 2013 at 2:32 PM, Kris Borchers <kris@redhat.com> wrote:

I will say +1 for this as long as none of the error messages returned introduce security issues by exposing too much info.

On Sep 18, 2013, at 4:27 AM, Matthias Wessendorf <matzew@apache.org> wrote:

Following up on [1], currently we give 400 (Bad Request), with no good meaning, when creating/updating a PushApplication or a Variant;

The HTML page(!) simply says "The request sent by the client was syntactically incorrect "

As suggested in [1] a good idea is using Bean Validation and returning better errors. 

The result would be a 400 (Bad Request) response, containing the errors and their messages as a JSON object (instead of a HTML page).

E.g.:

{"name":"may not be null"}

This is a) more meaningful and we can also get rid of our 'hard-coded' validation rules, like:

https://github.com/aerogear/aerogear-unifiedpush-server/blob/master/src/main/java/org/jboss/aerogear/unifiedpush/rest/registry/applications/AndroidVariantEndpoint.java#L79-L82

-Matthias

[1] https://issues.jboss.org/browse/AGPUSH-255

--
Matthias Wessendorf

blog: http://matthiaswessendorf.wordpress.com/
sessions: http://www.slideshare.net/mwessendorf
twitter: http://twitter.com/mwessendorf

_______________________________________________
aerogear-dev mailing list
aerogear-dev@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/aerogear-dev

_______________________________________________
aerogear-dev mailing list
aerogear-dev@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/aerogear-dev

--
Matthias Wessendorf

blog: http://matthiaswessendorf.wordpress.com/
sessions: http://www.slideshare.net/mwessendorf
twitter: http://twitter.com/mwessendorf

_______________________________________________
aerogear-dev mailing list
aerogear-dev@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/aerogear-dev