[jira] Commented: (AXIS2-92) Axis2 Addressing does not check for the legal URI values at right places

["Sanjiva Weerawarana (JIRA)" <ji...@apache.org>]

    [ http://issues.apache.org/jira/browse/AXIS2-92?page=comments#action_12316647 ] 

Sanjiva Weerawarana commented on AXIS2-92:
------------------------------------------

Given what we're doing with our private HTTP transport (i.e., getting rid of it), I'd rather not see us introducing effectively redundant code. If we really want to check whether the string is a real URI, then we could insert the following:

try { new java.net.URI (uriString); }

after every place where we read in a supposed URI as a string. I'd like to say we close this issue and if this becomes a problem then we go back and add this line everywhere as needed. (And also add "throws MalformedURIException" of course.)

Sanjiva.

> Axis2 Addressing does not check for the legal URI values at right places
> ------------------------------------------------------------------------
>
>          Key: AXIS2-92
>          URL: http://issues.apache.org/jira/browse/AXIS2-92
>      Project: Apache Axis 2.0 (Axis2)
>         Type: Bug
>     Reporter: Srinath Perera

>
> FYI: based on thread "[Axis2] Fault in handling values in addressing parameters"
> Most values (messageID,Action, address in a EPR ) in the Adressing is
> typed as URI in the Spec but our implementation and the dispatching
> treats them as string.
> This relaxation aollow our Axis2 to send wrong Addressing headers
> e.g.
> <wsa:MessageID> 11234 </wsa:MessageID>
> I belive we should check for and reject the non URI values that are
> set, Shall I add a JIRA issue?
> Thanks
> Srinath
>  
>  
> --------------------------------------------------------------------------------
> This was done intentionally as some people suggested me that java URI
> handling is very slow. So I put String for them purposely.
> But as you proposed we can do a check on these values in
> AddressingOutHandler.
> What abt others think on this ....
> Regards,
> Chinthaka
> --------------------------------------------------------------------------------
> +1 for checking in the out handler but I also think parsing those
> strings as URIs is waaay overkill. Maybe a small "parser" to check that
> the strings have a scheme would be sufficient.
> Sanjiva.
> --------------------------------------------------------------------------------
> +1 to use a efficient our URI class .. I think rather than do it in
> the out handler we should change the signatures of our methods to
> accept our URI and do the check in the URI constructor.
> Thanks
> Srinath
>  
>  

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira