Fred, Ruchith, all,

first of all - thanks to Fred to take actions on all the open issues :-)

As for WSS-54: in the orginal implementation the "handleUsernameToken()"
checked the both types of passwords. After some discussions on the mailing
list (back in 2004, WSS4J's stoneage :-)  ) we modified the behaviour to
check only the hashed passwords. The main reason was (as far as I can remember):
- only for hashed passwords the WS-Security specs define how the validate
  it (using nonce, created time etc)
- the plain password is just "plain" text - no validation is specified, thus
  we decided not to implement a check into the handler but to leave the
  check to ther server application. You may refer to the follwoing archived 
  e-mail discussion:
  http://mail-archives.apache.org/mod_mbox/ws-fx-dev/200409.mbox/[EMAIL 
PROTECTED]

IMHO implementing this patch brakes a behaviour that WSS4J provides since long 
and thus may break applications.

Regards,
Werner

> -----Ursprüngliche Nachricht-----
> Von: ext Fred Dushin [mailto:[EMAIL PROTECTED] 
> Gesendet: Dienstag, 15. April 2008 01:51
> An: wss4j-dev
> Betreff: WSS-54
> 
> Hi Ruchith,
> 
> Could I ask you to take a look at Colm's patch for WSS-54?
> 
> https://issues.apache.org/jira/browse/WSS-54
> 
> I'm +1 on the change, but I see you had some important comments in the
> Jira trail, and before committing the change (or asking you to), I'd
> like to make sure you're in agreement with it.
> 
> Thanks!
> -Fred
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
> 
> 

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to