Arno Garrels wrote:
> Huch, I am Arno? I said that an attempt to send the mail even though 
> authentication failed were an easy task from user code. 
> Also after the EHLO-response has been received property 
> AuthTypesSupported holds a list of auth-types supported by the *server.
> If that list were empty you could skip the call to method Auth. The 
> high-level methods may be less flexible though.

I think we are agreeing there, or maybe I am misunderstanding you? I 
apologize if so, but I thought we both meant exactly the same thing: 
that it is easier and better to do this on the application side rather 
than in the component.  Is this not what you meant by "user code"?

But if Veit is using the high-level methods, then he that would be a 
problem, as you state above.  I keep forgetting about these, since I 
don't use them.

> However I won't do that,
> because if the user provides a password he wants to authenticate and thus
> should see an error if authentication failed.

I agree.  So perhaps the best solution for Veit's problem is to not use 
the "highest-level" method (the one that connects, authenticates and 
does all at once), so that he has time to not set the password (if 
that's what he wants to do); or else just fail the connection (as is 
happening) and let the user know that authentication is not supported.


To unsubscribe or change your settings for TWSocket mailing list
please goto
Visit our website at

Reply via email to