[EMAIL PROTECTED] wrote:
>> So far we've hit OSError, IOError, AttributeError and I think one or two 
>> others.  Needless to say, I am not amused.
> 
> Out of curiosity, what were the others?  I was under the impression that
> EnvironmentError would catch OSError and IOError, but I don't know what
> else is going wrong.

I wasn't using EnvironmentError before; just a comment.

>> As such, instead of continuing to find out (the hard way) all the 
>> possible failure cases for this apparently horribly buggy library 
>> function, I'd prefer to simply catch any exception and fallback.
> 
> Ok.  I'm just asking if there's a chance we'll surpress a legtimiate
> error.  If it's the case that any negative result out of this function
> means we should just use something else, then it's not really a problem.

That's why it was written the way it was originally, but I'm happy to 
change it since you still have concerns.

>> I'm just tired of having to workaround this function.
> 
> I hope you'll forgive me for asking, but why not remove it completely
> then?

In hopes that when it does get fixed properly by the Python folks, that 
it will also be the better way to get the active username.

Updated webrev:
http://cr.opensolaris.org/~swalker/pkg-4010-2/

-- 
Shawn Walker
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss

Reply via email to