Re: [PATCH] advapi32: succeed for null pointer in ChangeServiceConfig2

[Bernhard Übelacker]

Hello Nikolay,
thanks for your review.

Am 19.03.2015 um 13:34 schrieb Nikolay Sivov:
> In this case it's better to return earlier, no need to step into 
> try..expect block if you know you won't call anything.

So would it be safe to move the assignments out of the try..except
block (please see attached patch)?
> Is it actually called with infolevel 1?

Yes, when the error happened it was called with
SERVICE_CONFIG_DESCRIPTION, but in my patch I missed
that I just wanted to early return in that case.

Or should we just do the call to the RPC server and decide there?

Kind regards,
Bernhard