[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gnumed-devel] address@hidden: Re: gmCfg dependancy on gmPG]
From: |
Karsten Hilbert |
Subject: |
Re: [Gnumed-devel] address@hidden: Re: gmCfg dependancy on gmPG] |
Date: |
Thu, 26 Feb 2004 14:58:16 +0100 |
User-agent: |
Mutt/1.3.22.1i |
> Well, one reason for having getFirstMatchingDBSet() was that you don't need
> to
> create a connection to the database first.
I realized that and added the lazy import code to those two
wrappers.
> This, in turn, is a requirement to initialize gmCfgSQL. So I would rather say
> no unless you auto-connect to the backend in gmCfgSQL.__init__().
I wonder exactly why I didn't do that in the first place.
Probably exactly because of trying to not depend on "import
gmPG". That needs cleaning up one way or the other IMO.
> I guess that's the only possibility, then. I personally feel that splitting
> gmCfg in gmCfgFile and cmCfgSQL would be cleaner than adding all these "lazy
> import" code, but that may be a matter of taste.
It's not that much code, really, but the half-breed
dependancy-or-not on an imported gmPG needs cleanup, I guess.
I'd say moving connection building into gmCfgSQL.__init__()
and do a lazy import in there is the way to go. Your wrappers
still make sense...
Karsten
--
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346