A startup puzzle

Edward K. Ream edreamleo at charter.net
Mon Oct 6 17:27:40 EDT 2003


My apologies if this is a duplicate.  There was a glitch in sending this...

> Anyone else detect a whiff of irony?  "It's never caused me a problem!
> Well, except for the problem I just had, but that doesn't count."

The short answer is: "no, the problem doesn't count".

As I said elsewhere, I am rethinking importing *, and the reason has to do
with scalability.  At some point the problems of keeping track of the
globals (pollution, if you must) outweighs the convenience and clarity in
the code.  This is essentially a management issue unrelated to the technical
problem I originally posed.

Sure, not importing * solves that technical problem too, but so what?
Initialization is always tricky, but solving initialization problems really
should have little or no bearing on overall design questions.  So maybe I'll
eliminate * and maybe I won't.  For sure I won't be influenced by whether it
makes initialization easier.  And no doubt I'll keep the proxy class:  I
really like being able to write app.gui.x instead of app().gui.x.  Hmm.  If
I eliminate * I could just put a __call__ method in the leoApp class...

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edreamleo at charter.net
Leo: Literate Editor with Outlines
Leo: http://webpages.charter.net/edreamleo/front.html
--------------------------------------------------------------------






More information about the Python-list mailing list