[PATCH] A compromise on case - another suggestion

[Will Ware]

Simon Brunning (sbrunning at trisystems.co.uk) wrote:
> [regarding Nick M's proposal for error messages]
> I envision this being enabled by default, to help newbies like
> myself. Expert users should be able to disable the check via a
> command line option if they wish.

Is there any performance hit to the clearer error messages? I would think
(not having studied the patch) that, for code with no errors in it, the
penalty would be only for testing the error, not for how it would be
processed if it did happen. If that's right, there's no incremental
performance cost for the improved error messages. Assuming experts don't
have such brittle egos as to find the extra information offensive, it'd
make sense to enable them all the time.

If it's correct at all, the no-additional-performance-hit reasoning should
also apply to people who want to hook in fancier near-miss-detection
functions (swapping letters, leaving a letter out, etc). Python could
be the world's first dyslexia-friendly programming language. Cool.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com