[racket-dev] Error message structure (error-message overhaul)

From: Neil Van Dyke (neil at neilvandyke.org)
Date: Wed Jun 20 13:34:17 EDT 2012

Eli Barzilay wrote at 06/20/2012 01:07 PM:
>> Performance-wise, for exceptions involving paths, if resolving a
>> complete path happens to be expensive...
> (One of the nice things about errors is that performance is usually
> not an issue...)

But sometimes is, such as doing something performance-intensive a raised 
exception isn't unusual, by design (and the library routine raising the 
exception cannot be changed to not raise an exception, such as some 
filesystem routines).  The question is whether doing it the more 
complicated way is gives much better performance than the easy way.

I don't care much about this right now.  Just mentioning it in case now 
is a good time to be mentioning it.

Neil V.

Posted on the dev mailing list.