On Mon, Nov 05, 2007 at 12:39:23PM +0000, Pierre Habouzit wrote: > I like the kind of code that I allow to write better (I tend to > dislike big fat global variables), though it's obvious that Johannes > patch is a lot simpler and I like that. We discussed it further, and what came out is that instead of supporting quite complicated recursion mechanisms (or even a non so complicated one), we can just define an OPT__DIFFOPTIONS(diffopts)/OPT__REVOPTIONS(rev) macro that would inline the needed options. That's an idea I had but dismissed. Though, maybe it's not _that_ ugly, is clearly simpler, and one can argue that it's in the logical continuation of OPT__QUIET and friends. What do you think ? If it's the road we decide to take, then my documentation patch (2/4), the parseopt fix (my 1/4 or johannes 1/3) and Johannes usage generator enhancement (his 3/3) are still to be taken. -- ·O· Pierre Habouzit ··O madcoder@xxxxxxxxxx OOO http://www.madism.org
Attachment:
pgp5qVqIXYbfr.pgp
Description: PGP signature
