On Thu, Oct 04, 2007 at 02:57:58PM +0000, Kristian Høgsberg wrote: > > On Thu, 2007-10-04 at 01:11 +0200, Pierre Habouzit wrote: > > On Wed, Oct 03, 2007 at 09:45:01PM +0000, Kristian Høgsberg wrote: > > > The option parser takes argc, argv, an array of struct option > > > and a usage string. Each of the struct option elements in the array > > > describes a valid option, its type and a pointer to the location where the > > > value is written. The entry point is parse_options(), which scans through > > > the given argv, and matches each option there against the list of valid > > > options. During the scan, argv is rewritten to only contain the > > > non-option command line arguments and the number of these is returned. > > > > if we are going in that direction (and I believe it's a good one), we > > should be sure that the model fits with other commands as well. And as I > > said on IRC, I believe the most "horrible" (as in complex) option parser > > in git is the one from git-grep. > > > > A migration of git-grep on that API should be tried first. If this > > works well enough, I believe that the rest of the git commands will be > > migrated easily enough. (with maybe small addition to parse-option.[hc] > > but the hardcore things should have been met with git-grep already I > > think). > > I'm not sure - we can go with the current proposal and add new options > types and probably the callback option type I suggested as we go. I > don't want to block builtin-commit on figuring out what the perfect > option parser should look like and what I sent out earlier work for > commit. I think the way you handled the strbuf rewrites worked pretty > well; extending and rewriting the API as you put it to use in more and > more places. We can do the same thing with parse_options(). Of course we can do that, or junio said that some people talked about popt some time ago. I understand that you don't want to block the git-commit work, but doing things right from the beginning is often a big win on the long term. I don't know popt, and I don't know if it has sufficient expressivity. For sure I don't like getopt_long APIs at all, so if popt is as cumbersome, rolling our own based on the current parse_options you propose is probably a good choice. -- ·O· Pierre Habouzit ··O madcoder@xxxxxxxxxx OOO http://www.madism.org
Attachment:
pgp441LiBgJnd.pgp
Description: PGP signature
