[Python-Dev] Re: Call for reviewer! (original) (raw)
Barry A. Warsaw bwarsaw@beopen.com
Wed, 16 Aug 2000 09:21:59 -0400 (EDT)
- Previous message: [Python-Dev] Re: Call for reviewer!
- Next message: [Python-Dev] RE: Call for reviewer!
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
I used to think getopt needed a lot of changes, but I'm not so sure anymore. getopt's current API works fine for me and I use it in all my scripts. However,
"DG" == David Goodger <dgoodger@bigfoot.com> writes:
DG> The incompatibility was introduced because the current
DG> getopt() returns an empty string as the optarg (second element
DG> of the tuple) for an argumentless option. I changed it to
DG> return None. Otherwise, it's impossible to differentiate
DG> between an argumentless option '-a' and an empty string
DG> argument '-a ""'. But I could rework it to remove the
DG> incompatibility.
I don't think that's necessary. In my own use, if I /know/ -a doesn't have an argument (because I didn't specify as "a:"), then I never check the optarg. And it's bad form for a flag to take an optional argument; it either does or it doesn't and you know that in advance.
-Barry
- Previous message: [Python-Dev] Re: Call for reviewer!
- Next message: [Python-Dev] RE: Call for reviewer!
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]