On Mon, Aug 20, 2007 at 05:31:00PM -0400, Jay Loden wrote: > Robert Dailey wrote: > > Well, I don't know what is wrong with people then. I don't see how > > required arguments are of bad design. Some command-line applications are > > built around performing tasks based on information received. Compilers, > > for example. A compiler can't do much of anything unless you give it at > > the very least a filename. So, a --file command would most definitely be > > one required argument. Anyway, I'm not trying to start a debate on this > > issue. I have my own implementation for required arguments at the > > moment, I am just a little bit surprised that this module doesn't make > > it convenient. It would definitely help on code duplication. > > > > Thanks for your response. >
> I tend to agree...while "required option" may be an oxymoron in > English, I can think of quite a few scripts I've written myself (in > various languages) that needed at least some kind of user input to > operate. At least the documentation points to some examples for > helpful hints, example code is a lot better than nothing ;) The idea with optparse is not that programs should not require certain information on the command line; rather, the idea is that this information should be positional arguments, not 'options'. That is, to use the compiler example: compiler file is preferred if a file argument is necessary. compiler --file file is not preferred. -- Due to some violent content, viewer discretion is advised. -- http://mail.python.org/mailman/listinfo/python-list
