On Fri, Feb 20, 2015 at 4:13 PM, Jason Friedman <jsf80...@gmail.com> wrote: > os.chdir(a_path) > return_list = [os.path.join(a_path, x) for x in glob.glob(a_glob) if > os.path.isfile(x)] > os.chdir(previous_dir) > return reversed(sorted(return_list, key=os.path.getmtime)) > > It's a shame that glob.glob does not take an arbitrary directory as an > optional argument if one does not want to scan the current directory.
If you look at the glob module's docs, you'll see how it's implemented: https://docs.python.org/2/library/glob.html You may want to use fnmatch, or possibly just prepend a path name onto your glob. That way, you wouldn't need to change directory, which is almost always a bad idea for a library function. Instead of the reversed() call, you can simply add reverse=True to the sorted() call. And since you're not making use of the original list, you could use its sort() method (which has key and reverse parameters too) to sort a bit more efficiently. But I think your code is pretty good! ChrisA -- https://mail.python.org/mailman/listinfo/python-list
