>Yes, transcode should definitely handle globbing.
>I think this is the way most commandline programs works and the users are used
>to it.
Technically speaking, it's the shell that handles globbing; when you
do something like "cat *", the shell expands "*" to all of the filesnames
in the directory, and what cat itself sees is "cat file1 file2 file3...".
If you did this with transcode, you'd end up with multiple parameters to
the -i option, which (under the current code) wouldn't be processed. I'm
not even sure it would be possible to handle that without having to write
our own replacement for getopt_long().
That aside, I'm not averse to wildcard support in the long run, to
the extent that its meaning is not ambiguous. For example, "-i *" seems
pretty clear to me--concatenate all files--but what about "-i * -p *"?
Do you concatenate the video and audio streams separately, or match each
audio stream up with a video stream? Both are valid interpretations, so
I'd suggest only allowing wildcards for -i.
In any case, this is not a 1.1.0 issue, particularly when you start
to think about having to match up stream parameters between each input
file. I think we ought to drop directory mode entirely for 1.1.0 and
reintroduce it after we get the import system working under NMS, which
should make things much easier.
--Andrew Church
[EMAIL PROTECTED]
http://achurch.org/
