completion functions clean up

completion functions clean up

[Oliver Kiddle]

I decided to go through the completion functions and do a bit of a
clean up. Some of this is little more than adjustments so that more
stuff conforms to the conventions I recently documented in
Etc/completion-style-guide but in many cases I made improvements, fixes
or brought it up-to-date with the latest command version.

Can _archie be removed? Archie servers don't seem to exist anymore.

I'll move _global_tags to the Unix/Type directory. Should fakeroot
perhaps be under the Debian directory - though useable elsewhere in
theory, it never is as far as I can tell.

I've not included the patch because it would be big and not very
interesting. You can always check changes in CVS. The updated files
are:

Base/_nothing                Redhat/_rpm
Cygwin/_cygpath              Unix/_apm
Cygwin/_cygrunsrv            Unix/_arping
Cygwin/_getclip              Unix/_cvs
Cygwin/_getfacl              Unix/_fakeroot
Cygwin/_mkshortcut           Unix/_fsh
Cygwin/_pscp                 Unix/_global
Cygwin/_putclip              Unix/_global_tags
Debian/_auto-apt             Unix/_imagemagick
Debian/_bug                  Unix/_irssi
Debian/_debchange            Unix/_mysql_utils
Debian/_debfoster            Unix/_pbm
Debian/_debsign              Unix/_rar
Debian/_dpkg                 Unix/_rsync
Debian/_dpkg_source          Unix/_slrn
Debian/_dput                 Unix/_user_admin
Debian/_dupload              Unix/_vorbis
Debian/_update-alternatives  X/_acroread
Linux/_iptables              X/_x_color
Linux/_pkgtool

I'll apply the majority of this to 4.0 too.

There's still more along these lines that can be done. If you know of
any other functions that could do with some attention, let me know.

Oliver

Re: completion functions clean up

[Clint Adams]

> I'll move _global_tags to the Unix/Type directory. Should fakeroot
> perhaps be under the Debian directory - though useable elsewhere in
> theory, it never is as far as I can tell.

I know people who use it on Solaris; also there is a fakeroot on the
Hurd (which is a different design and implementation). . . although,
since there is not yet a non-Debian Hurd, this is currently irrelevant.