Re: More complete tmux support

[Frank Terbeck <ft@bewatermyfriend.org>]

Jasper Lievisse Adriaanse wrote:
> Hi,

Hey there!

> We've had this patch for a long time in OpenBSD ports, it's to complete tmux
> support.

Great! I've been meaning to take a look for missing and changed features
for ages... but. Laziness. :)

However, something seems wrong with the patch.

[...]
> Subject: [PATCH] Add more complete tmux support.
>
> by pea@openbsd.org
> ---
>  Completion/Unix/Command/_tmux | 1491 +++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 1491 insertions(+)
[...]
> diff --git a/Completion/Unix/Command/_tmux b/Completion/Unix/Command/_tmux
[...]
> +# --- Sub-command functions ---
> +# These *must* be called _tmux-*(); The description generation relies on
> +# them being names that way. *No* other functions may match that pattern.
> +# Other utility functions should be named __tmux-*() (see below).
> +#
> +# Another thing, the description generation needs, is handling of
> +# $tmux_describe: If that parameter is non-empty, the sub-command function
> +# should only print a description of the sub-command it handles and return
> +# immidiately after doing so.
> +#
> +# To add support for a new sub-command, you only have to add a new
> +# _tmux-<foo>() function below (preferably alphabetically sorted), that
> +# behaves like described above; and add a alias->command pair in the
> +# _tmux_aliasmap associative array above (if the comand in fact has an
> +# alias). The rest should just work[tm].
[...]

I don't understand why this comment is added here.

[...]
> +function _tmux-send-prefix() {
> +    [[ -n ${tmux_describe} ]] && print "Send the prefix key to a window" && return
> +    local -a args
> +    args=('-t[Choose a target pane]:panes:__tmux-panes')
> +    _arguments ${args}
> +}
[...]

There are multiple pre-existing function added as well.

[...]
> +_tmux
> +#compdef tmux
> +
> +# tmux <http://tmux.sf.net> completion for zsh <http://zsh.sf.net>.
> +#
> +# Configuration:
> +#

And then here the file starts all over again?


Would be great, if you could take a look at what went wrong here and
resend.


Regards, Frank

-- 
In protocol design, perfection has been reached not when there is
nothing left to add, but when there is nothing left to take away.
                                                  -- RFC 1925