Re: Initialisation of dired-x for install-info guessing

[Stefan Monnier]

> Well just to give some examples, I have downloaded the info file of
> Bash and that of Texinfo on my machine and I have installed them with
> install-info by hand. I also installed all the manuals that were
> coming with the MinGW tool-chain with install-info. I am also writing
> myself 3 manuals for emacs extension packages (one is for
> emacs-template which is still under discussion with Christoph Wedler,
> and the two other are for extension packages I am developping myself),
> and I had also to install them with install-info.

I must admit that I don't compile packages by hand very often nowadays,
but back when I did that, I very rarely used install-info since usually
the info files were installed by `make' instead.

> I recognize that it is not that often, but it happens...

Fair enough.  To tell you truth, since this rule is only for files that
end in ".info" it doesn't have to be frequent, it only has to be more
frequent than other commands that one might apply to .info files.
So `install-info' is probably a good candidate, indeed.

> There are a number of options to call install-info, like for any other
> commands. But this is also true for any of the commands that are
> invoked by means of the dired-x default shell guess.  I think that the
> shape 

>     install-info file.info dir

> is the most common... but you should cross check this with Karl Berry...

That was my impression as well.  So why is the suggested command using
the weird `for' loop (which will make it apply to all kinds of other
unrelated files).


        Stefan