Re: master 695f679: Remove ; ; ; ###tramp-autoload cookie from Tramp defcustoms (Bug#47063)

[Michael Albinus]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

Hi Stefan,

>> Of course, we can live w/o that predicate. We did until now. But OTOH,
>> it doesn't hurt to check this, that's why this kind of check has been
>> added to Emacs.
>
> What I mean is that this is a predicate which should be most useful to
> those who never use `tramp-crypt` (e.g. because `encfs` is not
> installed), so only activating it when `tramp-crypt` is loaded makes it
> *much* less useful.

I don't want to activate the predicate before tramp-crypt.el is loaded,
because this would counteract the idea to keep tramp-crypt related code
in tramp-crypt.el.

> This predicate is only useful for those users who don't have
> `encfs` installed.  Which proportion of those users do you think has
> `tramp-crypt` loaded in their Emacs session?  My guess is that this is
> pretty close to 0%.

Well, it is also useful for users who have tried
tramp-crypt-add-directory, and see afterwards that it doesn't work for
them. But yes, this is a small number of users.

OTOH, it doesn't hurt. I don't see any price we have to pay for this
predicate, in terms of performance or whatsoever. People who don't use
tramp-crypt.el are not affected in any way.

And I have extended the predicate just now to be more useful for the
other command in tramp-crypt.el, tramp-crypt-remove-directory.

>         Stefan

Best regards, Michael.