bug#24229: 25.1; Can't set EPA program

[Daiki Ueno <ueno@gnu.org>]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Daiki Ueno <ueno@gnu.org>
>> Cc: nick@tenpoint.co.nz,  24229@debbugs.gnu.org
>> Date: Thu, 18 Aug 2016 09:14:27 +0200
>> 
>> Eli Zaretskii <eliz@gnu.org> writes:
>> 
>> >> So, we really need to take account of user's intention here; check
>> >> whether the variable set by user or it is the default value.  With
>> >> 'setq' it is not possible to have such distinction.
>> >
>> > Doesn't Customize provide some functions that allow such a
>> > distinction?  The Customize UI surely is able to tell when a variable
>> > was "changed not through Customize", isn't that what you want?
>> 
>> This is what we are currently doing, and the bug is saying that it
>> should be done without Customize.  It doesn't make sense to me though.
>
> What I meant is that Customize does have a way of telling that a value
> was changed using setq, so evidently there's some infrastructure for
> that which you could use.  Am I missing something?

Do you mean custom-variable-state or anything else?  The function just
compares a value with the default value.  In that case, we cannot
determine whether a user explicitly sets the variable to "gpg2", or the
default value "gpg2" is used.

If the former, we want to use "gpg2" binary anyway.  But if the latter
and "gpg2" is GnuPG 2.0, we want to try other versions when available.
This is not possible with that function.

Regards,
-- 
Daiki Ueno