>>>> "Jan" == Jan Vroonhof
<vroonhof(a)math.ethz.ch> writes:
Jan> "Stephen J. Turnbull" <turnbull(a)sk.tsukuba.ac.jp> writes:
> Changing `face-spec-set' is the safer, more conservative
> approach. I don't like the third idea of saving 'default for
> reasons discussed in a separate post.
Jan> [This is all 21.0 talk, what would be more proper long term
Jan> solution is not the question now]
Agreed. "Safer" is still relevant to the current release. No?
Especially since Per points out that face-spec-set is one of the
recommended face APIs for Gnus. Things exposed to user-land ought to
be safe where the extra power is unnecessary; those calling
`reset-face' should know what they're doing.
Jan> Personally I do not care where one changes the code but what
Jan> do want to propose replacing the (reset-face 'default) in
Jan> face-spec-set in. A no-op (either by not calling reset-face
Jan> or having reset-face be a no-op on default)?
My current preference is to wrap the `reset-face' as follows (I'm not
sure at the moment what the right equality test is):
(unless (equal face 'default) (reset-face face ...))
Jan> I think that is wrong.
Your example is correct. Too bad. Custom is inadequate here, because
you can't win. Your suggested default-saving breaks any later changes
the user makes to 'default. I gave an example of how it could break.
Of course Custom basically says "faces are mine, don't touch," and we
could document that and ignore it for 21.0, except that Mule users
have no choice but to use non-Custom interfaces like `set-face-font'.
If you want to implement default-saving in Custom for 21.0, OK. I
myself don't want to hack up a temporary fix that both Hrvoje and Per
are likely to have reservations about. Custom already has mechanisms
for saving existing settings; it already uses them and you will have
to make sure that this default-saving behavior does not conflict with
them. Not worth it, IMO.
So this is more "post-21.0" "what should Custom really be" talk, IMO.
--
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091
__________________________________________________________________________
__________________________________________________________________________
What are those two straight lines for? "Free software rules."