[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#35508: 27.0.50; Fine-ordering of functions on hooks
From: |
Eli Zaretskii |
Subject: |
bug#35508: 27.0.50; Fine-ordering of functions on hooks |
Date: |
Sat, 11 May 2019 16:54:42 +0300 |
> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Cc: 35508@debbugs.gnu.org
> Date: Sat, 11 May 2019 09:26:10 -0400
>
> >> Other objections?
> > Thanks. Should we perhaps change 100 to 110 and 90 to 100?
>
> You mean make it go https://en.wikipedia.org/wiki/Up_to_eleven ? ;-)
Yes. The idea is that 100 is easier to remember than either 90 or
110.
> > And perhaps not document the 110 value? Just a thought.
>
> I think we do want to document values greater than what `t` does: it can
> be important (e.g. for syntax-ppss-flush-cache) to make sure we stay
> closer to the end than those hooks appended via `t` for weaker reasons
> (e.g. because they don't want to be before some other function, although
> they don't really care if they're the very last one or not).
>
> Also I think it's important to use the same convention as for add-function.
We can always document that in comments.
> But what I wonder is whether we should enforce the convention: currently
> we don't in add-function (and in this add-hook patch), so you can use
> a depth of 8345 if you feel like: it's really just a convention.
>
> Also, maybe the docs should insist on the fact that 100/-100 should
> basically never be used since they imply that you're 100% sure that
> noone will ever need to come before/after you, and you can never be sure
> 100%.
We could have checkdoc complain about values we don't want to see in
application code.