[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: featurep instead of bound tests
From: |
Dan Nicolaescu |
Subject: |
Re: featurep instead of bound tests |
Date: |
Thu, 24 Jan 2008 21:49:04 -0800 |
Stefan Monnier <address@hidden> writes:
> >> --- gnus/ChangeLog 24 Jan 2008 07:47:37 -0000 1.633
> >> +++ gnus/ChangeLog 24 Jan 2008 15:14:50 -0000 1.634
> >> @@ -1,3 +1,9 @@
> >> +2008-01-24 Dan Nicolaescu <address@hidden>
> >> +
> >> + * sieve.el (sieve-make-overlay, sieve-overlay-put,
sieve-overlays-at):
> >> + * message.el (message-beginning-of-line): Use featurep instead
of bound
> >> + tests in order to resolve conditionals at compile time.
>
> > In the past, I've been told that it's preferable to use fbound tests
> > instead of Emacs flavor or version tests. So I'm surprised by changes
> > like these...
>
> Agreed. I see 2 better solutions:
> 1 - use (require 'outline)
> 2 - teach the byte-optimizer that (fboundp 'make-overlay) will always be t.
That has the problem that XEmacs would have to teach its compiler that
it is always false. Or if the test was (fboundp 'make-extent) emacs will
have to know it's always false.
IMO for extents vs overlays and zmacs-* the featurep test is the best
option.