On Sat, Jun 11, 2005 at 06:13:51PM +0300, Nikodemus Siivola wrote:
> On Sat, 11 Jun 2005, Christophe Rhodes wrote:
> >> (if (fixnump r2)
> >>- r2
> >>- (reg-tn-encoding r2))
> >>+ r2
> >>+ (reg-tn-encoding r2))
> >(if (foo)
> > (bar)
> > (baz))
> >I'm afraid I don't know where that can come from.
> I did not try to apply the patch and check the results, but are
> you sure that's not just a combination of tabs and -/+? I've seen
> such strangeness before in diffs with source looking just right.
> ...which brings me to my actual point: the great detabification.
> I'd like to see it, how about others?
I'm for it.
As it happens, for a couple of weeks I have been running the
experiment of putting
;;;; -*- mode: Lisp; indent-tabs-mode: nil -*-
at the head of my actively edited local CL code. I encourage others to
try it who, like me, don't have a global knowledge of Emacs behavior
and aren't sure whether they'd be happy with INDENT-TABS-MODE=NIL. As
far as I can see, it gets rid of some small annoyances without adding
any new ones: not a big win, of course, but a small clear tidiness
The one drawback I see is that by touching most of the lines of the
code, detabification will tend to make it tricky to get "cvs annotate"
information. For all I know it might already be a solved problem to do
an ignore-whitespace version of "cvs annotate", but even if it isn't,
I would cheerfully trade friction at "cvs annotate" time for less
whitespace/indentation friction at diff/patch time.
If people generally like the idea, the detabification should be
trivial to do early in any release cycle.
William Harold Newman <william.newman@...>
PGP key fingerprint 85 CE 1C BA 79 8D 51 8C B9 25 FB EE E0 C3 E5 7C
Ubi saeva indignatio ulterius cor lacerare nequit. -- Jonathan Swift's epitaph