bug#1621: Please move cperl micro-docs to top of the cperl-mode.el file

["era eriksson" <era@iki.fi>]

On Sun, 15 Mar 2009 21:03:14 -0700, "Ilya Zakharevich"
<ilya@Math.Berkeley.EDU> said:
> On Mon, Mar 16, 2009 at 04:10:57AM +0200, era eriksson wrote:
> > Another thing I've been meaning to propose would be for cperl-mode 
> > to offer to install itself into your .emacs when you invoke it for 
> > the first time, sort of like how (put 'dangerous-command 'disabled 
> > nil) occasionally gets added to your .emacs after asking for 
> > permission.  (IIRC narrow-to-region is one of these "dangerous" 
> > commands, and some of the horizontal scrolling commands which I 
> > never use except by mistake.)
> 
> Myself, I would hate it if it were like this: supposed I'm annoyed by
> this "feature" - should it also ask whether to put a command into
> .emacs which would disable this question?  ;-)

The 'disabled logic is fairly unobtrusive IMHO.  My suggestion would be
to (defvar 'cperl-mode-install-setup-flag 'ask) and then either add
(setq 'cperl-mode-install-setup-flag nil) or 't to the user's .emacs if
the user doesn't specifically select "continue to ask me" instead of
"yes" or "no".

The main logic would simply inspect the value during load and do or do
not (there is no try :-) depending on what it says.

To my uneducated $self this sounds like it would need to go in a
separate .el file which would have some sort of ;;;###autoload but maybe
there is a more elegant way.

> > Do you think it would be appropriate to remove the +wontfix tag from
> > this bug?
> 
> I do not know what "the +wontfix" means.  The situation is as in "I
> considered your suggestions, I appreciated your suggestions, and I
> even mentioned possibly reconsidering your suggestions in my TODO
> files".  ;-)

I'm talking about the bug's status in the Emacs Bug Tracking System at
http://emacsbugs.donarmstrong.com/1621 -- if Emacs maintainers don't
object, I'll change it to "forwarded <ilya@Math.Berkeley.EDU>" and take
away the wontfix tag.  I hope it's not too far of a stretch to interpret
your answer to mean that you don't really care about the Emacs BTS
status of this bug report.

(The bug tracking system is an installation of the Debian BTS which by
design doesn't ever obfuscate email addresses.  I really hate this
"feature" because it's a regular fly paper for spam.  Maybe you would
prefer if I put something else than your email address in the
"forwarded" field, although I guess by and large the damage is already
done.)

/* era */

-- 
If this were a real .signature, it would suck less.  Well, maybe not.