Re: [Bug-AUCTeX] Re: 2005-05-02; environment insertion on active region unpractical

[Ralf Angeli]

* David Kastrup (2005-05-04) writes:

> Ralf Angeli <address@hidden> writes:

>From the headers of my previous message:

Received: from localhost ([127.0.0.1] helo=lists.gnu.org)
        by lists.gnu.org with esmtp (Exim 4.43)
        id 1DTLEU-0001ON-D9
        for address@hidden; Wed, 04 May 2005 10:52:14 -0400
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
        id 1DSvzC-0007ut-TZ
        for address@hidden; Tue, 03 May 2005 07:54:48 -0400

Boy, that's awful.  More than a day ...

> I don't insist on that, but at least C-x C-x should jump between start
> and end of environment after insertion of an environment around the
> active region.  If point is at the end of the environment, putting the
> mark right at the end of the \begin{...} line would help facilitating
> adding environment arguments.  Although we don't have a similar help
> available when doing a "normal" C-c C-e, which should not, I think,
> change the mark or mark stack.

Thinking about this, I think your last suggestion -- leaving mark and
mark ring alone -- is the only sensible thing to do.  I don't think
anyone will expect the mark being placed at the start of the
environment after inserting it.  As a user I would rather expect it to
be where I left it, i.e. either at the start or the end of the
originally marked region, depending on it being marked from bottom to
top or the other way round.  That means we just have to make sure that
the functions for inserting environments don't do stuff like
`exchange-point-and-mark'.

For jumping to the start of an environment we have `C-M-a'.  Which is,
uh, not documented in the manual.  Great.  Maybe we should add a
chapter about commands for navigating in LaTeX, ConTeXt etc. files.

-- 
Ralf