[ESS-bugs] transient mark and noweb

[Martin Maechler]

Hi Kasper,

thank you for this.

>>>>> "KDH" == Kasper Daniel Hansen <khansen at stat.Berkeley.EDU>
>>>>>     on Fri, 20 Jul 2007 16:34:51 -0700 writes:

    KDH> Hi All I was getting really tired of the fact that
    KDH> transient mark mode does not play well with noweb-mode,
    KDH> so I did a google search and lo and behold a patch had
    KDH> been submitted to ess-bugs a while ago by Markus
    KDH> Triska. I don't know how to do a followup to such a
    KDH> report, hence this new email. The beginning of that
    KDH> mail is

was 'Feb 12' (to ESS-bugs;  I didn't know that google would find
    	      ESS-bugs archives!  interesting!)
ESS-core: Note that I have added the correct 'References:' header
 	  to this e-mail, so you should see Markus Triska's two
 	  messages with Stephen's reply in between, properly threaded...
	  in case you have a threaded e-mail view and the
	  messages in the same mailbox

    >> Hi ESS-Team,
    >> 
    >> first of all, thank you so much for ESS, it makes my work
    >> much easier.  Second, I recommend the following
    >> ess-write-to-dribble-buffer:

    >> (defun ess-write-to-dribble-buffer (text)
    >>   "Write TEXT to dribble buffer."
    >>   (unless (buffer-live-p ess-dribble-buffer)
    >>     ;; ESS dribble buffer must be re-created.
    >>     (setq ess-dribble-buffer (get-buffer-create "*ESS*")))
    >>   (let (deactivate-mark)
    >>     (with-current-buffer ess-dribble-buffer
    >>       (goto-char (point-max))
    >>       (insert text))))

    >> The most important change is that deactivate-mark is set
    >> to nil, so that the region, if active, isn't
    >> deactivated. Without this, in transient-mark-mode in GNU
    >> Emacs, highlighting of the region is lost when moving
    >> over chunks, and M-h (= mark-paragraph) on chunks doesn't
    >> work as expected.

Stephen had replied on Feb 23, saying that he would look at this
"over the weekend" and Markus furhter repied, mentioning a few
wrong nestings of safe-recursion / safe-restriction pairs.

It seems Stephen's weekend had been filled with more exciting
matters, and unfortunately, both very pertinent reports from
Markus are still TODO.

I'm trying (and then will commit) the first one now.


    KDH> I have applied this patch to ess.el and it works
    KDH> beautifully. I am using the latest version of Carbon
    KDH> Emacs under Mac OS X Tiger using a Power PC. The
    KDH> versions are GNU Emacs 22.1.1
    KDH> (powerpc-apple-darwin8.9.0, Carbon Version 1.6.0) of
    KDH> 2007-06-16 on g5.tokyo.stp.isas.jaxa.jp ESS version
    KDH> 5.3.4

    KDH> The posting (and replies) contain some other bug fixes,
    KDH> which I have not tried.

{indeed, I've mentioned them above}

    KDH> Could this at least go into the next ESS?

yes, it could ... ;-)
and it will!

Regards, Martin