[ESS-bugs] ess-mode 5.2.4; ess-help "s e" fails for Xemacs

[Martin Maechler]

>>>>> "StEgl" == Stephen Eglen <S.J.Eglen at damtp.cam.ac.uk>
>>>>>     on Wed, 5 Jan 2005 13:36:59 +0000 writes:

    >> 
    >> No, unfortunately not `as easy'; the message is
    >> 
    >> Invalid section key: e
    >> ----------------------
    StEgl> ...

    >> The *-alist looks ok, well, at least identical to the one I see
    >> in GNU emacs :
    >> 

    >> Could it be that my version of Xemacs doesn't give correct
    >> values of `last-command-char' ?

    StEgl> Doing it with last-command-char does indeed seem odd. Checking the SVN
    StEgl> log file, I couldn't see any mention of this function being edited, so
    StEgl> does it mean that this use of last-command-char is an ancient (pre 
    StEgl> 97) trick?  We should be able to do it "right" without resorting to
    StEgl> such hackery I'd have thought.

Yes, I bet this is very old code; has been there in old versions
of S-mode AFAIR.

    StEgl> "e" should be code 101 (in decimal), and that's what I see in my
    StEgl> buffers.  You may need to use "edebug" on ess-skip-to-help-section and
    StEgl> see whether it really has code 101.

It says ?e only
(wheras the keys of ess-help-sec-keys-alist have 101).

I really don't know xemacs enough {and see again and again how
much I have forgotten about emacs-lisp};
trying
    (equal ?e 101)
gives nil in my Xemacs 
but 't' in my GNU emacs.

Given Tony's feedback, and 
a test on my debian-sid machine where I get xemacs 21.4.16,
and the same problem,
I think this must be a local problem.

Could it be that the    
      ?e <--> 100 equivalence
is (something like) locale/mule/language dependent and doesn't
work for me for that reason?

Well, in any case this doesn't seem like a global xemacs
problem and definitely not a show stopper for release of 5.2.4.

Martin