Re: status of the VMS port ?

[Dan Nicolaescu]

Richard Stallman <address@hidden> writes:

  >     RMS said at some point that someone is working on making emacs work on
  >     VMS.  Who is that person?  
  > 
  > Sorry, I do not remember.
  > 
  >     Unless someone comes forward and announces that he will work on getting
  >     the VMS in good shape, can we deprecate the port in 22.2 and remove it
  >     after that?
  > 
  > Please do not be in such a rush to delete things!

You read too much into that.  Let's just give VMS an equal opportunity
with the other platforms: to stay or go based on existence of users
+ willing maintainers.

Right now it's in a state of limbo: it does not build, but it's not dead
either...

  > Before we delete this, we ought to do more than just wait and see if
  > "someone comes forward".  We should make a real effort to try to find
  > the person.  One way that we should certainly try is to search the
  > emacs-devel archives for the past 5 years for "VMS".

I found this page:

http://www.emacswiki.org/cgi-bin/wiki/EmacsOnVMS


For your benefit the contents are below.  The updates mentioined there
are almost 4 years old...


EmacsOnVMS

   I'm currently porting GNU Emacs 21.2 to OpenVMS? 7.3-1, and
   summarizing progress in this space. When the work is done,
   only build tips will remain. --ttn

   Current status
          Rather than make everyone wait while I puzzle out the
          copyright status of old VMS-specific files, I've decided to
          expose the work-in-progress so others can play, too. Details
          are available on the Emacs for VMS homepage:

     * http://www.gnuvola.org/software/emacs-for-vms/

   What works (that is, reproducibly w/o too much manual tweaking)
          Configuration, "dump", file read/write, byte compilation,
          timers and unixoid select(2) emulation, multiple
          subprocesses (synch and asynch - implies dired, comint,
          compile, and so forth), colors on console, filename
          completion (with "magic" right square brace and slash key
          bindings), network operations via open-network-stream.
          Yes, that means M-x zone in color over ssh from a VMS
          box. "When the going gets weird, the weird turn pro." (HST)

   Release methodology
          The "public releases" will actually be in the form of checking
          things into the GNU CVS repositories, updating the
          Emacs for VMS page, and making the occasional announcement to
          gnu-emacs-sources.

   Commentary
          Feel free to ignore the lunatic ramblings in this section.

     * (2004-04-02) In debugging next-error (for M-x compile), I notice
       that var compilation-error-regexp-alist for Emacs
       19.28 used to support specification of a function to do further
       parsing, beyond that which func
       compilation-parse-errors would do. Dropping this support makes
       for a cleaner design, evidently. Parsing must be one
       of the least-loved branches of computer science.
     * (2004-05-01) I asked the GNU Autoconf maintainers for vms-branch
       write privs but the response was not encouraging, so
       I've decided to publish "Autoconf for VMS" independently until a
       time (if any) when merging maintenance efforts
       becomes desirable.
          +
       http://mail.gnu.org/archive/html/autoconf/2004-04/msg00131.html
          +
       http://mail.gnu.org/archive/html/automake/2004-04/msg00205.html
          + http://www.gnuvola.org/software/autoconf-for-vms/
     * (2005-03-22) There is a Bourne shell script PREP-ZIP available in
       the distribution directory that can be used to
       perform the first half of the bootstrap relatively painlessly (on
       unixoid hosts). Observant readers will notice a
       workaround for a methodology bug in the script.
     
____________________________________________________________________________________________________________________

   Field Reports

   Since there is no mailing list, this section is for those intrepid
   spirits wishing to share experiences and tips building
   Emacs for VMS.
     * Autoconf for VMS requires GNU m4, and has been tested with
   Richard Levitte's port of GNU m4 version 1.2. Newer
       versions may or may not work. There is now a directory "m4" w/
   source and an Alpha binary in the Autoconf for VMS
       distribution dir.