[Info-vax] Some questions on software for VMS 7.3 VAX

[lists at openmailbox.org]

On Thu, 21 Jan 2016 17:36:10 +0100
hb via Info-vax <info-vax at rbnsn.com> wrote:

> On 01/21/2016 04:58 PM, lists at openmailbox.org wrote:
> 
> > %DCL-W-ACTIMAGE, error activating image BLD_ROOT:[LOCAL.BIN]EMACS-21_2
> > -CLI-E-IMGNAME, image file DUA0:[REDACTED.][LOCAL.BIN]EMACS-21_2.EXE;
> > -RMS-E-DNF, directory not found
> > -SYSTEM-W-NOSUCHFILE, no such file
> 
> I apologize, my fault. I should have checked GNU_STARTUP.COM for
> BLD_ROOT and what it is set to.
> 
> > I guess one thing at a time. What should I do now? Thanks a lot for
> > helping with this. Sorry it took me awhile to get to this. Usually
> > whenever I have something interesting to try on my home systems I get
> > to work overtime that day. I will never know how they know...
> > 
> 
> (Re)-Define BLD_ROOT, the rooted logical. Assumed your current directory
> when restoring the saveset was DUA1:[USER] then
> 
> $ define BLD_ROOT /trans=(conc,term) DUA1:[USER.]
> 
> It seems, when BLD_ROOT is defined before GNU_STARTUP runs, the command
> procedure will not change it.
> 
> Or change GNU_STARTUP.COM. It seems in line 278 BLD_ROOT is defined.

What I did was to copy the gnu_startup.com to my own version and modify
that to try a few things. I did global changes on your filespec disk and
user directory to mine and then I changed [LOCAL. to [EMACS. to match my
setup after following your instructions about how to expand the backup
saveset to my desired directory. Thanks for your clear emails, they really
help. This gets me past having to set BLD_ROOT each time.

This got me further and Emacs did start (well, it started to start). But
then I got additional messages that various files couldn't be found. They
were still using the LOCAL directory. Apparently this is baked-in fairly
deeply. I don't know if it's just scripts that could be tracked-down or
whether it's also in code.

I guess I could keep trying to track down the scripts and rename them or I
could just live with the local directory name. The former seems like the
better solution since the directories are nested to max depth for VMS as I
understand it, which would mean I could never have a directory named local
wherever I decide to root this whole thing if I ever get it working and
can't hide it in a higher-level directory.

Falling back to using the baked-in directory name it gets further

[Spawning a new Kept EMACS]

but I get a message for various emulated terminals that they aren't
supported. For example coming in through Putty's telnet I get:

emacs: Terminal type vt102-132 is not defined.
If that is not the actual type of terminal you have,
use the Bourne shell command `TERM=... export TERM' (C-shell:
`setenv TERM ...') to specify the correct type. It may be necessary
to do `unset TERMCAP' (C-Shell: `unsetenv TERMCAP') as well.

Using xterm and telnet I get a similar message:

emacs: Terminal type vt100-132 is not defined.

So I used the default xterm size and it also failed similarly:

emacs: Terminal type vt400-80 is not defined.

Do you know how to resolve the terminal issue? Looks like it's fairly close
but there would still be the issue of how to make it globally available
since the directory structure and names are baked-in...

Thanks a lot for the help.

-- 
Please do not copy me on mailing list replies. I read the mailing list.
RSA 4096 fingerprint 7940 3F02 16D3 AFEE F2F8  ACAA 557C 4B36 98E4 4D49