The directory specification can include paths defined in
operating system environment variables.
but it doesn't say how to do that. Any ideas? e.g. are they
specified as $VAR, or %VAR%, or some other way?
--
Ben Pfaff
b...@cs.stanford.edu
http://benpfaff.org
CD 'c:\sales\data\july.sav'.
CD 'c:\sales\data\july'.
CD 'c:\sales\dqta\july'.
If c:\sales\data is a valid directory:
· The first CD command will ignore the filename july.sav and set the
working directory to c:\sales\data.
· If the subdirectory july exists, the second CD command will change
the working directory to c:\sales\data\july; otherwise, it will change
the working directory to c:\sales\data.
· The third CD command will fail if the dqta subdirectory doesn't
exist.
> I'm not quite sure what it is your asking, but does this help?
>
> CD 'c:\sales\data\july.sav'.
> CD 'c:\sales\data\july'.
> CD 'c:\sales\dqta\july'.
No.
Let me restate the question. The documentation for the CD
command in release 15 says:
The directory specification can include paths defined in
operating system environment variables.
I assume that this means that, if the environment contains a
variable named, say, DATADIR, I should be able to CD to it. If
so, then how do I write the command for that. Is it:
CD '%DATADIR%'.
or
CD '$DATADIR'.
(to take a couple of wild guesses) or some other syntax entirely?
The documentation does not say.
--
"The road to hell is paved with convenient shortcuts."
--Peter da Silva
OK. That is because we support platform conventions. But I think
that $DATADIR works everywhere.
Correction: it is the platform convention only. $ does not work on
Windows.