Craig A. Berry wrote:
>>FWIW, I am using the generate PERL_SETUP.COM (slightly modified) to test the perl
>build without installing it.
>
>
> That should work, though you may have to define PERL_CORE in the environment
> for some tests. To run the whole test suite you can of course just do
> MM(S|K) TEST. The canonical way to run individual tests is
>
> $ @[.vms]test .exe "" -"v" [-.lib.File.Spec.t]rel2abs2rel.t
>
I just changed the following in PERL_SETUP.COM:
--- perl_setup.com;2 Tue May 28 15:43:58 2002
+++ perl_setup.com;1 Tue May 28 14:51:38 2002
@@ -2,7 +2,7 @@
$! Perl_Setup.com Tue May 28 13:50:46 UTC-5:00 2002
$! This perl configured & administered by [EMAIL PROTECTED]
$!
-$ define/translation=concealed perl_root DKA0:[USER.JPEACOCK.PERLSRC.PERL.]
+$ define/translation=concealed perl_root DKA0:[USER.JPEACOCK.PERLSRC.PERL_ROOT.]
$ ext = ".exe"
$ perl :== $perl_root:[000000]Perl'ext'
$ define PerlShr perl_root:[000000]PerlShr'ext'
and then ran it. That way I can run tests by just calling
perl harness [-.lib.File.Spec.t]rel2abs2rel.t
or
perl [-.lib.File.Spec.t]rel2abs2rel.t
so I am not using TEST.COM at all (since none of _my_ scripts will be using it
either). It still seems to me that safe_rel() should not choke on rooted logicals.
John
--
John Peacock
Director of Information Research and Technology
Rowman & Littlefield Publishing Group
4720 Boston Way
Lanham, MD 20706
301-459-3366 x.5010
fax 301-429-5747
