Yitzchak Scott-Thoennes wrote:
On Thu, Jan 19, 2006 at 04:19:04PM -0500, John E. Malmberg wrote:
Yitzchak Scott-Thoennes wrote:
Would you be able to try out the patch adding Module::Build?
Specifically I was wondering if it would fix the result of the
lib/Extutils/t/Manifest.t test.
Arrg. I'm sorry, I only just noticed I hijacked an existing thread.
I must have been asleep at the keyboard.
No, this has nothing to do with the Manifest.t failure, sorry.
Looking at the failing test, does VMS even have a concept of an
executable bit? perlport says:
C<-r>, C<-w>, C<-x>, and C<-o> tell whether the file is accessible,
which may not reflect UIC-based file protections. (VMS)
but I'm not sure what it means by accessible there.
VMS does have the concept of an executable bit. Generally read access
also grants execute access because it means that someone could just make
a copy. I do not know if the test for C<-x> takes that into account.
On OpenVMS, usually execute access is set for the default protection if
read or higher access is set in the default protection, so a test that
expects the execute protection to be off is likely to fail.
Execute access with out read access allows a program to be run, but no
copies to be made or displayed. This can not currently be done with
Perl scripts on OpenVMS. Doing so would require some changes in Perl
and require that Perl be installed as a trusted image with READALL
privilege to use to get around the file permissions blocking the script
from being read. An interesting concept, perl scripts that can be run,
but not viewed, copied, or debugged. I am not sure that there would be
much demand for that feature.
-John
[EMAIL PROTECTED]
Personal Opinion Only
