Craig A. Berry wrote:
On Oct 23, 2008, at 7:35 PM, John E. Malmberg wrote:
Craig A. Berry wrote:
At
first glance the readdir part doesn't look right; there shouldn't need
to be another completely different way of setting up the macros for
one extension, but I don't know offhand why the existing macros would
not be visible in File::Glob.
According to the comments in iperlsys.h, it is the place that the
macros need to be in order to make the wrappers visible to Perl.
I don't think there's any reason for the opendir/readdir part of the
patch. Running ext/File/Glob/t/basic.t under the debugger and doing a
SET TRACE/CALL/NOSYS shows definitively that it's already using the
routines in vms.c:
%DEBUG-I-DYNIMGSET, setting image PL_FILE__GLOB
trace on calls at BSD_GLOB\g_opendir\%LINE 73519+88 in THREAD 1
73519: return(PerlDir_open(buf));
%DEBUG-I-DYNIMGSET, setting image DBGPERLSHR
trace on calls at VMS\Perl_opendir\%LINE 120525+16 in THREAD 1
120525: Newx(dir, VMS_MAXRSS, char);
Is that with a threaded build?
When I stepped into the PerlDir_open() in PL_FILE__GLOB, using a
threaded Perl, it appeared to step into the a VMS threaded shared image
instead of DBGPERLSHR.EXE
-John
[EMAIL PROTECTED]
Personal Opinion Only
