John E. Malmberg wrote:
Initial results on 25308 blead on OpenVMS/Alpha.
EAGLE> search blead.log failed
t/op/goto.................................FAILED at test 53
Appears to be known issue with VMS adding extra new lines to piped
output.
t/op/gv...................................FAILED at test 67
Appears to be passing a multiple line Perl program on the command line,
which VMS does not appear to be currently able to handle.
ext/Compress/Zlib/t/03examples............FAILED at test 10
ext/Compress/Zlib/t/06gzdopen.............FAILED--unexpected output at
test 0
New tests to VMS, investigation needed.
lib/Archive/Tar/t/02_methods..............FAILED--unexpected output at
test 68
New test to VMS, preliminary investigation shows a pathname that is
longer than Perl on VMS can handle.
lib/ExtUtils/ParseXS/t/basic..............FAILED--unexpected output at
test 3
VMS specific code needs to be written for the PERL module being tested.
Needs option file and fixed compile/link syntax.
lib/ExtUtils/t/basic......................FAILED at test 67
This test fails on VMS if the logical name "BIN" is defined and the DECC
feature DECC$UNIX_PATH_BEFORE_LOGNAME is not not also enabled.
The GNV product defines the BIN logical name in the system table, so
there is no way to locally undefine it by non-privileged users.
lib/ExtUtils/t/FIRST_MAKEFILE.............FAILED at test 4
Patch submitted to MAKEMAKER.
lib/ExtUtils/t/PL_FILES...................FAILED at test 3
Patch submitted to MAKEMAKER.
lib/File/Copy.............................FAILED at test 20
This test will only pass now on VMS if three conditions exist.
1. The test directory is an ODS-5 volume.
2. The hardware is not a VAX.
3. The DECC feature DECC$EFS_FILE_TIMESTAMPS is enabled.
lib/Math/BigInt/t/mbi_rand................FAILED--unexpected output at
test 60
Known restriction on VMS with a work around. Real fix is to find out
how to fix the extra new lines being added to piped output.
lib/Test/Simple/t/create..................FAILED at test 7
Patch for create.t submitted to me to test by Michael G. Schwern
seems to fix.
Failed 16 test scripts out of 921
Now down to 4 scripts failing for unknown reasons, 2 for ZLIB, 1 for
TAR, and one for ParseXS.
-John
[EMAIL PROTECTED]
Personal Opinion Only
