At 11:28 AM -0500 1/3/03, [EMAIL PROTECTED] wrote: >OK here is the patch that could be applied to perl@18376 >to affect the change that I said I wanted. With it >applied (as well as the two test fixes that started this >discussion thread) I see only these `mmk test` failures: > >t/run/switchi........................FAILED at test 2 >ext/Devel/Peek/Peek..................FAILED at test 21
I have fixes for those two that just need to be posted. >lib/charnames........................FAILED at test 74 This one is a bit stranger and I haven't resolved it yet. >lib/File/Find/t/find.................FAILED at test 1 >lib/File/Find/t/taint................FAILED at test 1 >lib/Net/hostent......................FAILED at test 5 >lib/Net/Ping/t/450_service...........FAILED at test 13 >Failed 7 test scripts out of 671, 98.96% okay. I do not see any of those failures on OpenVMS Alpha 7.3-1, Compaq C 6.5, TCP/IP Services 5.3 ECO 1. The File::Find tests can be thrown off if there is a SYS logical name defined. The Net stuff sometimes depends on network services that may or may not be defined. >The patch to vms/vms.c and vms/perlvms.pod appears as: Looks good. >So the questions I have are: 1. Does anyone besides me want to >see this change in vmsperl? 2. If the answer to the previous >question is "yes" then which interface is preferred?: If it adds a capability we didn't have and doesn't break anything, go for it. I agree that adding a new switch is probably not a good idea, and even though the triple quoting is a bit ugly, it's really what anyone who knows how DCL works would expect. -- ________________________________________ Craig A. Berry mailto:[EMAIL PROTECTED] "... getting out of a sonnet is much more difficult than getting in." Brad Leithauser
