Hi Guys, On Aug 24, 2007, at 5:55 PM, Craig A. Berry wrote:
Good catch. It should of course be C<=~> rather than C<=>. Somehow I've managed to be both slow and hasty in my patch review and application lately. I will fix this and the bizarre use of whitespace while I'm at it. John, in the future, please try to follow the use of spaces and tabs that the surrounding code uses.As far as whether this should be in the library rather than the test, let me explain what's happening and then you tell me. As I understand it, the test is trying to determine if C</disk1/foo/bar> is part of C</disk1/foo/bar/baz>. Except in VMS syntax, that would mean trying to determine whether C<disk1:[foo.bar]> is part of C<disk1:[foo.bar.baz]>. Because we have both a directory delimiter (dot) and a directory spec terminator (right bracket), we have to trim the right bracket from the first one to make it successfully match the second one. Since we're asserting the same truth -- that one path spec is the leading part of the other -- it seems to me ok to have this in the test only.
Attached is Archive::Extract 0.23_01, with the VMS fixes as discussed.
John, Craig if you could take a look and run this on VMS and tell me if
it's ok to promote this to 0.24, I'd be much obliged.
Cheers,
--
Jos Boumans
"Whenever you find you are on the side of the majority,
it is time to pause and reflect." - Mark Twain
Archive-Extract-0.23_01.tar.gz
Description: GNU Zip compressed data
