On Tue, 3 Sep 2002 12:43:18 -0400 (EDT), James Olin Oden <[EMAIL PROTECTED]> wrote:
>Right. I was thinking about that, also. I am not sure what one should >do in that case. I almost want to say assume the package being installed >is newest...almost. Unfortunately one is not always installing a package. For example a (semi-)automatic update tool might just be checking if a particular version is newer. > That problem gets even worse when you allow arbitrary >functions that compare versions. I would guess that the problem would >not occur fairly often, though. When this occurs, perhaps --force would be >the appropriate thing to do. Well, --force (or --oldpackage, which I think is not always implied by --force for some reason) is always a fix for the problem, but a solution that gets away from this would be nice. >On the other hand if tommorow Perl went to >an integer versioning scheme you could have a real mess...sigh. I think translation of version numbers is the best option, but to do it for perl stuff would require Redhat to agree with me, or a lot of work on my part. Ganesh _______________________________________________ Redhat-devel-list mailing list [EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-devel-list
