From: "dman" <[EMAIL PROTECTED]>
> On Mon, Jan 28, 2002 at 02:24:17PM -0000, Matt Sergeant wrote: > | I've been saying all along - in perl $VERSION is a > | number. So it does numeric comparison, just like I described. > > Sorry, I have trouble remembering which operator in perl is numeric > and which is string based. > > Given this clarification, I assert that CPAN is really broken because > version numbers aren't floating point numbers. Indeed, how would > "2.2.1" compare with "2.2.2"? It is illogical to have multiple > decimal points in a float literal, though perl silently ignores that. > In fact, trying it out for myself, neither is less than the other! I've been through this before, and indeed I used to think like you do. But like everything in life, there is more than one way to do it. While linux may do major.minor.patchlevel, as do many many applications including RCS/CVS and most version control systems, Perl on the other hand does not. It's simply a different way of doing it. Whether that "different" sucks in your opinion or not doesn't really matter a whole tin of beans[1] in the wide scheme of things, because in Perl we have CPAN, and CPAN treats $VERSION as a floating point number, and does comparisons that way, and there ain't no hell nor fire going to change the way CPAN works - there are nearly four THOUSAND modules on CPAN you would have to change to make sure they updated their scheme of working. Sorry, but if you want to play in this garden, you have to use their toys. Matt. [1] You think it sucks, I think it sucks, almost everyone thinks it sucks. _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
