> > > > > > I keep going round and round in my head about this. I keep > > > wondering if there is a clean way to introduce this w/o breaking > > > templates. > > > > The easiest way would be to introduce new operators. > > -1 uh... no. please, no.
+1 on your -1 from me! :) > i'm not opposed to having some minor B.C. breaks in 1.5. if > we make clear public note of them, then we can just point > that out to those who don't pay attention and coming whining > to the user list. > > if people are relying on #set( $foo = 5 / 2 ) to make > "$foo" == '2', then chances are they're doing > not-very-proper-designer-ish things to begin with. even if > it's legit stuff, it means that they should be savy enough to > repair their templates when upgrading to 1.5 (assuming we > properly warn of this). We should always keep in mind though, that upgrading might be a hard thing to do if you have a lot of templates. For the number issue it would be good to have a switch like "Warn if changed numerical handling detected" and then add checks to the operators, that put a warning into the logfile, if there are false results (like 5/2). This would happen at runtime of course (since you cannot do this without a lot of false positives while parsing). OTOH I think that we should take the chance to eliminate all the "flaws" we find like whitespace handling (#* *# different than ##) and escaping. So we should only have ONE release that breaks something... But that's clear I think. Regards, Peter --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
