> From: Alex Tweedly > > Can you please define the problem more precisely ? > > It seems to me there are 3 possible answers to come out from any > solution (or maybe more) > > 1 --> odd > 2 --> even > 2.5 --> neither odd nor even (or, if you prefer, "not whole") > > AFAICS, all the proposed solutions so far have been > incomplete, because > they failed to produce "neither" (i.e. they tested for > evenness, so if > not even then you'd have to assume it was odd).
I thought the original question stated the assumption that the number was whole. Normally, one tests to see if a number is an integer by comparing it to its integer part. In LC you just ask if it "is an integer". It's interesting how people approach things differently when they're using an interpreted scripting language. For instance, in C, one would never divide by two and then see if the remainder is nonzero, because a divide is a comparatively slow operation; one would use a bitwise AND, since that is typically a single-clock op. But in LC, that difference is swamped by the time spent processing the script. And in C, one would NEVER base the analysis on the printed representation, because that would involve generating the printed representation. I don't know how LC represents things, but I would think that it would represent values by the combination of a numeric representation and a string representation, either one of which (but not both) may be missing, deferring generation until it is needed. If that's the case, just asking for the last character of a number might very well cause the representation to be generated, which would be significantly costlier than doing arithmetic on it. -- Ciao, Paul D. DeRocco Paul mailto:pdero...@ix.netcom.com _______________________________________________ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
