On Thu, 8 Aug 2002, Stephen Turner wrote: > I thought when I first read the problems that the factorial would be much > shorter than the postorder
i thought that, too... > (like 35 vs 70 or something), and that it would > be the postorder that sorted people out. As it turned out, the factorial was > only just shorter, and it was the factorial that sorted people out. (In > fact, I was one of two players with postorder shorter than factorial. When > I told Ton this he said "That's because your factorial is so long". Ouch!) my problem was the 10**6 instead of 1e6. then my factorial and postorder would have been the same length. now it's so obvious... i still don't really get that thing with ~4, but i'll sleep over it... i like the postorder solutions, because there are quite efficient. the normal approach would be (what i did first, too) to write a recursive function, because trees somehow belong to "recursive". function(root) = root . function(child) . function(child2). but then i realized that the declaration of a sub just takes away too many space, and i remembered that a) every algorithm can be done non-recursive and b) perl regexes are more powerful than one might think. the regex solution might be a little bit slower than recursive but it needs less memory. anyway, glad that i made it to place 8, never expected that... =) looking forward to the next golf, tina -- http://www.tinita.de/ \ enter__| |__the___ _ _ ___ http://Movies.tinita.de/ \ / _` / _ \/ _ \ '_(_-< of http://PerlQuotes.tinita.de/ \ \ _,_\ __/\ __/_| /__/ perception
