toys: Tue Feb 26 01:20:00 2008

Tue Feb 26 01:20:00 2008 is an exploration of words with ambiguous T9 (phone text interface) encodings. The interesting bit is the scaling issue... or lack of one: it handled a 230K wordlist, producing 10k result sets, in about 30 seconds... on an EEEpc. Completely naive implementation, no optimization (and thus no wasted optimization :-) I did use string.maketrans and string.translate but that was more because I'd used them before and they're actually a pretty natural fit for this task. (I did switch from lists to sets, but that wasn't performance-related either - the wordlist I was working from had a variety of entries that weren't unique when downcased, and it didn't occur to me to just use sort -fu on it.)