On Thu, 25 Nov 2010 19:36:32 +0000, Nick Treleaven wrote: > On Thu, 25 Nov 2010 09:08:11 -0500, Jason House wrote: > >> I believe dmd looks for up to two errors when looking for a candidate >> match. I agree that for single letter variables that doesn't make a lot >> of sense. Even for two letter symbols it probably doesn't make much >> sense either. > > It's perhaps a minor quirk, but dmd could be changed to look for names > with a Levenshtein distance of MIN(2, variable_name.length).
Actually a distance of 1 is probably best for variable_name.length == 2 too, thus preventing 'i' for 'to'. > That would prevent 'to' being suggested for 'i', but could still match > 'ii' or 'j'.
