On Wednesday 21 September 2011, 00:20:09, Casey McCann wrote:
This plays havoc with the search tree used internally by Set and Map, the result being that if you have any NaN values in the data structure, you may not be able to find other values anymore. Because NaN values never compare equal to themselves, I'm not sure if it's even possible to remove them from the structure,
filter (not . isNaN) resp. filterWithKey (\k _ -> not $ isNaN k)
and because of tree rebalancing I'm not sure how to predict what the impact of one or more NaNs would be over multiple operations on the data structure.
Yuck. Don't even try to predict that (unless you absolutely have to).
In short: Using Doubles in a Set, or as the key to a Map, should be regarded as a bug until proven otherwise (i.e., proving that NaN will never be inserted).
If you'd like to see an explicit demonstration (which you can try in GHCi yourself!) see here: http://stackoverflow.com/questions/6399648/what-happens-to-you-if-you-br eak-the-monad-laws/6399798#6399798 where I use it as an example of why it's important for type class instances to obey the relevant laws.
Nice and short.