On Tuesday 20 September 2011, 23:56:53, Evan Laforge wrote:
I actually think the brokenness of Ord for floating point values is worse in many ways, as demonstrated by the ability to insert a value into a Data.Set.Set and have other values "disappear" from the set as a result. Getting an unexpected element in a list doesn't really seem as bad as silently corrupting entire data structures.
Whoah, that's scary. What are some examples of this happening? Does this mean it's unsafe to store Doubles in a Map?
Too lazy to work out the details, but since NaN ? x = False for ? any of <, >, <=, >=, (==, /=) and compare is defined on terms of these, all results of compare involving a NaN are GT. member and insert in Data.set use compare to find out where to go, so inserting NaNs puts them at the max position. Insert a couple, and rebalancing can put one above non-NaN values, oops.