Junio C Hamano <gitster@xxxxxxxxx> writes: > As you are, I am on the fence. > > I do not think :(icase) pathspec is something we want to optimize > for, but I still like this new hash function primarily because I > suspect that it will increase the number of paths that you can cram > into the filter without getting their hashes collided (hence getting > false positive), under the assumption that real projects won't try > to store too many pair of paths that are only different in their > case... Sorry, but no, I do not think there is such upside. It may have effects on the actual hash values to downcase paths that are originally camelCased, but reducing the entropy of input paths that way shouldn't have effect on the overall distribution and rate of collision in any meaningful way (otherwise the chosen underlying hash function would be broken). So, sorry for the noise.
