>>970162
wew, this is the Kode with Klossy I expected.
>bloomlib.c
This is my favorite, it looks like homework from someone learning C at Jewschool. C datastructure for speed that uses doubles, calculates a mystery error value that shouldn't be exposed and isn't used internally, requires the array to be malloced, stores the count of bits separately from the count of bytes rather than round to a byte for who knows what reason, assertions attempting to catch bad values allow integer overflow, protection from a stack overflow in bf_get_index is only present in callers rather than the function itself, and supports loading and saving the filter to a file for no reason.
>bloom.c
Now that's just suspicious as it has many of the same design mistakes and structure as the one above and chose the same hash and was written in the same year. Bet you've found a class project.
>pajeetscript
didn't read.
>not counting murmur that everyone has to write
Why the fuck would you have to write, or use, murmur hash? A hash for this purpose is often as simple as modulus with a prime, and that's the normal state bloom filters are found in real code rather than massive overgeneralized data structures for homework. What hashes you use are a decision to be made by the user of the filter anyway, not the author of the filter. And murmur is obsolete (these were from 2012, though).