Now the Bloom filter itself. Adding an item sets the k bits at its double-hash indices - that single move is the entire write path, and everything else reads from it.
Build a Bloom filter of m bits and k hashes whose Add sets the k bits for an item.
A Bloom filter is just the bit array from the last chapter plus the double-hash rule, wrapped together with its two parameters: m, the number of bits, and k, the number of hash positions per item. Add is the whole write side - compute the k indices for the item and set each of those bits. Nothing records which item was added, only that some item lit those bits, and that is exactly why the filter is so small.
Adding the same item twice is a no-op on the second try: the bits are already set, so setting them again changes nothing. That idempotence is worth pinning now because it is what lets many items share the array - each just contributes its handful of set bits, and collisions between items are expected rather than exceptional. The indices 7, 21, 30 are "cat"’s three positions from the previous lesson, now written into the filter.
type Bloom struct {bits *Bitsm, k int}func NewBloom(m, k int) *Bloom { return &Bloom{bits: NewBits(m), m: m, k: k} }func (f *Bloom) Add(data []byte) {// set each of the k double-hash indices}