import hashlib


class BloomFilter:
    def __init__(self, size=2048, num_hashes=3):
        self.size = size
        self.num_hashes = num_hashes
        self.bits = bytearray(size // 8 + 1)

    def _hash_positions(self, item):
        h = hashlib.sha256(item.encode("utf-8")).digest()
        for i in range(self.num_hashes):
            val = int.from_bytes(h[i*4:(i+1)*4], "big") % self.size
            yield val

    def add(self, item):
        for pos in self._hash_positions(item):
            self.bits[pos // 8] |= 1 << (pos % 8)

    def might_contain(self, item):
        return all(
            bool(self.bits[pos // 8] & (1 << (pos % 8)))
            for pos in self._hash_positions(item)
        )

    @property
    def bytes(self):
        return bytes(self.bits)

    @classmethod
    def from_bytes(cls, data, size=2048, num_hashes=3):
        bf = cls(size=size, num_hashes=num_hashes)
        bf.bits = bytearray(data)
        return bf

    @staticmethod
    def from_items(items, size=2048, num_hashes=3):
        bf = BloomFilter(size=size, num_hashes=num_hashes)
        for item in items:
            bf.add(item)
        return bf

    @staticmethod
    def from_tags(tags, size=2048, num_hashes=3):
        bf = BloomFilter(size=size, num_hashes=num_hashes)
        for tag in tags:
            bf.add(tag.strip().lower())
        return bf