Blind Indexing

A blind index is calculated by using hash functions and/or key-stretching algorithms of the plaintext, using an appropriate key.

Blind indexes come in two modes (slow, fast). Both variants accept a parameter called the "bit length", which affects the truncation length of the resulting cryptographic output. Additionally, if truncated to a bit length that is not an even multiple of 8, it will clear the least significant bits in the trailing byte.