Struct BloomTokenLog

pub struct BloomTokenLog(/* private fields */);

Bloom filter-based TokenLog

Parameterizable over an approximate maximum number of bytes to allocate. Starts out by storing used tokens in a hash set. Once the hash set becomes too large, converts it to a bloom filter. This achieves a memory profile of linear growth with an upper bound.

Divides time into periods based on lifetime and stores two filters at any given moment, for each of the two periods currently non-expired tokens could expire in. As such, turns over filters as time goes on to avoid bloom filter false positive rate increasing infinitely over time.

Implementations

impl BloomTokenLog

pub fn new_expected_items(max_bytes: usize, expected_hits: u64) -> BloomTokenLog

Construct with an approximate maximum memory usage and expected number of validation token usages per expiration period

Calculates the optimal bloom filter k number automatically.

pub fn new(max_bytes: usize, k_num: u32) -> BloomTokenLog

Construct with an approximate maximum memory usage and a bloom filter k number

If choosing a custom k number, note that BloomTokenLog always maintains two filters between them and divides the allocation budget of max_bytes evenly between them. As such, each bloom filter will contain max_bytes * 4 bits.

Trait Implementations

impl Default for BloomTokenLog

Default to 20 MiB max memory consumption and expected one million hits

With the default validation token lifetime of 2 weeks, this corresponds to one token usage per 1.21 seconds.

fn default() -> BloomTokenLog

Returns the “default value” for a type.

impl TokenLog for BloomTokenLog

fn check_and_insert( &self, nonce: u128, issued: SystemTime, lifetime: Duration, ) -> Result<(), TokenReuseError>

Record that the token was used and, ideally, return a token reuse error if the token may have been already used previously