Struct NamespaceHasher 

pub struct NamespaceHasher { /* private fields */ }

A utility struct for namespacing keys using XXH3 hashing.

This ensures that keys are uniquely identified within a given namespace, even if they share the same suffix. By hashing both the namespace and key separately before combining them, it prevents unintended collisions.

Example:

use simd_r_drive::utils::NamespaceHasher;

let hasher = NamespaceHasher::new(b"opt");
let namespaced_key = hasher.namespace(b"my_key");
assert_eq!(namespaced_key.len(), 16, "Namespaced key should be exactly 16 bytes");

Implementations

impl NamespaceHasher

pub fn new(prefix: &[u8]) -> Self

Creates a new NamespaceHasher with a given prefix.

The prefix itself is hashed using XXH3 to ensure a unique namespace identifier. This avoids collisions between different namespaces while keeping the hashing fast.

Arguments

• prefix: A byte slice representing the namespace prefix.

Returns

• A NamespaceHasher instance with a precomputed prefix hash.

pub fn namespace(&self, key: &[u8]) -> Vec<u8>

Computes a namespaced key, returning it as a 16-byte vector.

The final namespaced key is derived by:

1. Hashing the key separately to ensure uniqueness.
2. Combining it with the precomputed namespace hash.
3. Returning the concatenation of both hashes as a 16-byte key.

This ensures that:

• Different namespaces do not generate overlapping keys.
• Keys within a namespace remain uniquely identifiable.

Arguments

• key: A byte slice representing the key to be namespaced.

Returns

• A Vec<u8> containing the 16-byte namespaced key (8-byte prefix hash + 8-byte key hash).