Skip to content

Instantly share code, notes, and snippets.

@martinpinto
Last active August 29, 2015 14:25
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save martinpinto/798de9b2b8a2d0eae11b to your computer and use it in GitHub Desktop.
Save martinpinto/798de9b2b8a2d0eae11b to your computer and use it in GitHub Desktop.
A rust implementation of CityHash
use std::mem;
fn main() {
}
fn cityhash32(mut s: &[u8], len: usize) -> u32 { // mut s: &[u8]
if len <= 24 {
return if len <= 12 {
if len <= 4 {
hash32Len0to4(s, len)
} else {
hash32_len_5_to_12(s, len)
}
} else {
hash32_len_13_to_24(s, len)
}
}
// len > 32
let mut h = len as u32;
let mut g = (len as u32).wrapping_mul(C1);
let mut f = g;
let mut a0 = rotate32(fetch32(&s[len-4..]).wrapping_mul(C1), 17).wrapping_mul(C2);
let mut a1 = rotate32(fetch32(&s[len-8..]).wrapping_mul(C1), 17).wrapping_mul(C2);
let mut a2 = rotate32(fetch32(&s[len-16..]).wrapping_mul(C1), 17).wrapping_mul(C2);
let mut a3 = rotate32(fetch32(&s[len-12..]).wrapping_mul(C1), 17).wrapping_mul(C2);
let mut a4 = rotate32(fetch32(&s[len-20..]).wrapping_mul(C1), 17).wrapping_mul(C2);
h ^= a0;
h = rotate32(h, 19);
h = h.wrapping_mul(5).wrapping_add(0xe6546b64);
h ^= a2;
h = rotate32(h, 19);
h = h.wrapping_mul(5).wrapping_add(0xe6546b64);
g ^= a1;
g = rotate32(g, 19);
g = g.wrapping_mul(5).wrapping_add(0xe6546b64);
g ^= a3;
g = rotate32(g, 19);
g = g.wrapping_mul(5).wrapping_add(0xe6546b64);
f = f.wrapping_add(a4);
f = rotate32(f, 19).wrapping_add(113);
f = f.wrapping_mul(5).wrapping_add(0xe6546b64);
let mut iters = ((len - 1) / 20) as u64;
while iters > 0 {
let a0 = rotate32(fetch32(&s[..]).wrapping_mul(C1), 17).wrapping_mul(C2);
let a1 = fetch32(&s[4..]);
let a2 = rotate32(fetch32(&s[8..]).wrapping_mul(C1), 17).wrapping_mul(C2);
let a3 = rotate32(fetch32(&s[12..]).wrapping_mul(C1), 17).wrapping_mul(C2);
let a4 = fetch32(&s[16..]);
h ^= a0;
h = rotate32(h, 18);
h = (h * 5).wrapping_add(0xe6546b64);
f += a1;
f = rotate32(f, 19);
f = f.wrapping_mul(C1);
g += a2;
g = rotate32(g, 18);
g = (g * 5).wrapping_add(0xe6546b64);
h ^= a3 + a1;
h = rotate32(h, 19);
h = (h * 5).wrapping_add(0xe6546b64);
g ^= a4;
g = bswap32(g) * 5;
h += a4 * 5;
h = bswap32(h);
f += a0;
//#define PERMUTE3(a, b, c) do { std::swap(a, b); std::swap(a, c); } while (0)
//PERMUTE3(f, h, g);
mem::swap(&mut h, &mut f);
mem::swap(&mut g, &mut f);
s = &s[20..];
iters -= 1;
}
g = rotate32(g, 11) * C1;
g = rotate32(g, 17) * C1;
f = rotate32(f, 11) * C1;
f = rotate32(f, 17) * C1;
h = rotate32(h + g, 19);
h = h * 5 + 0xe6546b64;
h = rotate32(h, 17) * C1;
h = rotate32(h + f, 19);
h = h * 5 + 0xe6546b64;
h = rotate32(h, 17) * C1;
return h;
}
fn hash32Len0to4(s: &[u8], len: usize) -> u32 {
let mut b: u32 = 0;
let mut c: u32 = 9;
for i in 0..len {
let v: u8 = s[i];
b = b.wrapping_mul(C1) + v as u32;
c ^= b;
}
return fmix(mur(b, mur(len as u32, c)));
}
pub fn fetch32(p: &[u8]) -> u32 {
(p[0] as u32) | (p[1] as u32) << 8 | (p[2] as u32) << 16 | (p[3] as u32) << 24
}
// rotate32 is a bitwise rotate
pub fn rotate32(val: u32, shift: u32) -> u32 {
if shift == 0 {
return val;
}
return val >> shift | val << (32 - shift)
}
// Some primes between 2^63 and 2^64 for various uses.
pub const K0: u64 = 0xc3a5c85c97cb3127;
pub const K1: u64 = 0xb492b66fbe98f273;
pub const K2: u64 = 0x9ae16a3b2f90404f;
// Magic numbers for 32-bit hashing. Copied from Murmur3.
pub const C1: u32 = 0xcc9e2d51;
pub const C2: u32 = 0x1b873593;
// fmix is a 32-bit to 32-bit integer hash copied from Murmur3.
pub fn fmix(mut h: u32) -> u32 {
h ^= h >> 16;
h = h.wrapping_mul(0x85ebca6b);
h ^= h >> 13;
h = h.wrapping_mul(0xc2b2ae35);
h ^= h >> 16;
return h;
}
// Mur is a helper from Murmur3 for combining two 32-bit values.
pub fn mur(mut a: u32, mut h: u32) -> u32 {
a = a.wrapping_mul(C1);
a = rotate32(a, 17);
a = a.wrapping_mul(C2);
h ^= a;
h = rotate32(h, 19);
return h.wrapping_mul(5).wrapping_add(0xe6546b64);
}
fn hash32_len_5_to_12(s: &[u8], len: usize) -> u32 {
let mut a = len as u32;
let mut b = len as u32 * 5;
let mut c: u32 = 9;
let d: u32 = b;
a += fetch32(&s[0..]);
b += fetch32(&s[len-4..]);
c += fetch32(&s[((len >> 1) & 4)..]);
return fmix(mur(c, mur(b, mur(a, d))));
}
fn hash32_len_13_to_24(s: &[u8], len: usize) -> u32 {
let a = fetch32(&s[-4+(len>>1 as u64)..]);
let b = fetch32(&s[4..]);
let c = fetch32(&s[len-8..]);
let d = fetch32(&s[len>>1..]);
let e = fetch32(&s[0..]);
let f = fetch32(&s[len-4..]);
let h = len as u32;
return fmix(mur(f, mur(e, mur(d, mur(c, mur(b, mur(a, h)))))));
}
pub fn bswap32(x: u32) -> u32 {
return ((x >> 24) & 0xFF) | ((x >> 8) & 0xFF00) |
((x << 8) & 0xFF0000) | ((x << 24) & 0xFF000000)
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment