162 lines
5.0 KiB
Go
162 lines
5.0 KiB
Go
|
package farm
|
||
|
|
||
|
import (
|
||
|
"encoding/binary"
|
||
|
"math/bits"
|
||
|
)
|
||
|
|
||
|
func shiftMix(val uint64) uint64 {
|
||
|
return val ^ (val >> 47)
|
||
|
}
|
||
|
|
||
|
func hashLen16(u, v uint64) uint64 {
|
||
|
return hash128to64(uint128{u, v})
|
||
|
}
|
||
|
|
||
|
func hashLen16Mul(u, v, mul uint64) uint64 {
|
||
|
// Murmur-inspired hashing.
|
||
|
a := (u ^ v) * mul
|
||
|
a ^= (a >> 47)
|
||
|
b := (v ^ a) * mul
|
||
|
b ^= (b >> 47)
|
||
|
b *= mul
|
||
|
return b
|
||
|
}
|
||
|
|
||
|
func hashLen0to16(s []byte) uint64 {
|
||
|
slen := uint64(len(s))
|
||
|
if slen >= 8 {
|
||
|
mul := k2 + slen*2
|
||
|
a := binary.LittleEndian.Uint64(s[0:0+8]) + k2
|
||
|
b := binary.LittleEndian.Uint64(s[int(slen-8) : int(slen-8)+8])
|
||
|
c := bits.RotateLeft64(b, -37)*mul + a
|
||
|
d := (bits.RotateLeft64(a, -25) + b) * mul
|
||
|
return hashLen16Mul(c, d, mul)
|
||
|
}
|
||
|
|
||
|
if slen >= 4 {
|
||
|
mul := k2 + slen*2
|
||
|
a := binary.LittleEndian.Uint32(s[0 : 0+4])
|
||
|
return hashLen16Mul(slen+(uint64(a)<<3), uint64(binary.LittleEndian.Uint32(s[int(slen-4):int(slen-4)+4])), mul)
|
||
|
}
|
||
|
if slen > 0 {
|
||
|
a := s[0]
|
||
|
b := s[slen>>1]
|
||
|
c := s[slen-1]
|
||
|
y := uint32(a) + (uint32(b) << 8)
|
||
|
z := uint32(slen) + (uint32(c) << 2)
|
||
|
return shiftMix(uint64(y)*k2^uint64(z)*k0) * k2
|
||
|
}
|
||
|
return k2
|
||
|
}
|
||
|
|
||
|
// This probably works well for 16-byte strings as well, but it may be overkill
|
||
|
// in that case.
|
||
|
func hashLen17to32(s []byte) uint64 {
|
||
|
slen := len(s)
|
||
|
mul := k2 + uint64(slen*2)
|
||
|
a := binary.LittleEndian.Uint64(s[0:0+8]) * k1
|
||
|
b := binary.LittleEndian.Uint64(s[8 : 8+8])
|
||
|
c := binary.LittleEndian.Uint64(s[slen-8:slen-8+8]) * mul
|
||
|
d := binary.LittleEndian.Uint64(s[slen-16:slen-16+8]) * k2
|
||
|
return hashLen16Mul(bits.RotateLeft64(a+b, -43)+bits.RotateLeft64(c, -30)+d, a+bits.RotateLeft64(b+k2, -18)+c, mul)
|
||
|
}
|
||
|
|
||
|
// Return a 16-byte hash for 48 bytes. Quick and dirty.
|
||
|
// Callers do best to use "random-looking" values for a and b.
|
||
|
func weakHashLen32WithSeedsWords(w, x, y, z, a, b uint64) (uint64, uint64) {
|
||
|
a += w
|
||
|
b = bits.RotateLeft64(b+a+z, -21)
|
||
|
c := a
|
||
|
a += x
|
||
|
a += y
|
||
|
b += bits.RotateLeft64(a, -44)
|
||
|
return a + z, b + c
|
||
|
}
|
||
|
|
||
|
// Return a 16-byte hash for s[0] ... s[31], a, and b. Quick and dirty.
|
||
|
func weakHashLen32WithSeeds(s []byte, a, b uint64) (uint64, uint64) {
|
||
|
return weakHashLen32WithSeedsWords(binary.LittleEndian.Uint64(s[0:0+8]),
|
||
|
binary.LittleEndian.Uint64(s[8:8+8]),
|
||
|
binary.LittleEndian.Uint64(s[16:16+8]),
|
||
|
binary.LittleEndian.Uint64(s[24:24+8]),
|
||
|
a,
|
||
|
b)
|
||
|
}
|
||
|
|
||
|
// Return an 8-byte hash for 33 to 64 bytes.
|
||
|
func hashLen33to64(s []byte) uint64 {
|
||
|
slen := len(s)
|
||
|
mul := k2 + uint64(slen)*2
|
||
|
a := binary.LittleEndian.Uint64(s[0:0+8]) * k2
|
||
|
b := binary.LittleEndian.Uint64(s[8 : 8+8])
|
||
|
c := binary.LittleEndian.Uint64(s[slen-8:slen-8+8]) * mul
|
||
|
d := binary.LittleEndian.Uint64(s[slen-16:slen-16+8]) * k2
|
||
|
y := bits.RotateLeft64(a+b, -43) + bits.RotateLeft64(c, -30) + d
|
||
|
z := hashLen16Mul(y, a+bits.RotateLeft64(b+k2, -18)+c, mul)
|
||
|
e := binary.LittleEndian.Uint64(s[16:16+8]) * mul
|
||
|
f := binary.LittleEndian.Uint64(s[24 : 24+8])
|
||
|
g := (y + binary.LittleEndian.Uint64(s[slen-32:slen-32+8])) * mul
|
||
|
h := (z + binary.LittleEndian.Uint64(s[slen-24:slen-24+8])) * mul
|
||
|
return hashLen16Mul(bits.RotateLeft64(e+f, -43)+bits.RotateLeft64(g, -30)+h, e+bits.RotateLeft64(f+a, -18)+g, mul)
|
||
|
}
|
||
|
|
||
|
func naHash64(s []byte) uint64 {
|
||
|
slen := len(s)
|
||
|
var seed uint64 = 81
|
||
|
if slen <= 32 {
|
||
|
if slen <= 16 {
|
||
|
return hashLen0to16(s)
|
||
|
}
|
||
|
return hashLen17to32(s)
|
||
|
}
|
||
|
if slen <= 64 {
|
||
|
return hashLen33to64(s)
|
||
|
}
|
||
|
// For strings over 64 bytes we loop.
|
||
|
// Internal state consists of 56 bytes: v, w, x, y, and z.
|
||
|
v := uint128{0, 0}
|
||
|
w := uint128{0, 0}
|
||
|
x := seed*k2 + binary.LittleEndian.Uint64(s[0:0+8])
|
||
|
y := seed*k1 + 113
|
||
|
z := shiftMix(y*k2+113) * k2
|
||
|
// Set end so that after the loop we have 1 to 64 bytes left to process.
|
||
|
endIdx := ((slen - 1) / 64) * 64
|
||
|
last64Idx := endIdx + ((slen - 1) & 63) - 63
|
||
|
last64 := s[last64Idx:]
|
||
|
for len(s) > 64 {
|
||
|
x = bits.RotateLeft64(x+y+v.lo+binary.LittleEndian.Uint64(s[8:8+8]), -37) * k1
|
||
|
y = bits.RotateLeft64(y+v.hi+binary.LittleEndian.Uint64(s[48:48+8]), -42) * k1
|
||
|
x ^= w.hi
|
||
|
y += v.lo + binary.LittleEndian.Uint64(s[40:40+8])
|
||
|
z = bits.RotateLeft64(z+w.lo, -33) * k1
|
||
|
v.lo, v.hi = weakHashLen32WithSeeds(s, v.hi*k1, x+w.lo)
|
||
|
w.lo, w.hi = weakHashLen32WithSeeds(s[32:], z+w.hi, y+binary.LittleEndian.Uint64(s[16:16+8]))
|
||
|
x, z = z, x
|
||
|
s = s[64:]
|
||
|
}
|
||
|
mul := k1 + ((z & 0xff) << 1)
|
||
|
// Make s point to the last 64 bytes of input.
|
||
|
s = last64
|
||
|
w.lo += (uint64(slen-1) & 63)
|
||
|
v.lo += w.lo
|
||
|
w.lo += v.lo
|
||
|
x = bits.RotateLeft64(x+y+v.lo+binary.LittleEndian.Uint64(s[8:8+8]), -37) * mul
|
||
|
y = bits.RotateLeft64(y+v.hi+binary.LittleEndian.Uint64(s[48:48+8]), -42) * mul
|
||
|
x ^= w.hi * 9
|
||
|
y += v.lo*9 + binary.LittleEndian.Uint64(s[40:40+8])
|
||
|
z = bits.RotateLeft64(z+w.lo, -33) * mul
|
||
|
v.lo, v.hi = weakHashLen32WithSeeds(s, v.hi*mul, x+w.lo)
|
||
|
w.lo, w.hi = weakHashLen32WithSeeds(s[32:], z+w.hi, y+binary.LittleEndian.Uint64(s[16:16+8]))
|
||
|
x, z = z, x
|
||
|
return hashLen16Mul(hashLen16Mul(v.lo, w.lo, mul)+shiftMix(y)*k0+z, hashLen16Mul(v.hi, w.hi, mul)+x, mul)
|
||
|
}
|
||
|
|
||
|
func naHash64WithSeed(s []byte, seed uint64) uint64 {
|
||
|
return naHash64WithSeeds(s, k2, seed)
|
||
|
}
|
||
|
|
||
|
func naHash64WithSeeds(s []byte, seed0, seed1 uint64) uint64 {
|
||
|
return hashLen16(naHash64(s)-seed0, seed1)
|
||
|
}
|