mleku
7480e9232c
Refactored the Evaluate function to utilize a new XOR helper for improved readability and modularity. The XOR function handles the bitwise operation, streamlining the bit difference calculation and supporting potential reuse in related computations.
55 lines
1.5 KiB
Go
55 lines
1.5 KiB
Go
package nilsimsa
|
|
|
|
import (
|
|
"fmt"
|
|
)
|
|
|
|
func Evaluate(a, b []byte) (bitDiffCount int, err error) {
|
|
// Check if the lengths of the strings are the same
|
|
if len(a) != len(b) {
|
|
return 0, fmt.Errorf("byte strings are of different lengths %d and %d", len(a), len(b))
|
|
}
|
|
if len(a) != 32 || len(b) != 32 {
|
|
return 0, fmt.Errorf("input nilsimsa hashes must be 32 bytes each, got %d and %d",
|
|
len(a), len(b))
|
|
}
|
|
|
|
// Count the differing bits
|
|
var c []byte
|
|
c, err = XOR(a, b)
|
|
for i := 0; i < len(a); i++ {
|
|
bitDiffCount += countBits(c[i])
|
|
}
|
|
return bitDiffCount, nil
|
|
}
|
|
|
|
// Count the number of `1` bits in a byte
|
|
func countBits(x byte) int {
|
|
count := 0
|
|
for x > 0 {
|
|
count += int(x & 1) // Add the last bit
|
|
x >>= 1 // Right shift the bits
|
|
}
|
|
return count
|
|
}
|
|
|
|
// XOR is a simple bitwise XOR on two nilsimsa hashes. This is used in the simple distance count
|
|
// in Evaluate but could also be used against multiple samples of text from the same author to
|
|
// create a metric of the variety of their text by XORing all of them together.
|
|
func XOR(a, b []byte) (c []byte, err error) {
|
|
// Check if the lengths of the strings are the same
|
|
if len(a) != len(b) {
|
|
return nil, fmt.Errorf("byte strings are of different lengths %d and %d", len(a), len(b))
|
|
}
|
|
if len(a) != 32 || len(b) != 32 {
|
|
return nil, fmt.Errorf("input nilsimsa hashes must be 32 bytes each, got %d and %d",
|
|
len(a), len(b))
|
|
}
|
|
c = make([]byte, 32)
|
|
for i := 0; i < len(a); i++ {
|
|
// XOR the bytes and count the number of `1` bits
|
|
c[i] = a[i] ^ b[i]
|
|
}
|
|
return
|
|
}
|