Move normalized_edit_distance function to distance.c

This commit is contained in:
Dimitri Lozeve 2020-06-11 19:31:22 +02:00
parent 3f074d2bf9
commit c71e9b2cd7
3 changed files with 15 additions and 11 deletions

View file

@ -12,3 +12,15 @@ unsigned int hamming(size_t len, const unsigned char s1[static len],
} }
return res; return res;
} }
double normalized_edit_distance(
unsigned int block_size, unsigned int blocks_count,
const unsigned char buf[static block_size * blocks_count]) {
double avg_dist = 0;
for (size_t i = 0; i < blocks_count - 1; ++i) {
avg_dist +=
hamming(block_size, &buf[i * block_size], &buf[(i + 1) * block_size]);
}
return avg_dist / (double)block_size / ((double)blocks_count - 1.0);
}

View file

@ -5,5 +5,8 @@
unsigned int hamming(size_t len, const unsigned char s1[static len], unsigned int hamming(size_t len, const unsigned char s1[static len],
const unsigned char s2[static len]); const unsigned char s2[static len]);
double normalized_edit_distance(
unsigned int block_size, unsigned int blocks_count,
const unsigned char buf[static block_size * blocks_count]);
#endif /* DISTANCE_H */ #endif /* DISTANCE_H */

View file

@ -8,17 +8,6 @@
#define BUF_SIZE 4096 #define BUF_SIZE 4096
double normalized_edit_distance(
unsigned int keysize, unsigned int blocks_count,
const unsigned char buf[static keysize * blocks_count]) {
double avg_dist = 0;
for (size_t i = 0; i < blocks_count - 1; ++i) {
avg_dist += hamming(keysize, &buf[i * keysize], &buf[(i + 1) * keysize]);
}
return avg_dist / (double)keysize / ((double)blocks_count - 1.0);
}
int main(int argc, char *argv[]) { int main(int argc, char *argv[]) {
if (argc < 2) { if (argc < 2) {
printf("Usage: %s <filename>\n", argv[0]); printf("Usage: %s <filename>\n", argv[0]);