@matheusbsbs / matheus-bagliano / commits / 1706306a54
Fork 0

diffcore-rename: similarity estimator fix.

The "similarity" logic was giving added material way too much negative weight. What we wanted to see was how similar the post-change image was compared to the pre-change image, so the natural definition of similarity is how much common things are there, relative to the post-change image's size. This simplifies things a lot. Signed-off-by: Junio C Hamano <junkio@cox.net>

Junio C Hamano committed Mar 2, 2006 at 22:11 UTC 1706306a54cfb5f1bf65f2b054aab2a5a7dba8e7
1 file changed +8 -12
diffcore-rename.c
+8 -12
@@ -170,19 +170,15 @@ static int estimate_similarity(struct diff_filespec *src,
170 &src_copied, &literal_added))
171 return 0;
172
173 - /* Extent of damage */
174 - if (src->size + literal_added < src_copied)
175 - delta_size = 0;
176 - else
177 - delta_size = (src->size - src_copied) + literal_added;
178 -
179 - /*
180 - * Now we will give some score to it. 100% edit gets 0 points
181 - * and 0% edit gets MAX_SCORE points.
173 + /* How similar are they?
174 + * what percentage of material in dst are from source?
175 */
183 - score = MAX_SCORE - (MAX_SCORE * delta_size / base_size);
184 - if (score < 0) return 0;
185 - if (MAX_SCORE < score) return MAX_SCORE;
176 + if (dst->size < src_copied)
177 + score = MAX_SCORE;
178 + else if (!dst->size)
179 + score = 0; /* should not happen */
180 + else
181 + score = src_copied * MAX_SCORE / dst->size;
182 return score;
183 }
184