Added file size based rename detection optimization
Prior to this change, files that were very different in size (enough so that they could not have enough in common to be detected as renames) were still having their scores calculated. I added an optimization to skip such files. For example, if the rename detection threshold is 60%, the larger file is 200kb, and the smaller file is 50kb, the pair cannot be counted as a rename since they cannot possibly share 60% of their content in common. (200*.6=120, 120>50) Change-Id: Icd8315412d5de6292839778e7cea7fe6f061b0fc
This commit is contained in:
parent
a8b29afd82
commit
64b9458640
|
@ -205,6 +205,14 @@ private int buildMatrix(ProgressMonitor pm) throws IOException {
|
|||
//
|
||||
matrix = new long[srcs.size() * dsts.size()];
|
||||
|
||||
long[] srcSizes = new long[srcs.size()];
|
||||
long[] dstSizes = new long[dsts.size()];
|
||||
|
||||
// Init the size arrays to some value that indicates that we haven't
|
||||
// calculated the size yet. Since sizes cannot be negative, -1 will work
|
||||
Arrays.fill(srcSizes, -1);
|
||||
Arrays.fill(dstSizes, -1);
|
||||
|
||||
// Consider each pair of files, if the score is above the minimum
|
||||
// threshold we need record that scoring in the matrix so we can
|
||||
// later find the best matches.
|
||||
|
@ -231,6 +239,26 @@ private int buildMatrix(ProgressMonitor pm) throws IOException {
|
|||
continue;
|
||||
}
|
||||
|
||||
long srcSize = srcSizes[srcIdx];
|
||||
if (srcSize < 0) {
|
||||
srcSize = size(srcEnt.oldId.toObjectId());
|
||||
srcSizes[srcIdx] = srcSize;
|
||||
}
|
||||
|
||||
long dstSize = dstSizes[dstIdx];
|
||||
if (dstSize < 0) {
|
||||
dstSize = size(dstEnt.newId.toObjectId());
|
||||
dstSizes[dstIdx] = dstSize;
|
||||
}
|
||||
|
||||
long max = Math.max(srcSize, dstSize);
|
||||
long min = Math.min(srcSize, dstSize);
|
||||
if (min * 100 / max < renameScore) {
|
||||
// Cannot possibly match, as the file sizes are so different
|
||||
pm.update(1);
|
||||
continue;
|
||||
}
|
||||
|
||||
SimilarityIndex d = hash(dstEnt.newId.toObjectId());
|
||||
int score = s.score(d);
|
||||
|
||||
|
@ -259,6 +287,10 @@ private SimilarityIndex hash(ObjectId objectId) throws IOException {
|
|||
return r;
|
||||
}
|
||||
|
||||
private long size(ObjectId objectId) throws IOException {
|
||||
return repo.openObject(objectId).getSize();
|
||||
}
|
||||
|
||||
private static int score(long value) {
|
||||
return (int) (value >>> SCORE_SHIFT);
|
||||
}
|
||||
|
|
Loading…
Reference in New Issue