SimilarityRenameDetector: Avoid allocating source index
If the only file added is really small, and all of the deleted
files are really big, none of the permutations will match up due
to the sizes being too far apart to fit the current rename score.
Avoid allocating the really big deleted SimilarityIndex by deferring
its construction until at least one add along that row has a
reasonable chance of matching it.
This avoids expending a lot of CPU time looking at big deleted
binary files when a small modified text file was broken due to a
high percentage of changed lines.
Change-Id: I11ae37edb80a7be1eef8cc01d79412017c2fc075
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/diff/SimilarityRenameDetector.java b/org.eclipse.jgit/src/org/eclipse/jgit/diff/SimilarityRenameDetector.java
index bf1bbda..f47caf9 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/diff/SimilarityRenameDetector.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/diff/SimilarityRenameDetector.java
@@ -229,20 +229,14 @@ private int buildMatrix(ProgressMonitor pm) throws IOException {
// later find the best matches.
//
int mNext = 0;
- for (int srcIdx = 0; srcIdx < srcs.size(); srcIdx++) {
+ SRC: for (int srcIdx = 0; srcIdx < srcs.size(); srcIdx++) {
DiffEntry srcEnt = srcs.get(srcIdx);
if (!isFile(srcEnt.oldMode)) {
pm.update(dsts.size());
continue;
}
- SimilarityIndex s;
- try {
- s = hash(OLD, srcEnt);
- } catch (TableFullException tableFull) {
- tableOverflow = true;
- continue;
- }
+ SimilarityIndex s = null;
for (int dstIdx = 0; dstIdx < dsts.size(); dstIdx++) {
DiffEntry dstEnt = dsts.get(dstIdx);
@@ -282,6 +276,15 @@ private int buildMatrix(ProgressMonitor pm) throws IOException {
continue;
}
+ if (s == null) {
+ try {
+ s = hash(OLD, srcEnt);
+ } catch (TableFullException tableFull) {
+ tableOverflow = true;
+ continue SRC;
+ }
+ }
+
SimilarityIndex d;
try {
d = hash(NEW, dstEnt);
