Improve runtime of delta computation #553
Conversation
Codecov Report
@@ Coverage Diff @@
## master #553 +/- ##
============================================
+ Coverage 90.46% 90.48% +0.01%
- Complexity 1586 1590 +4
============================================
Files 177 177
Lines 4899 4919 +20
Branches 542 546 +4
============================================
+ Hits 4432 4451 +19
Misses 293 293
- Partials 174 175 +1
Continue to review full report at Codecov.
|
| referencesByHash = new HashMap(); | ||
| referencesByFingerprint = new HashMap(); | ||
|
|
||
| for (Issue issue : fixedIssues) { |
There was a problem hiding this comment.
More readable:
| for (Issue issue : fixedIssues) { | |
| for (Issue issue : referenceIssues) { |
There was a problem hiding this comment.
This is not the same and would case regression test failures.
| @@ -57,12 +70,31 @@ private void matchIssuesByFingerprint(final Report currentIssues) { | |||
| } | |||
| } | |||
|
|
|||
| private <K> void addIssueToMap(final Map<K, List<Issue>> map, final K key, final Issue issue) { | |||
| List<Issue> issues = map.get(key); | |||
There was a problem hiding this comment.
This would put more pressure on GC creating possibly unneeded object.
| } | ||
| return Optional.empty(); | ||
| } | ||
|
|
||
| private List<Issue> findReferenceByEquals(final Issue current) { | ||
| List<Issue> references = referencesByHash.get(current.hashCode()); |
There was a problem hiding this comment.
| List<Issue> references = referencesByHash.get(current.hashCode()); | |
| List<Issue> references = referencesByHash.getOrDefault(current.hashCode(), new ArrayList<>()); |
There was a problem hiding this comment.
This would put more pressure on GC creating possibly unneeded object.
| for (Issue reference : fixedIssues) { | ||
| if (current.equals(reference)) { | ||
| equalIssues.add(reference); | ||
| if (references != null) { |
There was a problem hiding this comment.
Can be removed if default is returned
uhafner
added
enhancement
Currently the difference of reports use a quadratic algorithm take takes hours when issues are several thousands.
The changes done copes with that.