LUCENE-7691 - add explicit size to arraylist creation if known

lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/HunspellStemFilterFactory.java

@@ -87,10 +87,9 @@ public void inform(ResourceLoader loader) throws IOException {
     String dicts[] = dictionaryFiles.split(",");
 
     InputStream affix = null;
-    List<InputStream> dictionaries = new ArrayList<>();
+    List<InputStream> dictionaries = new ArrayList<>(dicts.length);
 
     try {
-      dictionaries = new ArrayList<>();
       for (String file : dicts) {
         dictionaries.add(loader.openResource(file));
       }

lucene/analysis/common/src/java/org/apache/lucene/analysis/miscellaneous/CapitalizationFilterFactory.java

@@ -89,7 +89,7 @@ public CapitalizationFilterFactory(Map<String, String> args) {
 
     k = getSet(args, OK_PREFIX);
     if (k != null) {
-      okPrefix = new ArrayList<>();
+      okPrefix = new ArrayList<>(k.size());
       for (String item : k) {
         okPrefix.add(item.toCharArray());
       }

lucene/analysis/common/src/java/org/apache/lucene/analysis/synonym/SynonymGraphFilter.java

@@ -438,7 +438,7 @@ private void bufferOutputTokens(BytesRef bytes, int matchInputLength) {
     // TODO: we could encode this instead into the FST:
 
     // 1st pass: count how many new nodes we need
-    List<List<String>> paths = new ArrayList<>();
+    List<List<String>> paths = new ArrayList<>(count);
     for(int outputIDX=0;outputIDX<count;outputIDX++) {
       int wordID = bytesReader.readVInt();
       synonyms.words.get(wordID, scratchBytes);

lucene/analysis/common/src/java/org/apache/lucene/analysis/util/AbstractAnalysisFactory.java

@@ -293,8 +293,9 @@ protected final List<String> splitFileNames(String fileNames) {
     if (fileNames == null)
       return Collections.<String>emptyList();
 
-    List<String> result = new ArrayList<>();
-    for (String file : fileNames.split("(?<!\\\\),")) {
+    String[] fileNamesArr = fileNames.split("(?<!\\\\),");
+    List<String> result = new ArrayList<>(fileNamesArr.length);
+    for (String file : fileNamesArr) {
       result.add(file.replaceAll("\\\\(?=,)", ""));
     }
 

lucene/analysis/kuromoji/src/tools/java/org/apache/lucene/analysis/ja/util/TokenInfoDictionaryBuilder.java

@@ -68,8 +68,9 @@ public boolean accept(File dir, String name) {
         return name.endsWith(".csv");
       }
     };
-    ArrayList<File> csvFiles = new ArrayList<>();
-    for (File file : new File(dirname).listFiles(filter)) {
+    File[] foundFiles = new File(dirname).listFiles(filter);
+    ArrayList<File> csvFiles = new ArrayList<>(foundFiles.length);
+    for (File file : foundFiles) {
       csvFiles.add(file);
     }
     Collections.sort(csvFiles);

lucene/analysis/smartcn/src/java/org/apache/lucene/analysis/cn/smart/hhmm/BiSegGraph.java

@@ -167,10 +167,10 @@ public int getToCount() {
   public List<SegToken> getShortPath() {
     int current;
     int nodeCount = getToCount();
-    List<PathNode> path = new ArrayList<>();
     PathNode zeroPath = new PathNode();
     zeroPath.weight = 0;
     zeroPath.preNode = 0;
+    List<PathNode> path = new ArrayList<>(nodeCount + 1);
     path.add(zeroPath);
     for (current = 1; current <= nodeCount; current++) {
       double weight;

lucene/analysis/stempel/src/java/org/egothor/stemmer/Gener.java

@@ -78,7 +78,6 @@ public Gener() {}
   @Override
   public Trie optimize(Trie orig) {
     List<CharSequence> cmds = orig.cmds;
-    List<Row> rows = new ArrayList<>();
     List<Row> orows = orig.rows;
     int remap[] = new int[orows.size()];
 
@@ -90,7 +89,7 @@ public Trie optimize(Trie orig) {
     }
 
     Arrays.fill(remap, -1);
-    rows = removeGaps(orig.root, orows, new ArrayList<Row>(), remap);
+    List<Row> rows = removeGaps(orig.root, orows, new ArrayList<Row>(), remap);
 
     return new Trie(orig.forward, remap[orig.root], cmds, rows);
   }

lucene/analysis/stempel/src/java/org/egothor/stemmer/Lift.java

@@ -88,7 +88,6 @@ public Lift(boolean changeSkip) {
   @Override
   public Trie optimize(Trie orig) {
     List<CharSequence> cmds = orig.cmds;
-    List<Row> rows = new ArrayList<>();
     List<Row> orows = orig.rows;
     int remap[] = new int[orows.size()];
 
@@ -97,7 +96,7 @@ public Trie optimize(Trie orig) {
     }
 
     Arrays.fill(remap, -1);
-    rows = removeGaps(orig.root, orows, new ArrayList<Row>(), remap);
+    List<Row> rows = removeGaps(orig.root, orows, new ArrayList<Row>(), remap);
 
     return new Trie(orig.forward, remap[orig.root], cmds, rows);
   }

lucene/analysis/stempel/src/java/org/egothor/stemmer/MultiTrie.java

@@ -186,7 +186,7 @@ public void add(CharSequence key, CharSequence cmd) {
    */
   @Override
   public Trie reduce(Reduce by) {
-    List<Trie> h = new ArrayList<>();
+    List<Trie> h = new ArrayList<>(tries.size());
     for (Trie trie : tries)
       h.add(trie.reduce(by));
 

lucene/analysis/stempel/src/java/org/egothor/stemmer/MultiTrie2.java

@@ -277,7 +277,7 @@ public CharSequence[] decompose(CharSequence cmd) {
    */
   @Override
   public Trie reduce(Reduce by) {
-    List<Trie> h = new ArrayList<>();
+    List<Trie> h = new ArrayList<>(tries.size());
     for (Trie trie : tries)
       h.add(trie.reduce(by));
 

lucene/backward-codecs/src/java/org/apache/lucene/codecs/lucene54/Lucene54DocValuesProducer.java

@@ -1531,7 +1531,7 @@ public long ramBytesUsed() {
 
     @Override
     public Collection<Accountable> getChildResources() {
-      List<Accountable> resources = new ArrayList<>();
+      List<Accountable> resources = new ArrayList<>(2);
       resources.add(Accountables.namedAccountable("term bytes", terms));
       resources.add(Accountables.namedAccountable("term addresses", termAddresses));
       return Collections.unmodifiableList(resources);

lucene/codecs/src/java/org/apache/lucene/codecs/blockterms/BlockTermsReader.java

@@ -843,7 +843,7 @@ public long ramBytesUsed() {
 
   @Override
   public Collection<Accountable> getChildResources() {
-    List<Accountable> resources = new ArrayList<>();
+    List<Accountable> resources = new ArrayList<>(2);
     if (indexReader != null) {
       resources.add(Accountables.namedAccountable("term index", indexReader));
     }

lucene/codecs/src/java/org/apache/lucene/codecs/blockterms/FixedGapTermsIndexReader.java

@@ -263,7 +263,7 @@ public long ramBytesUsed() {
 
     @Override
     public Collection<Accountable> getChildResources() {
-      List<Accountable> resources = new ArrayList<>();
+      List<Accountable> resources = new ArrayList<>(2);
       if (termOffsets != null) {
         resources.add(Accountables.namedAccountable("term lengths", termOffsets));
       }

lucene/codecs/src/java/org/apache/lucene/codecs/memory/DirectDocValuesProducer.java

@@ -631,7 +631,7 @@ public long ramBytesUsed() {
 
     @Override
     public Collection<Accountable> getChildResources() {
-      List<Accountable> resources = new ArrayList<>();
+      List<Accountable> resources = new ArrayList<>(2);
       if (address != null) {
         resources.add(Accountables.namedAccountable("addresses", RamUsageEstimator.sizeOf(address)));
       }

lucene/codecs/src/java/org/apache/lucene/codecs/memory/MemoryDocValuesProducer.java

@@ -774,7 +774,7 @@ public long ramBytesUsed() {
 
     @Override
     public Collection<Accountable> getChildResources() {
-      List<Accountable> resources = new ArrayList<>();
+      List<Accountable> resources = new ArrayList<>(2);
       if (addresses != null) {
         resources.add(Accountables.namedAccountable("addresses", addresses));
       }

lucene/core/src/java/org/apache/lucene/codecs/DocValuesConsumer.java

@@ -476,7 +476,7 @@ public int nextDoc() throws IOException {
    * an Iterable that merges ordinals and values and filters deleted documents .
    */
   public void mergeSortedField(FieldInfo fieldInfo, final MergeState mergeState) throws IOException {
-    List<SortedDocValues> toMerge = new ArrayList<>();
+    List<SortedDocValues> toMerge = new ArrayList<>(mergeState.docValuesProducers.length);
     for (int i=0;i<mergeState.docValuesProducers.length;i++) {
       SortedDocValues values = null;
       DocValuesProducer docValuesProducer = mergeState.docValuesProducers[i];
@@ -647,7 +647,7 @@ public String toString() {
    */
   public void mergeSortedSetField(FieldInfo mergeFieldInfo, final MergeState mergeState) throws IOException {
 
-    List<SortedSetDocValues> toMerge = new ArrayList<>();
+    List<SortedSetDocValues> toMerge = new ArrayList<>(mergeState.docValuesProducers.length);
     for (int i=0;i<mergeState.docValuesProducers.length;i++) {
       SortedSetDocValues values = null;
       DocValuesProducer docValuesProducer = mergeState.docValuesProducers[i];

lucene/core/src/java/org/apache/lucene/codecs/FieldsConsumer.java

@@ -84,8 +84,8 @@ protected FieldsConsumer() {
    *  Implementations can override this method for more sophisticated
    *  merging (bulk-byte copying, etc). */
   public void merge(MergeState mergeState) throws IOException {
-    final List<Fields> fields = new ArrayList<>();
-    final List<ReaderSlice> slices = new ArrayList<>();
+    final List<Fields> fields = new ArrayList<>(mergeState.fieldsProducers.length);
+    final List<ReaderSlice> slices = new ArrayList<>(mergeState.fieldsProducers.length);
 
     int docBase = 0;
 

lucene/core/src/java/org/apache/lucene/codecs/StoredFieldsWriter.java

@@ -110,7 +110,7 @@ public int nextDoc() {
    *  Implementations can override this method for more sophisticated
    *  merging (bulk-byte copying, etc). */
   public int merge(MergeState mergeState) throws IOException {
-    List<StoredFieldsMergeSub> subs = new ArrayList<>();
+    List<StoredFieldsMergeSub> subs = new ArrayList<>(mergeState.storedFieldsReaders.length);
     for(int i=0;i<mergeState.storedFieldsReaders.length;i++) {
       StoredFieldsReader storedFieldsReader = mergeState.storedFieldsReaders[i];
       storedFieldsReader.checkIntegrity();

lucene/core/src/java/org/apache/lucene/codecs/TermVectorsWriter.java

@@ -196,7 +196,7 @@ public int nextDoc() {
    *  merging (bulk-byte copying, etc). */
   public int merge(MergeState mergeState) throws IOException {
 
-    List<TermVectorsMergeSub> subs = new ArrayList<>();
+    List<TermVectorsMergeSub> subs = new ArrayList<>(mergeState.termVectorsReaders.length);
     for(int i=0;i<mergeState.termVectorsReaders.length;i++) {
       TermVectorsReader reader = mergeState.termVectorsReaders[i];
       if (reader != null) {

lucene/core/src/java/org/apache/lucene/codecs/compressing/CompressingStoredFieldsIndexReader.java

@@ -191,7 +191,7 @@ public long ramBytesUsed() {
 
   @Override
   public Collection<Accountable> getChildResources() {
-    List<Accountable> resources = new ArrayList<>();
+    List<Accountable> resources = new ArrayList<>(2);
 
     long docBaseDeltaBytes = RamUsageEstimator.shallowSizeOf(docBasesDeltas);
     for (PackedInts.Reader r : docBasesDeltas) {

lucene/core/src/java/org/apache/lucene/codecs/compressing/CompressingStoredFieldsWriter.java

@@ -502,7 +502,7 @@ public int merge(MergeState mergeState) throws IOException {
        * If all readers are compressed and they have the same fieldinfos then we can merge the serialized document
        * directly.
        */
-      List<CompressingStoredFieldsMergeSub> subs = new ArrayList<>();
+      List<CompressingStoredFieldsMergeSub> subs = new ArrayList<>(mergeState.storedFieldsReaders.length);
       for(int i=0;i<mergeState.storedFieldsReaders.length;i++) {
         if (matching.matchingReaders[i] &&
             mergeState.storedFieldsReaders[i] instanceof CompressingStoredFieldsReader) {

lucene/core/src/java/org/apache/lucene/codecs/lucene60/Lucene60PointsReader.java

@@ -142,7 +142,7 @@ public long ramBytesUsed() {
 
   @Override
   public Collection<Accountable> getChildResources() {
-    List<Accountable> resources = new ArrayList<>();
+    List<Accountable> resources = new ArrayList<>(readers.size());
     for(Map.Entry<Integer,BKDReader> ent : readers.entrySet()) {
       resources.add(Accountables.namedAccountable(readState.fieldInfos.fieldInfo(ent.getKey()).name,
                                                   ent.getValue()));

lucene/core/src/java/org/apache/lucene/geo/SimpleGeoJSONPolygonParser.java

@@ -79,16 +79,16 @@ public Polygon[] parse() throws ParseException {
     if (polyType.equals("Polygon")) {
       return new Polygon[] {parsePolygon(coordinates)};
     } else {
-      List<Polygon> polygons = new ArrayList<>();
+      Polygon[] polygons = new Polygon[coordinates.size()];
       for(int i=0;i<coordinates.size();i++) {
         Object o = coordinates.get(i);
         if (o instanceof List == false) {
           throw newParseException("elements of coordinates array should be an array, but got: " + o.getClass());
         }
-        polygons.add(parsePolygon((List<Object>) o));
+        polygons[i] = (parsePolygon((List<Object>) o));
       }
 
-      return polygons.toArray(new Polygon[polygons.size()]);
+      return polygons;
     }
   }
 
@@ -217,11 +217,11 @@ private boolean isValidGeometryPath(String path) {
   }
 
   private Polygon parsePolygon(List<Object> coordinates) throws ParseException {
-    List<Polygon> holes = new ArrayList<>();
     Object o = coordinates.get(0);
     if (o instanceof List == false) {
       throw newParseException("first element of polygon array must be an array [[lat, lon], [lat, lon] ...] but got: " + o);
     }
+    List<Polygon> holes = new ArrayList<>(coordinates.size());
     double[][] polyPoints = parsePoints((List<Object>) o);
     for(int i=1;i<coordinates.size();i++) {
       o = coordinates.get(i);

lucene/core/src/java/org/apache/lucene/index/CoalescedUpdates.java

@@ -54,15 +54,15 @@ void update(FrozenBufferedUpdates in) {
       queries.put(query, BufferedUpdates.MAX_INT);
     }
 
-    List<DocValuesUpdate> numericPacket = new ArrayList<>();
+    List<DocValuesUpdate> numericPacket = new ArrayList<>(in.numericDVUpdates.length);
     numericDVUpdates.add(numericPacket);
     for (NumericDocValuesUpdate nu : in.numericDVUpdates) {
       NumericDocValuesUpdate clone = new NumericDocValuesUpdate(nu.term, nu.field, (Long) nu.value);
       clone.docIDUpto = Integer.MAX_VALUE;
       numericPacket.add(clone);
     }
 
-    List<DocValuesUpdate> binaryPacket = new ArrayList<>();
+    List<DocValuesUpdate> binaryPacket = new ArrayList<>(in.binaryDVUpdates.length);
     binaryDVUpdates.add(binaryPacket);
     for (BinaryDocValuesUpdate bu : in.binaryDVUpdates) {
       BinaryDocValuesUpdate clone = new BinaryDocValuesUpdate(bu.term, bu.field, (BytesRef) bu.value);

lucene/core/src/java/org/apache/lucene/index/IndexWriter.java

@@ -4320,7 +4320,7 @@ private int mergeMiddle(MergePolicy.OneMerge merge, MergePolicy mergePolicy) thr
 //      System.out.println("[" + Thread.currentThread().getName() + "] IW.mergeMiddle: merging " + merge.getMergeReaders());
 
       // Let the merge wrap readers
-      List<CodecReader> mergeReaders = new ArrayList<>();
+      List<CodecReader> mergeReaders = new ArrayList<>(merge.readers.size());
       for (SegmentReader reader : merge.readers) {
         mergeReaders.add(merge.wrapForMerge(reader));
       }

lucene/core/src/java/org/apache/lucene/index/MultiDocValues.java

@@ -913,7 +913,7 @@ public long ramBytesUsed() {
 
     @Override
     public Collection<Accountable> getChildResources() {
-      List<Accountable> resources = new ArrayList<>();
+      List<Accountable> resources = new ArrayList<>(3);
       resources.add(Accountables.namedAccountable("global ord deltas", globalOrdDeltas));
       resources.add(Accountables.namedAccountable("first segments", firstSegments));
       resources.add(Accountables.namedAccountable("segment map", segmentMap));

lucene/core/src/java/org/apache/lucene/search/Boolean2ScorerSupplier.java

@@ -161,8 +161,8 @@ public int freq() throws IOException {
       long minCost = Math.min(
           requiredNoScoring.stream().mapToLong(ScorerSupplier::cost).min().orElse(Long.MAX_VALUE),
           requiredScoring.stream().mapToLong(ScorerSupplier::cost).min().orElse(Long.MAX_VALUE));
-      List<Scorer> requiredScorers = new ArrayList<>();
-      List<Scorer> scoringScorers = new ArrayList<>();
+      List<Scorer> requiredScorers = new ArrayList<>(requiredNoScoring.size() + requiredScoring.size());
+      List<Scorer> scoringScorers = new ArrayList<>(requiredScoring.size());
       for (ScorerSupplier s : requiredNoScoring) {
         requiredScorers.add(s.get(randomAccess || s.cost() > minCost));
       }
@@ -206,7 +206,7 @@ protected boolean lessThan(ScorerSupplier a, ScorerSupplier b) {
       }
       return new MinShouldMatchSumScorer(weight, optionalScorers, minShouldMatch);
     } else {
-      final List<Scorer> optionalScorers = new ArrayList<>();
+      final List<Scorer> optionalScorers = new ArrayList<>(optional.size());
       for (ScorerSupplier scorer : optional) {
         optionalScorers.add(scorer.get(randomAccess));
       }

lucene/core/src/java/org/apache/lucene/search/BooleanWeight.java

@@ -49,7 +49,7 @@ final class BooleanWeight extends Weight {
     this.query = query;
     this.needsScores = needsScores;
     this.similarity = searcher.getSimilarity(needsScores);
-    weights = new ArrayList<>();
+    weights = new ArrayList<>(query.clauses().size());
     for (BooleanClause c : query) {
       Weight w = searcher.createWeight(c.getQuery(), needsScores && c.isScoring(), boost);
       weights.add(w);

lucene/core/src/java/org/apache/lucene/search/ConjunctionScorer.java

@@ -66,7 +66,7 @@ public int freq() {
 
   @Override
   public Collection<ChildScorer> getChildren() {
-    ArrayList<ChildScorer> children = new ArrayList<>();
+    ArrayList<ChildScorer> children = new ArrayList<>(scorers.length);
     for (Scorer scorer : scorers) {
       children.add(new ChildScorer(scorer, "MUST"));
     }

lucene/core/src/java/org/apache/lucene/search/DisjunctionMaxQuery.java

@@ -179,7 +179,7 @@ public Query rewrite(IndexReader reader) throws IOException {
     }
 
     boolean actuallyRewritten = false;
-    List<Query> rewrittenDisjuncts = new ArrayList<>();
+    List<Query> rewrittenDisjuncts = new ArrayList<>(disjuncts.length);
     for (Query sub : disjuncts) {
       Query rewrittenSub = sub.rewrite(reader);
       actuallyRewritten |= rewrittenSub != sub;

lucene/core/src/java/org/apache/lucene/search/DisjunctionScorer.java

@@ -195,7 +195,7 @@ public final float score() throws IOException {
 
   @Override
   public final Collection<ChildScorer> getChildren() {
-    ArrayList<ChildScorer> children = new ArrayList<>();
+    ArrayList<ChildScorer> children = new ArrayList<>(subScorers.size());
     for (DisiWrapper scorer : subScorers) {
       children.add(new ChildScorer(scorer.scorer, "SHOULD"));
     }

lucene/core/src/java/org/apache/lucene/search/ExactPhraseScorer.java

@@ -53,8 +53,8 @@ public PostingsAndPosition(PostingsEnum postings, int offset) {
     this.docScorer = docScorer;
     this.needsScores = needsScores;
 
-    List<DocIdSetIterator> iterators = new ArrayList<>();
-    List<PostingsAndPosition> postingsAndPositions = new ArrayList<>();
+    List<DocIdSetIterator> iterators = new ArrayList<>(postings.length);
+    List<PostingsAndPosition> postingsAndPositions = new ArrayList<>(postings.length);
     for(PhraseQuery.PostingsAndFreq posting : postings) {
       iterators.add(posting.postings);
       postingsAndPositions.add(new PostingsAndPosition(posting.postings, posting.position));

lucene/core/src/java/org/apache/lucene/search/IndexSearcher.java

@@ -604,7 +604,7 @@ public C call() throws Exception {
         }));
       }
 
-      final List<C> collectedCollectors = new ArrayList<>();
+      final List<C> collectedCollectors = new ArrayList<>(topDocsFutures.size());
       for (Future<C> future : topDocsFutures) {
         try {
           collectedCollectors.add(future.get());

lucene/core/src/java/org/apache/lucene/search/MinShouldMatchSumScorer.java

@@ -115,7 +115,7 @@ protected boolean lessThan(Long a, Long b) {
       addLead(new DisiWrapper(scorer));
     }
 
-    List<ChildScorer> children = new ArrayList<>();
+    List<ChildScorer> children = new ArrayList<>(scorers.size());
     for (Scorer scorer : scorers) {
       children.add(new ChildScorer(scorer, "SHOULD"));
     }

lucene/core/src/java/org/apache/lucene/search/SearcherLifetimeManager.java

@@ -246,7 +246,7 @@ public synchronized void prune(Pruner pruner) throws IOException {
     // (not thread-safe since the values can change while
     // ArrayList is init'ing itself); must instead iterate
     // ourselves:
-    final List<SearcherTracker> trackers = new ArrayList<>();
+    final List<SearcherTracker> trackers = new ArrayList<>(searchers.size());
     for(SearcherTracker tracker : searchers.values()) {
       trackers.add(tracker);
     }