List of usage examples for com.google.common.hash BloomFilter put
public boolean put(T object)
From source file:org.apache.niolex.common.guava.GuavaHashing.java
/** * @param args/*from w w w .j a v a 2s.com*/ */ public static void main(String[] args) { // Common Hash HashFunction hf = Hashing.goodFastHash(32); HashCode code = hf.hashObject(new Person(1, "Jiyun", "Xie", 1984), PersonFunnel.INSTANCE); System.out.println("Code1 => " + code.asInt()); code = hf.hashObject(new Person(1, "Jiyun", "Xie", 1985), PersonFunnel.INSTANCE); System.out.println("Code2 => " + code.asInt()); // Consistent Hashing HashFunction hf2 = Hashing.goodFastHash(64); code = hf2.hashObject(new Person(1, "Jiyun", "Xie", 1984), PersonFunnel.INSTANCE); System.out.println("Code3 => " + code.asLong()); long hash = code.asLong(); int bucket = Hashing.consistentHash(code, 100); System.out.println("Bucket1 => " + bucket); bucket = Hashing.consistentHash(hash, 101); System.out.println("Bucket2 => " + bucket); for (int i = 0; i < 10; ++i) { System.out.println("HashTo5 => " + Hashing.consistentHash(i, 5)); System.out.println("HashTo6 => " + Hashing.consistentHash(i, 6)); } // BloomFilter BloomFilter<Person> friends = BloomFilter.create(PersonFunnel.INSTANCE, 500, 0.02); for (int i = 0; i < 500; ++i) { friends.put(new Person(i, "Jiyun", "Xie", 1984 + i)); } int k = 0; for (int i = 0; i < 500; ++i) { if (!friends.mightContain(new Person(i, "Jiyun", "Xie", 1984 + i))) { System.out.println("Error1 => " + i); ++k; } } System.out.println("fnp => (should be 0)" + ((double) k / 500)); k = 0; for (int i = 0; i < 1000; i += 2) { if (friends.mightContain(new Person(i, "Jiyun", "Xie", 1984 + i))) { //System.out.println("Error2 => " + i); ++k; } } System.out.println("fpp => " + ((double) k / 500)); }
From source file:com.jordanwilliams.heftydb.test.performance.table.file.TableBloomFilterPerformance.java
public static void main(String[] args) throws Exception { TestFileHelper.createTestDirectory(); KeyValueGenerator keyValueGenerator = new KeyValueGenerator(); Value value = new Value(keyValueGenerator.testValue(100)); System.out.println("Writing bloom filter"); Paths paths = ConfigGenerator.testPaths(); TableBloomFilterWriter filterWriter = TableBloomFilterWriter.open(1, paths, RECORD_COUNT); BloomFilter<Key> guavaFilter = BloomFilter.create(new Funnel<Key>() { @Override/*from www.j a v a 2s . c o m*/ public void funnel(Key key, PrimitiveSink primitiveSink) { primitiveSink.putBytes(key.data().array()); } }, RECORD_COUNT, 0.01); for (int i = 0; i < RECORD_COUNT; i++) { value.data().rewind(); filterWriter.write(new Key(ByteBuffers.fromString(i + ""), i)); guavaFilter.put(new Key(ByteBuffers.fromString(i + ""), i)); } filterWriter.finish(); System.out.println("Reading bloom filter"); TableBloomFilter tableBloomFilter = TableBloomFilter.read(1, paths); double hits = 0; double misses = 0; double ghits = 0; double gmisses = 0; for (int i = RECORD_COUNT * 2; i > RECORD_COUNT; i--) { if (tableBloomFilter.mightContain(new Key(ByteBuffers.fromString(i + ""), i))) { hits++; } else { misses++; } if (guavaFilter.mightContain(new Key(ByteBuffers.fromString(i + ""), i))) { ghits++; } else { gmisses++; } } System.out.println("False positive rate: " + hits / (hits + misses)); System.out.println("Guava positive rate: " + ghits / (ghits + gmisses)); TestFileHelper.cleanUpTestFiles(); }
From source file:com.edduarte.vokter.diff.DifferenceMatcher.java
@Override public Set<DifferenceMatcher.Result> call() { Stopwatch sw = Stopwatch.createStarted(); Set<Result> matchedDiffs = new ConcurrentHashSet<>(); DifferenceEvent lastAction = DifferenceEvent.nothing; BloomFilter<String> lastBloomFilter = null; for (Difference r : differences) { if (lastAction == DifferenceEvent.nothing || r.getAction() != lastAction) { // reset the bloom filter being used lastBloomFilter = BloomFilter.create((from, into) -> into.putUnencodedChars(from), 10); lastAction = r.getAction();//from ww w .j a va 2 s . c om } BloomFilter<String> bloomFilter = lastBloomFilter; bloomFilter.put(r.getOccurrenceText()); // check if AT LEAST ONE of the keywords has ALL of its words // contained in the diff text keywords.parallelStream().unordered().filter(kw -> kw.textStream().allMatch(bloomFilter::mightContain)) .map(kw -> new Pair<>(r, kw)) .filter((pair) -> pair.b().textStream().anyMatch(pair.a().getOccurrenceText()::equals)) .map((pair) -> { Difference diff = pair.a(); Keyword keyword = pair.b(); DifferenceEvent i = diff.getAction(); if (i == DifferenceEvent.inserted && !ignoreAdded) { return new Result(diff.getAction(), keyword, diff.getSnippet()); } else if (i == DifferenceEvent.deleted && !ignoreRemoved) { return new Result(diff.getAction(), keyword, diff.getSnippet()); } return null; }).filter(diff -> diff != null).forEach(matchedDiffs::add); } sw.stop(); logger.info("Completed difference matching for keywords '{}' in {}", keywords.toString(), sw.toString()); return matchedDiffs; }
From source file:com.edduarte.argus.diff.DifferenceMatcher.java
@Override public Set<DifferenceMatcher.Result> call() { Stopwatch sw = Stopwatch.createStarted(); Set<Result> matchedDiffs = new ConcurrentHashSet<>(); DifferenceAction lastAction = DifferenceAction.nothing; BloomFilter<String> lastBloomFilter = null; for (Difference r : differences) { if (lastAction == DifferenceAction.nothing || r.getAction() != lastAction) { // reset the bloom filter being used lastBloomFilter = BloomFilter.create((from, into) -> into.putUnencodedChars(from), 10); lastAction = r.getAction();// w w w . j a v a 2 s . c om } BloomFilter<String> bloomFilter = lastBloomFilter; bloomFilter.put(r.getOccurrenceText()); // check if AT LEAST ONE of the keywords has ALL of its words // contained in the diff text keywords.parallelStream().unordered().filter(kw -> kw.textStream().allMatch(bloomFilter::mightContain)) .map(kw -> new Pair<>(r, kw)) .filter((pair) -> pair.b().textStream().anyMatch(pair.a().getOccurrenceText()::equals)) .map((pair) -> { Difference diff = pair.a(); Keyword keyword = pair.b(); DifferenceAction i = diff.getAction(); if (i == DifferenceAction.inserted && !ignoreAdded) { return new Result(diff.getAction(), keyword, diff.getSnippet()); } else if (i == DifferenceAction.deleted && !ignoreRemoved) { return new Result(diff.getAction(), keyword, diff.getSnippet()); } return null; }).filter(diff -> diff != null).forEach(matchedDiffs::add); } sw.stop(); logger.info("Completed difference matching for keywords '{}' in {}", keywords.toString(), sw.toString()); return matchedDiffs; }
From source file:graphene.ingest.batchoptimizers.BasicBatchOptimizer.java
/** * Checks to see if the relationship is unique, and if it is, adds it to the * bloom filter.//from w w w .j a v a2s .c om * * @param from * @param to * @param rel * @return */ public boolean is_unique_rel(BloomFilter<String> bf, Long from, String rel, Long to) { String key = "" + from + to + rel; if (bf.mightContain(key)) { return false; } else { bf.put(key); return true; } }
From source file:etri.sdn.controller.module.routing.Route.java
/** * This method returns a bloom filter (set) that contains all node-port tuple * inside the route. This bloom filter is used to test a link is within the * route very fast.// ww w.j a v a2 s . co m * * @return BloomFilter<NodePortTuple> */ public BloomFilter<NodePortTuple> getBloomFilter() { BloomFilter<NodePortTuple> ret = BloomFilter.create(RouteFunnel.INSTANCE, 20); for (NodePortTuple npt : this.switchPorts) { ret.put(npt); } return ret; }