MAHOUT-1652: Java 7 update
Project: http://git-wip-us.apache.org/repos/asf/mahout/repo Commit: http://git-wip-us.apache.org/repos/asf/mahout/commit/87c15bea Tree: http://git-wip-us.apache.org/repos/asf/mahout/tree/87c15bea Diff: http://git-wip-us.apache.org/repos/asf/mahout/diff/87c15bea Branch: refs/heads/master Commit: 87c15bea46081baab98db9a0dbd480c5dac5ceb4 Parents: ec8a5a0 Author: Suneel Marthi <[email protected]> Authored: Wed Mar 25 02:43:44 2015 -0400 Committer: Suneel Marthi <[email protected]> Committed: Wed Mar 25 02:43:44 2015 -0400 ---------------------------------------------------------------------- .../taste/example/email/MailToPrefsDriver.java | 4 +- .../cf/taste/example/email/MailToRecMapper.java | 4 +- .../taste/example/kddcup/DataFileIterator.java | 2 +- .../taste/example/kddcup/KDDCupDataModel.java | 8 +-- .../track1/svd/ParallelArraysSGDFactorizer.java | 4 +- .../example/kddcup/track2/Track2Callable.java | 2 +- .../kddcup/track2/TrackItemSimilarity.java | 2 +- .../classifier/df/mapreduce/TestForest.java | 2 +- .../sequencelearning/hmm/PosTagger.java | 3 +- .../tools/ClusterQualitySummarizer.java | 12 ++-- .../model/cassandra/CassandraDataModel.java | 20 +++--- .../impl/model/jdbc/AbstractJDBCDataModel.java | 6 +- .../mahout/cf/taste/web/RecommenderServlet.java | 4 +- .../mahout/clustering/cdbw/CDbwEvaluator.java | 12 ++-- .../clustering/conversion/InputMapper.java | 10 +-- .../clustering/evaluation/ClusterEvaluator.java | 2 +- .../mahout/clustering/lda/LDAPrintTopics.java | 6 +- .../text/MultipleTextFileInputFormat.java | 4 +- .../mahout/text/PrefixAdditionFilter.java | 2 +- .../text/ReadOnlyFileSystemDirectory.java | 4 +- .../mahout/text/SequenceFilesFromDirectory.java | 4 +- .../SequenceFilesFromLuceneStorageMRJob.java | 6 +- .../text/SequenceFilesFromMailArchives.java | 4 +- .../mahout/text/WholeFileRecordReader.java | 6 +- .../mahout/text/WikipediaToSequenceFile.java | 10 +-- .../org/apache/mahout/utils/MatrixDumper.java | 2 +- .../apache/mahout/utils/SequenceFileDumper.java | 16 ++--- .../org/apache/mahout/utils/SplitInput.java | 12 ++-- .../org/apache/mahout/utils/SplitInputJob.java | 2 +- .../utils/clustering/AbstractClusterWriter.java | 2 +- .../mahout/utils/vectors/VectorDumper.java | 5 +- .../mahout/utils/vectors/VectorHelper.java | 9 +-- .../utils/vectors/lucene/ClusterLabels.java | 4 +- .../text/LuceneSegmentRecordReaderTest.java | 25 ++++--- .../text/TestSequenceFilesFromDirectory.java | 9 +-- .../mahout/utils/TestConcatenateVectorsJob.java | 2 +- .../mahout/cf/taste/common/Refreshable.java | 2 +- .../apache/mahout/cf/taste/hadoop/als/ALS.java | 5 +- .../hadoop/als/ParallelALSFactorizationJob.java | 14 ++-- .../cf/taste/hadoop/als/PredictionMapper.java | 2 +- .../item/AggregateAndRecommendReducer.java | 4 +- .../mahout/cf/taste/hadoop/item/IDReader.java | 13 ++-- .../hadoop/item/UserVectorSplitterMapper.java | 3 + .../AbstractDifferenceRecommenderEvaluator.java | 4 +- .../GenericRecommenderIRStatsEvaluator.java | 2 +- .../impl/model/GenericBooleanPrefDataModel.java | 4 +- .../cf/taste/impl/model/GenericDataModel.java | 4 +- .../cf/taste/impl/model/MemoryIDMigrator.java | 2 +- .../PlusAnonymousConcurrentUserDataModel.java | 6 +- .../cf/taste/impl/model/file/FileDataModel.java | 12 ++-- .../taste/impl/model/file/FileIDMigrator.java | 4 +- .../neighborhood/CachingUserNeighborhood.java | 2 +- .../impl/recommender/CachingRecommender.java | 4 +- .../recommender/ItemAverageRecommender.java | 2 +- .../recommender/ItemUserAverageRecommender.java | 4 +- .../cf/taste/impl/recommender/NullRescorer.java | 4 +- .../SamplingCandidateItemsStrategy.java | 2 +- .../cf/taste/impl/recommender/TopItems.java | 8 +-- .../similarity/AveragingPreferenceInferrer.java | 2 +- .../impl/similarity/CachingItemSimilarity.java | 2 +- .../impl/similarity/CachingUserSimilarity.java | 2 +- .../impl/similarity/GenericItemSimilarity.java | 6 +- .../impl/similarity/GenericUserSimilarity.java | 4 +- .../MultithreadedBatchItemSimilarities.java | 4 +- .../mahout/cf/taste/model/IDMigrator.java | 2 - .../classifier/naivebayes/BayesUtils.java | 4 +- .../naivebayes/test/BayesTestMapper.java | 2 +- .../naivebayes/test/TestNaiveBayesDriver.java | 6 +- .../sequencelearning/hmm/BaumWelchTrainer.java | 5 +- .../sequencelearning/hmm/ViterbiEvaluator.java | 5 +- .../mahout/clustering/AbstractCluster.java | 2 +- .../WeightedPropertyVectorWritable.java | 2 +- .../clustering/fuzzykmeans/FuzzyKMeansUtil.java | 2 +- .../clustering/iterator/ClusterIterator.java | 2 +- .../mahout/clustering/kmeans/KMeansUtil.java | 2 +- .../clustering/kmeans/RandomSeedGenerator.java | 4 +- .../cvb/InMemoryCollapsedVariationalBayes0.java | 2 +- .../mahout/clustering/lda/cvb/ModelTrainer.java | 2 +- .../mahout/clustering/lda/cvb/TopicModel.java | 2 +- .../mahout/clustering/spectral/VectorCache.java | 2 +- .../spectral/kmeans/EigenSeedGenerator.java | 4 +- .../streaming/cluster/BallKMeans.java | 4 +- .../postprocessor/ClusterCountReader.java | 4 +- .../ClusterOutputPostProcessorDriver.java | 2 +- .../org/apache/mahout/common/ClassUtils.java | 16 +---- .../org/apache/mahout/common/HadoopUtil.java | 6 +- .../java/org/apache/mahout/common/IOUtils.java | 6 +- .../java/org/apache/mahout/common/Pair.java | 4 +- .../org/apache/mahout/common/Parameters.java | 4 +- .../iterator/CopyConstructorIterator.java | 6 +- .../StableFixedSizeSamplingIterator.java | 4 +- .../common/iterator/StringRecordIterator.java | 2 +- .../sequencefile/SequenceFileDirIterable.java | 2 +- .../sequencefile/SequenceFileDirIterator.java | 2 +- .../SequenceFileDirValueIterable.java | 2 +- .../SequenceFileDirValueIterator.java | 2 +- .../sequencefile/SequenceFileIterable.java | 2 +- .../sequencefile/SequenceFileIterator.java | 2 +- .../sequencefile/SequenceFileValueIterable.java | 2 +- .../sequencefile/SequenceFileValueIterator.java | 2 +- .../org/apache/mahout/driver/MahoutDriver.java | 2 - .../main/java/org/apache/mahout/ep/State.java | 2 +- .../org/apache/mahout/math/MatrixUtils.java | 2 +- .../math/hadoop/DistributedRowMatrix.java | 2 +- .../math/hadoop/MatrixColumnMeansJob.java | 2 +- .../mahout/math/hadoop/TimesSquaredJob.java | 4 +- .../math/hadoop/similarity/SeedVectorUtil.java | 2 +- .../hadoop/similarity/cooccurrence/Vectors.java | 2 +- .../DistributedConjugateGradientSolver.java | 14 ++-- .../mahout/math/hadoop/stats/BasicStats.java | 2 +- .../hadoop/stochasticsvd/ABtDenseOutJob.java | 14 +--- .../math/hadoop/stochasticsvd/ABtJob.java | 12 +--- .../mahout/math/hadoop/stochasticsvd/BtJob.java | 14 ++-- .../math/hadoop/stochasticsvd/SSVDHelper.java | 74 ++++++++++---------- .../math/hadoop/stochasticsvd/SSVDSolver.java | 1 - .../mahout/math/hadoop/stochasticsvd/UJob.java | 1 - .../hadoop/stochasticsvd/qr/QRFirstStep.java | 3 +- .../hadoop/stochasticsvd/qr/QRLastStep.java | 4 +- .../mahout/math/neighborhood/BruteSearch.java | 8 +-- .../math/neighborhood/FastProjectionSearch.java | 8 +-- .../LocalitySensitiveHashSearch.java | 4 +- .../math/neighborhood/ProjectionSearch.java | 12 ++-- .../mahout/math/random/RandomProjector.java | 2 +- .../math/ssvd/SequentialOutOfCoreSvd.java | 44 +++--------- .../mahout/vectorizer/DictionaryVectorizer.java | 2 +- .../collocations/llr/CollocMapper.java | 12 ++-- .../encoders/CachingContinuousValueEncoder.java | 6 +- .../encoders/CachingStaticWordValueEncoder.java | 6 +- .../vectorizer/term/TFPartialVectorReducer.java | 2 +- .../mahout/vectorizer/term/TermCountMapper.java | 2 +- .../mahout/vectorizer/tfidf/TFIDFConverter.java | 2 +- pom.xml | 8 +-- 132 files changed, 324 insertions(+), 433 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToPrefsDriver.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToPrefsDriver.java b/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToPrefsDriver.java index 09bd817..1d150ae 100644 --- a/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToPrefsDriver.java +++ b/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToPrefsDriver.java @@ -55,7 +55,7 @@ import org.slf4j.LoggerFactory; /** * Convert the Mail archives (see {@link org.apache.mahout.text.SequenceFilesFromMailArchives}) to a preference - * file that can be consumed by the {@link org.apache.mahout.cf.taste.hadoop.pseudo.RecommenderJob}. + * file that can be consumed by the {@link org.apache.mahout.cf.taste.hadoop.item.RecommenderJob}. * <p/> * This assumes the input is a Sequence File, that the key is: filename/message id and the value is a list * (separated by the user's choosing) containing the from email and any references @@ -246,7 +246,7 @@ public final class MailToPrefsDriver extends AbstractJob { Path filesPattern = new Path(inputPath, OUTPUT_FILES_PATTERN); int i = 1; //start at 1, since a miss in the OpenObjectIntHashMap returns a 0 for (Pair<Writable, Writable> record - : new SequenceFileDirIterable<Writable, Writable>(filesPattern, PathType.GLOB, null, null, true, conf)) { + : new SequenceFileDirIterable<>(filesPattern, PathType.GLOB, null, null, true, conf)) { if (currentChunkSize > chunkSizeLimit) { Closeables.close(dictWriter, false); chunkIndex++; http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToRecMapper.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToRecMapper.java b/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToRecMapper.java index 2de08cc..91bbd17 100644 --- a/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToRecMapper.java +++ b/examples/src/main/java/org/apache/mahout/cf/taste/example/email/MailToRecMapper.java @@ -32,8 +32,8 @@ public final class MailToRecMapper extends Mapper<Text, Text, Text, LongWritable private static final Logger log = LoggerFactory.getLogger(MailToRecMapper.class); - private final OpenObjectIntHashMap<String> fromDictionary = new OpenObjectIntHashMap<String>(); - private final OpenObjectIntHashMap<String> msgIdDictionary = new OpenObjectIntHashMap<String>(); + private final OpenObjectIntHashMap<String> fromDictionary = new OpenObjectIntHashMap<>(); + private final OpenObjectIntHashMap<String> msgIdDictionary = new OpenObjectIntHashMap<>(); private String separator = "\n"; private int fromIdx; private int refsIdx; http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/DataFileIterator.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/DataFileIterator.java b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/DataFileIterator.java index d0c9f68..786e080 100644 --- a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/DataFileIterator.java +++ b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/DataFileIterator.java @@ -112,7 +112,7 @@ public final class DataFileIterator ratingsLeftToRead--; } - return new Pair<PreferenceArray,long[]>(currentUserPrefs, timestamps); + return new Pair<>(currentUserPrefs, timestamps); } @Override http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/KDDCupDataModel.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/KDDCupDataModel.java b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/KDDCupDataModel.java index 1c26027..4b62050 100644 --- a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/KDDCupDataModel.java +++ b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/KDDCupDataModel.java @@ -72,11 +72,11 @@ public final class KDDCupDataModel implements DataModel { Iterator<Pair<PreferenceArray,long[]>> dataIterator = new DataFileIterator(dataFile); if (samplingRate < 1.0) { - dataIterator = new SamplingIterator<Pair<PreferenceArray,long[]>>(dataIterator, samplingRate); + dataIterator = new SamplingIterator<>(dataIterator, samplingRate); } - FastByIDMap<PreferenceArray> userData = new FastByIDMap<PreferenceArray>(); - FastByIDMap<FastByIDMap<Long>> timestamps = new FastByIDMap<FastByIDMap<Long>>(); + FastByIDMap<PreferenceArray> userData = new FastByIDMap<>(); + FastByIDMap<FastByIDMap<Long>> timestamps = new FastByIDMap<>(); while (dataIterator.hasNext()) { @@ -86,7 +86,7 @@ public final class KDDCupDataModel implements DataModel { userData.put(userPrefs.getUserID(0), userPrefs); if (storeDates) { - FastByIDMap<Long> itemTimestamps = new FastByIDMap<Long>(); + FastByIDMap<Long> itemTimestamps = new FastByIDMap<>(); for (int i = 0; i < timestampsForPrefs.length; i++) { long timestamp = timestampsForPrefs[i]; if (timestamp > 0L) { http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track1/svd/ParallelArraysSGDFactorizer.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track1/svd/ParallelArraysSGDFactorizer.java b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track1/svd/ParallelArraysSGDFactorizer.java index 715ebc3..a99d54c 100644 --- a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track1/svd/ParallelArraysSGDFactorizer.java +++ b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track1/svd/ParallelArraysSGDFactorizer.java @@ -107,7 +107,7 @@ public class ParallelArraysSGDFactorizer implements Factorizer { int numPrefs = factorizablePreferences.numPreferences(); log.info("Mapping {} users...", numUsers); - userIDMapping = new FastByIDMap<Integer>(numUsers); + userIDMapping = new FastByIDMap<>(numUsers); int index = 0; LongPrimitiveIterator userIterator = factorizablePreferences.getUserIDs(); while (userIterator.hasNext()) { @@ -115,7 +115,7 @@ public class ParallelArraysSGDFactorizer implements Factorizer { } log.info("Mapping {} items", numItems); - itemIDMapping = new FastByIDMap<Integer>(numItems); + itemIDMapping = new FastByIDMap<>(numItems); index = 0; LongPrimitiveIterator itemIterator = factorizablePreferences.getItemIDs(); while (itemIterator.hasNext()) { http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/Track2Callable.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/Track2Callable.java b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/Track2Callable.java index 2454017..7ecc5fe 100644 --- a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/Track2Callable.java +++ b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/Track2Callable.java @@ -54,7 +54,7 @@ final class Track2Callable implements Callable<UserResult> { throw new IllegalArgumentException("Expecting 6 items for user but got " + userTest); } long userID = userTest.get(0).getUserID(); - TreeMap<Double,Long> estimateToItemID = new TreeMap<Double,Long>(Collections.reverseOrder()); + TreeMap<Double,Long> estimateToItemID = new TreeMap<>(Collections.reverseOrder()); for (int i = 0; i < testSize; i++) { long itemID = userTest.getItemID(i); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/TrackItemSimilarity.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/TrackItemSimilarity.java b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/TrackItemSimilarity.java index 54f2fa7..3012a84 100644 --- a/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/TrackItemSimilarity.java +++ b/examples/src/main/java/org/apache/mahout/cf/taste/example/kddcup/track2/TrackItemSimilarity.java @@ -34,7 +34,7 @@ final class TrackItemSimilarity implements ItemSimilarity { private final FastByIDMap<TrackData> trackData; TrackItemSimilarity(File dataFileDirectory) throws IOException { - trackData = new FastByIDMap<TrackData>(); + trackData = new FastByIDMap<>(); for (String line : new FileLineIterable(KDDCupDataModel.getTrackFile(dataFileDirectory))) { TrackData trackDatum = new TrackData(line); trackData.put(trackDatum.getTrackID(), trackDatum); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/classifier/df/mapreduce/TestForest.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/classifier/df/mapreduce/TestForest.java b/examples/src/main/java/org/apache/mahout/classifier/df/mapreduce/TestForest.java index 411c68f..9ce2104 100644 --- a/examples/src/main/java/org/apache/mahout/classifier/df/mapreduce/TestForest.java +++ b/examples/src/main/java/org/apache/mahout/classifier/df/mapreduce/TestForest.java @@ -227,7 +227,7 @@ public class TestForest extends Configured implements Tool { Random rng = RandomUtils.getRandom(); List<double[]> resList = Lists.newArrayList(); - if (dataFS.getFileStatus(dataPath).isDir()) { + if (dataFS.getFileStatus(dataPath).isDirectory()) { //the input is a directory of files testDirectory(outputPath, converter, forest, dataset, resList, rng); } else { http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/PosTagger.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/PosTagger.java b/examples/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/PosTagger.java index 4864021..4ba844d 100644 --- a/examples/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/PosTagger.java +++ b/examples/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/PosTagger.java @@ -27,7 +27,6 @@ import java.util.regex.Pattern; import com.google.common.base.Charsets; import com.google.common.collect.Lists; import com.google.common.collect.Maps; -import com.google.common.io.CharStreams; import com.google.common.io.Resources; import org.apache.mahout.math.Matrix; import org.slf4j.Logger; @@ -122,7 +121,7 @@ public final class PosTagger { List<Integer> observedSequence = Lists.newLinkedList(); List<Integer> hiddenSequence = Lists.newLinkedList(); - for (String line : CharStreams.readLines(Resources.newReaderSupplier(new URL(url), Charsets.UTF_8))) { + for (String line :Resources.readLines(new URL(url), Charsets.UTF_8)) { if (line.isEmpty()) { // new sentence starts int[] observedSequenceArray = new int[observedSequence.size()]; http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/examples/src/main/java/org/apache/mahout/clustering/streaming/tools/ClusterQualitySummarizer.java ---------------------------------------------------------------------- diff --git a/examples/src/main/java/org/apache/mahout/clustering/streaming/tools/ClusterQualitySummarizer.java b/examples/src/main/java/org/apache/mahout/clustering/streaming/tools/ClusterQualitySummarizer.java index ffc07a4..038a097 100644 --- a/examples/src/main/java/org/apache/mahout/clustering/streaming/tools/ClusterQualitySummarizer.java +++ b/examples/src/main/java/org/apache/mahout/clustering/streaming/tools/ClusterQualitySummarizer.java @@ -110,23 +110,23 @@ public class ClusterQualitySummarizer extends AbstractJob { List<Centroid> centroidsCompare = null; if (mahoutKMeansFormat) { SequenceFileDirValueIterable<ClusterWritable> clusterIterable = - new SequenceFileDirValueIterable<ClusterWritable>(new Path(centroidFile), PathType.GLOB, conf); + new SequenceFileDirValueIterable<>(new Path(centroidFile), PathType.GLOB, conf); centroids = Lists.newArrayList(IOUtils.getCentroidsFromClusterWritableIterable(clusterIterable)); } else { SequenceFileDirValueIterable<CentroidWritable> centroidIterable = - new SequenceFileDirValueIterable<CentroidWritable>(new Path(centroidFile), PathType.GLOB, conf); + new SequenceFileDirValueIterable<>(new Path(centroidFile), PathType.GLOB, conf); centroids = Lists.newArrayList(IOUtils.getCentroidsFromCentroidWritableIterable(centroidIterable)); } if (centroidCompareFile != null) { if (mahoutKMeansFormatCompare) { SequenceFileDirValueIterable<ClusterWritable> clusterCompareIterable = - new SequenceFileDirValueIterable<ClusterWritable>(new Path(centroidCompareFile), PathType.GLOB, conf); + new SequenceFileDirValueIterable<>(new Path(centroidCompareFile), PathType.GLOB, conf); centroidsCompare = Lists.newArrayList( IOUtils.getCentroidsFromClusterWritableIterable(clusterCompareIterable)); } else { SequenceFileDirValueIterable<CentroidWritable> centroidCompareIterable = - new SequenceFileDirValueIterable<CentroidWritable>(new Path(centroidCompareFile), PathType.GLOB, conf); + new SequenceFileDirValueIterable<>(new Path(centroidCompareFile), PathType.GLOB, conf); centroidsCompare = Lists.newArrayList( IOUtils.getCentroidsFromCentroidWritableIterable(centroidCompareIterable)); } @@ -134,7 +134,7 @@ public class ClusterQualitySummarizer extends AbstractJob { // Reading in the "training" set. SequenceFileDirValueIterable<VectorWritable> trainIterable = - new SequenceFileDirValueIterable<VectorWritable>(new Path(trainFile), PathType.GLOB, conf); + new SequenceFileDirValueIterable<>(new Path(trainFile), PathType.GLOB, conf); Iterable<Vector> trainDatapoints = IOUtils.getVectorsFromVectorWritableIterable(trainIterable); Iterable<Vector> datapoints = trainDatapoints; @@ -144,7 +144,7 @@ public class ClusterQualitySummarizer extends AbstractJob { // Also adding in the "test" set. if (testFile != null) { SequenceFileDirValueIterable<VectorWritable> testIterable = - new SequenceFileDirValueIterable<VectorWritable>(new Path(testFile), PathType.GLOB, conf); + new SequenceFileDirValueIterable<>(new Path(testFile), PathType.GLOB, conf); Iterable<Vector> testDatapoints = IOUtils.getVectorsFromVectorWritableIterable(testIterable); printSummaries(ClusteringUtils.summarizeClusterDistances(testDatapoints, centroids, http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/cassandra/CassandraDataModel.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/cassandra/CassandraDataModel.java b/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/cassandra/CassandraDataModel.java index 68f70c3..b220993 100644 --- a/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/cassandra/CassandraDataModel.java +++ b/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/cassandra/CassandraDataModel.java @@ -133,12 +133,12 @@ public final class CassandraDataModel implements DataModel, Closeable { keyspace = HFactory.createKeyspace(keyspaceName, cluster); keyspace.setConsistencyLevelPolicy(new OneConsistencyLevelPolicy()); - userCache = new Cache<Long,PreferenceArray>(new UserPrefArrayRetriever(), 1 << 20); - itemCache = new Cache<Long,PreferenceArray>(new ItemPrefArrayRetriever(), 1 << 20); - itemIDsFromUserCache = new Cache<Long,FastIDSet>(new ItemIDsFromUserRetriever(), 1 << 20); - userIDsFromItemCache = new Cache<Long,FastIDSet>(new UserIDsFromItemRetriever(), 1 << 20); - userCountCache = new AtomicReference<Integer>(null); - itemCountCache = new AtomicReference<Integer>(null); + userCache = new Cache<>(new UserPrefArrayRetriever(), 1 << 20); + itemCache = new Cache<>(new ItemPrefArrayRetriever(), 1 << 20); + itemIDsFromUserCache = new Cache<>(new ItemIDsFromUserRetriever(), 1 << 20); + userIDsFromItemCache = new Cache<>(new UserIDsFromItemRetriever(), 1 << 20); + userCountCache = new AtomicReference<>(null); + itemCountCache = new AtomicReference<>(null); } @Override @@ -262,25 +262,25 @@ public final class CassandraDataModel implements DataModel, Closeable { Mutator<Long> mutator = HFactory.createMutator(keyspace, LongSerializer.get()); - HColumn<Long,Float> itemForUsers = new HColumnImpl<Long,Float>(LongSerializer.get(), FloatSerializer.get()); + HColumn<Long,Float> itemForUsers = new HColumnImpl<>(LongSerializer.get(), FloatSerializer.get()); itemForUsers.setName(itemID); itemForUsers.setClock(now); itemForUsers.setValue(value); mutator.addInsertion(userID, USERS_CF, itemForUsers); - HColumn<Long,Float> userForItems = new HColumnImpl<Long,Float>(LongSerializer.get(), FloatSerializer.get()); + HColumn<Long,Float> userForItems = new HColumnImpl<>(LongSerializer.get(), FloatSerializer.get()); userForItems.setName(userID); userForItems.setClock(now); userForItems.setValue(value); mutator.addInsertion(itemID, ITEMS_CF, userForItems); - HColumn<Long,byte[]> userIDs = new HColumnImpl<Long,byte[]>(LongSerializer.get(), BytesArraySerializer.get()); + HColumn<Long,byte[]> userIDs = new HColumnImpl<>(LongSerializer.get(), BytesArraySerializer.get()); userIDs.setName(userID); userIDs.setClock(now); userIDs.setValue(EMPTY); mutator.addInsertion(ID_ROW_KEY, USER_IDS_CF, userIDs); - HColumn<Long,byte[]> itemIDs = new HColumnImpl<Long,byte[]>(LongSerializer.get(), BytesArraySerializer.get()); + HColumn<Long,byte[]> itemIDs = new HColumnImpl<>(LongSerializer.get(), BytesArraySerializer.get()); itemIDs.setName(itemID); itemIDs.setClock(now); itemIDs.setValue(EMPTY); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractJDBCDataModel.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractJDBCDataModel.java b/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractJDBCDataModel.java index e3d4c3f..d9317c0 100644 --- a/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractJDBCDataModel.java +++ b/integration/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractJDBCDataModel.java @@ -218,7 +218,7 @@ public abstract class AbstractJDBCDataModel extends AbstractJDBCComponent implem this.cachedNumUsers = -1; this.cachedNumItems = -1; - this.itemPrefCounts = new Cache<Long,Integer>(new ItemPrefCountRetriever(getNumPreferenceForItemSQL)); + this.itemPrefCounts = new Cache<>(new ItemPrefCountRetriever(getNumPreferenceForItemSQL)); this.maxPreference = Float.NaN; this.minPreference = Float.NaN; @@ -311,7 +311,7 @@ public abstract class AbstractJDBCDataModel extends AbstractJDBCComponent implem Statement stmt = null; ResultSet rs = null; - FastByIDMap<PreferenceArray> result = new FastByIDMap<PreferenceArray>(); + FastByIDMap<PreferenceArray> result = new FastByIDMap<>(); try { conn = dataSource.getConnection(); @@ -356,7 +356,7 @@ public abstract class AbstractJDBCDataModel extends AbstractJDBCComponent implem Statement stmt = null; ResultSet rs = null; - FastByIDMap<FastIDSet> result = new FastByIDMap<FastIDSet>(); + FastByIDMap<FastIDSet> result = new FastByIDMap<>(); try { conn = dataSource.getConnection(); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/cf/taste/web/RecommenderServlet.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/cf/taste/web/RecommenderServlet.java b/integration/src/main/java/org/apache/mahout/cf/taste/web/RecommenderServlet.java index 45c1ef1..a5a89c6 100644 --- a/integration/src/main/java/org/apache/mahout/cf/taste/web/RecommenderServlet.java +++ b/integration/src/main/java/org/apache/mahout/cf/taste/web/RecommenderServlet.java @@ -102,10 +102,8 @@ public final class RecommenderServlet extends HttpServlet { } else { throw new ServletException("Bad format parameter: " + format); } - } catch (TasteException te) { + } catch (TasteException | IOException te) { throw new ServletException(te); - } catch (IOException ioe) { - throw new ServletException(ioe); } } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/clustering/cdbw/CDbwEvaluator.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/clustering/cdbw/CDbwEvaluator.java b/integration/src/main/java/org/apache/mahout/clustering/cdbw/CDbwEvaluator.java index 7a4c211..972dec1 100644 --- a/integration/src/main/java/org/apache/mahout/clustering/cdbw/CDbwEvaluator.java +++ b/integration/src/main/java/org/apache/mahout/clustering/cdbw/CDbwEvaluator.java @@ -211,11 +211,11 @@ public final class CDbwEvaluator { if (interClusterDensities != null) { return interClusterDensities; } - interClusterDensities = new TreeMap<Integer,Map<Integer,Double>>(); + interClusterDensities = new TreeMap<>(); // find the closest representative points between the clusters for (int i = 0; i < clusters.size(); i++) { int cI = clusters.get(i).getId(); - Map<Integer,Double> map = new TreeMap<Integer,Double>(); + Map<Integer,Double> map = new TreeMap<>(); interClusterDensities.put(cI, map); for (int j = i + 1; j < clusters.size(); j++) { int cJ = clusters.get(j).getId(); @@ -324,12 +324,12 @@ public final class CDbwEvaluator { if (minimumDistances != null) { return minimumDistances; } - minimumDistances = new TreeMap<Integer,Map<Integer,Double>>(); - closestRepPointIndices = new TreeMap<Integer,Map<Integer,int[]>>(); + minimumDistances = new TreeMap<>(); + closestRepPointIndices = new TreeMap<>(); for (int i = 0; i < clusters.size(); i++) { Integer cI = clusters.get(i).getId(); - Map<Integer,Double> map = new TreeMap<Integer,Double>(); - Map<Integer,int[]> treeMap = new TreeMap<Integer,int[]>(); + Map<Integer,Double> map = new TreeMap<>(); + Map<Integer,int[]> treeMap = new TreeMap<>(); closestRepPointIndices.put(cI, treeMap); minimumDistances.put(cI, map); List<VectorWritable> closRepI = representativePoints.get(cI); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/clustering/conversion/InputMapper.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/clustering/conversion/InputMapper.java b/integration/src/main/java/org/apache/mahout/clustering/conversion/InputMapper.java index ec8b3f5..9dbce5a 100644 --- a/integration/src/main/java/org/apache/mahout/clustering/conversion/InputMapper.java +++ b/integration/src/main/java/org/apache/mahout/clustering/conversion/InputMapper.java @@ -59,11 +59,7 @@ public class InputMapper extends Mapper<LongWritable, Text, Text, VectorWritable VectorWritable vectorWritable = new VectorWritable(result); context.write(new Text(String.valueOf(index)), vectorWritable); - } catch (InstantiationException e) { - throw new IllegalStateException(e); - } catch (IllegalAccessException e) { - throw new IllegalStateException(e); - } catch (InvocationTargetException e) { + } catch (InstantiationException | IllegalAccessException | InvocationTargetException e) { throw new IllegalStateException(e); } } @@ -77,9 +73,7 @@ public class InputMapper extends Mapper<LongWritable, Text, Text, VectorWritable try { Class<? extends Vector> outputClass = conf.getClassByName(vectorImplClassName).asSubclass(Vector.class); constructor = outputClass.getConstructor(int.class); - } catch (NoSuchMethodException e) { - throw new IllegalStateException(e); - } catch (ClassNotFoundException e) { + } catch (NoSuchMethodException | ClassNotFoundException e) { throw new IllegalStateException(e); } } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/clustering/evaluation/ClusterEvaluator.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/clustering/evaluation/ClusterEvaluator.java b/integration/src/main/java/org/apache/mahout/clustering/evaluation/ClusterEvaluator.java index 30f1a97..3cd06eb 100644 --- a/integration/src/main/java/org/apache/mahout/clustering/evaluation/ClusterEvaluator.java +++ b/integration/src/main/java/org/apache/mahout/clustering/evaluation/ClusterEvaluator.java @@ -129,7 +129,7 @@ public class ClusterEvaluator { * @return a Map<Integer, Vector> */ public Map<Integer,Vector> interClusterDistances() { - Map<Integer,Vector> distances = new TreeMap<Integer,Vector>(); + Map<Integer,Vector> distances = new TreeMap<>(); for (int i = 0; i < clusters.size(); i++) { Cluster clusterI = clusters.get(i); RandomAccessSparseVector row = new RandomAccessSparseVector(Integer.MAX_VALUE); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/clustering/lda/LDAPrintTopics.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/clustering/lda/LDAPrintTopics.java b/integration/src/main/java/org/apache/mahout/clustering/lda/LDAPrintTopics.java index 32d2421..cb8d935 100644 --- a/integration/src/main/java/org/apache/mahout/clustering/lda/LDAPrintTopics.java +++ b/integration/src/main/java/org/apache/mahout/clustering/lda/LDAPrintTopics.java @@ -147,7 +147,7 @@ public final class LDAPrintTopics { q.poll(); } if (q.size() < numWordsToPrint) { - q.add(new Pair<String,Double>(word, score)); + q.add(new Pair<>(word, score)); } } @@ -218,10 +218,10 @@ public final class LDAPrintTopics { } for (int i = 0; i < queues.size(); i++) { Queue<Pair<String,Double>> queue = queues.get(i); - Queue<Pair<String,Double>> newQueue = new PriorityQueue<Pair<String, Double>>(queue.size()); + Queue<Pair<String,Double>> newQueue = new PriorityQueue<>(queue.size()); double norm = expSums.get(i); for (Pair<String,Double> pair : queue) { - newQueue.add(new Pair<String,Double>(pair.getFirst(), Math.exp(pair.getSecond()) / norm)); + newQueue.add(new Pair<>(pair.getFirst(), Math.exp(pair.getSecond()) / norm)); } queues.set(i, newQueue); } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/text/MultipleTextFileInputFormat.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/text/MultipleTextFileInputFormat.java b/integration/src/main/java/org/apache/mahout/text/MultipleTextFileInputFormat.java index 32aa4a9..44df006 100644 --- a/integration/src/main/java/org/apache/mahout/text/MultipleTextFileInputFormat.java +++ b/integration/src/main/java/org/apache/mahout/text/MultipleTextFileInputFormat.java @@ -40,7 +40,7 @@ public class MultipleTextFileInputFormat extends CombineFileInputFormat<IntWrita public RecordReader<IntWritable, BytesWritable> createRecordReader(InputSplit inputSplit, TaskAttemptContext taskAttemptContext) throws IOException { - return new CombineFileRecordReader<IntWritable, BytesWritable>((CombineFileSplit) inputSplit, - taskAttemptContext, WholeFileRecordReader.class); + return new CombineFileRecordReader<>((CombineFileSplit) inputSplit, + taskAttemptContext, WholeFileRecordReader.class); } } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/text/PrefixAdditionFilter.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/text/PrefixAdditionFilter.java b/integration/src/main/java/org/apache/mahout/text/PrefixAdditionFilter.java index a13341b..9c0bc11 100644 --- a/integration/src/main/java/org/apache/mahout/text/PrefixAdditionFilter.java +++ b/integration/src/main/java/org/apache/mahout/text/PrefixAdditionFilter.java @@ -48,7 +48,7 @@ public final class PrefixAdditionFilter extends SequenceFilesFromDirectoryFilter protected void process(FileStatus fst, Path current) throws IOException { FileSystem fs = getFs(); ChunkedWriter writer = getWriter(); - if (fst.isDir()) { + if (fst.isDirectory()) { String dirPath = getPrefix() + Path.SEPARATOR + current.getName() + Path.SEPARATOR + fst.getPath().getName(); fs.listStatus(fst.getPath(), new PrefixAdditionFilter(getConf(), dirPath, getOptions(), writer, getCharset(), fs)); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/text/ReadOnlyFileSystemDirectory.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/text/ReadOnlyFileSystemDirectory.java b/integration/src/main/java/org/apache/mahout/text/ReadOnlyFileSystemDirectory.java index e97e35b..18c1252 100644 --- a/integration/src/main/java/org/apache/mahout/text/ReadOnlyFileSystemDirectory.java +++ b/integration/src/main/java/org/apache/mahout/text/ReadOnlyFileSystemDirectory.java @@ -79,7 +79,7 @@ public class ReadOnlyFileSystemDirectory extends BaseDirectory { try { FileStatus status = fs.getFileStatus(directory); if (status != null) { - isDir = status.isDir(); + isDir = status.isDirectory(); } } catch (IOException e) { log.error(e.getMessage(), e); @@ -99,7 +99,7 @@ public class ReadOnlyFileSystemDirectory extends BaseDirectory { try { FileStatus status = fs.getFileStatus(directory); if (status != null) { - isDir = status.isDir(); + isDir = status.isDirectory(); } } catch (IOException e) { log.error(e.getMessage(), e); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java b/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java index 9d08984..720078c 100644 --- a/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java +++ b/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java @@ -142,9 +142,7 @@ public class SequenceFilesFromDirectory extends AbstractJob { if (!StringUtils.isBlank(fileFilterClassName) && !PrefixAdditionFilter.class.getName().equals(fileFilterClassName)) { try { pathFilter = (PathFilter) Class.forName(fileFilterClassName).newInstance(); - } catch (InstantiationException e) { - throw new IllegalStateException(e); - } catch (IllegalAccessException e) { + } catch (InstantiationException | IllegalAccessException e) { throw new IllegalStateException(e); } } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromLuceneStorageMRJob.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromLuceneStorageMRJob.java b/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromLuceneStorageMRJob.java index 6c80aa0..5b45c48 100644 --- a/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromLuceneStorageMRJob.java +++ b/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromLuceneStorageMRJob.java @@ -58,11 +58,7 @@ public class SequenceFilesFromLuceneStorageMRJob { job.setNumReduceTasks(0); job.waitForCompletion(true); - } catch (IOException e) { - throw new RuntimeException(e); - } catch (InterruptedException e) { - throw new RuntimeException(e); - } catch (ClassNotFoundException e) { + } catch (IOException | InterruptedException | ClassNotFoundException e) { throw new RuntimeException(e); } } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java b/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java index 8b135bc..30c2a47 100644 --- a/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java +++ b/integration/src/main/java/org/apache/mahout/text/SequenceFilesFromMailArchives.java @@ -105,9 +105,9 @@ public final class SequenceFilesFromMailArchives extends AbstractJob { private static final Comparator<File> FILE_COMPARATOR = new CompositeFileComparator( DirectoryFileComparator.DIRECTORY_REVERSE, PathFileComparator.PATH_COMPARATOR); - private final Deque<MailProcessor> processors = new ArrayDeque<MailProcessor>(); + private final Deque<MailProcessor> processors = new ArrayDeque<>(); private final ChunkedWriter writer; - private final Deque<Long> messageCounts = new ArrayDeque<Long>(); + private final Deque<Long> messageCounts = new ArrayDeque<>(); public PrefixAdditionDirectoryWalker(MailProcessor processor, ChunkedWriter writer) { processors.addFirst(processor); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/text/WholeFileRecordReader.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/text/WholeFileRecordReader.java b/integration/src/main/java/org/apache/mahout/text/WholeFileRecordReader.java index 7489842..fdb3654 100644 --- a/integration/src/main/java/org/apache/mahout/text/WholeFileRecordReader.java +++ b/integration/src/main/java/org/apache/mahout/text/WholeFileRecordReader.java @@ -82,11 +82,7 @@ public class WholeFileRecordReader extends RecordReader<IntWritable, BytesWritab if (!StringUtils.isBlank(fileFilterClassName) && !PrefixAdditionFilter.class.getName().equals(fileFilterClassName)) { try { pathFilter = (PathFilter) Class.forName(fileFilterClassName).newInstance(); - } catch (ClassNotFoundException e) { - throw new IllegalStateException(e); - } catch (InstantiationException e) { - throw new IllegalStateException(e); - } catch (IllegalAccessException e) { + } catch (ClassNotFoundException | InstantiationException | IllegalAccessException e) { throw new IllegalStateException(e); } } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/text/WikipediaToSequenceFile.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/text/WikipediaToSequenceFile.java b/integration/src/main/java/org/apache/mahout/text/WikipediaToSequenceFile.java index 19f353c..1cde4cd 100644 --- a/integration/src/main/java/org/apache/mahout/text/WikipediaToSequenceFile.java +++ b/integration/src/main/java/org/apache/mahout/text/WikipediaToSequenceFile.java @@ -128,13 +128,7 @@ public final class WikipediaToSequenceFile { } runJob(inputPath, outputPath, catFile, cmdLine.hasOption(exactMatchOpt), all, removeLabels); - } catch (OptionException e) { - log.error("Exception", e); - CommandLineUtil.printHelp(group); - } catch (InterruptedException e) { - log.error("Exception", e); - CommandLineUtil.printHelp(group); - } catch (ClassNotFoundException e) { + } catch (OptionException | InterruptedException | ClassNotFoundException e) { log.error("Exception", e); CommandLineUtil.printHelp(group); } @@ -182,7 +176,7 @@ public final class WikipediaToSequenceFile { } Stringifier<Set<String>> setStringifier = - new DefaultStringifier<Set<String>>(conf, GenericsUtil.getClass(categories)); + new DefaultStringifier<>(conf, GenericsUtil.getClass(categories)); String categoriesStr = setStringifier.toString(categories); conf.set("wikipedia.categories", categoriesStr); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/utils/MatrixDumper.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/utils/MatrixDumper.java b/integration/src/main/java/org/apache/mahout/utils/MatrixDumper.java index 67cace4..8ab57be 100644 --- a/integration/src/main/java/org/apache/mahout/utils/MatrixDumper.java +++ b/integration/src/main/java/org/apache/mahout/utils/MatrixDumper.java @@ -68,7 +68,7 @@ public final class MatrixDumper extends AbstractJob { private static void exportCSV(Path inputPath, String outputFile, boolean doLabels) throws IOException { SequenceFileValueIterator<MatrixWritable> it = - new SequenceFileValueIterator<MatrixWritable>(inputPath, true, new Configuration()); + new SequenceFileValueIterator<>(inputPath, true, new Configuration()); Matrix m = it.next().get(); it.close(); PrintStream ps = getPrintStream(outputFile); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/utils/SequenceFileDumper.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/utils/SequenceFileDumper.java b/integration/src/main/java/org/apache/mahout/utils/SequenceFileDumper.java index a337be8..bf6691f 100644 --- a/integration/src/main/java/org/apache/mahout/utils/SequenceFileDumper.java +++ b/integration/src/main/java/org/apache/mahout/utils/SequenceFileDumper.java @@ -17,20 +17,18 @@ package org.apache.mahout.utils; -import java.io.File; -import java.io.OutputStreamWriter; -import java.io.Writer; -import java.util.List; - import com.google.common.base.Charsets; import com.google.common.collect.Lists; import com.google.common.io.Closeables; import com.google.common.io.Files; +import java.io.File; +import java.io.OutputStreamWriter; +import java.io.Writer; +import java.util.List; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.FileUtil; import org.apache.hadoop.fs.Path; -import org.apache.hadoop.io.Writable; import org.apache.mahout.common.AbstractJob; import org.apache.mahout.common.Pair; import org.apache.mahout.common.iterator.sequencefile.PathFilters; @@ -64,7 +62,7 @@ public final class SequenceFileDumper extends AbstractJob { Configuration conf = new Configuration(); Path input = getInputPath(); FileSystem fs = input.getFileSystem(conf); - if (fs.getFileStatus(input).isDir()) { + if (fs.getFileStatus(input).isDirectory()) { pathArr = FileUtil.stat2Paths(fs.listStatus(input, PathFilters.logsCRCFilter())); } else { pathArr = new Path[1]; @@ -92,14 +90,14 @@ public final class SequenceFileDumper extends AbstractJob { sub = Integer.parseInt(getOption("substring")); } boolean countOnly = hasOption("count"); - SequenceFileIterator<?, ?> iterator = new SequenceFileIterator<Writable, Writable>(path, true, conf); + SequenceFileIterator<?, ?> iterator = new SequenceFileIterator<>(path, true, conf); if (!hasOption("quiet")) { writer.append("Key class: ").append(iterator.getKeyClass().toString()); writer.append(" Value Class: ").append(iterator.getValueClass().toString()).append('\n'); } OpenObjectIntHashMap<String> facets = null; if (hasOption("facets")) { - facets = new OpenObjectIntHashMap<String>(); + facets = new OpenObjectIntHashMap<>(); } long count = 0; if (countOnly) { http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/utils/SplitInput.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/utils/SplitInput.java b/integration/src/main/java/org/apache/mahout/utils/SplitInput.java index 83ff3c6..834d5cd 100644 --- a/integration/src/main/java/org/apache/mahout/utils/SplitInput.java +++ b/integration/src/main/java/org/apache/mahout/utils/SplitInput.java @@ -289,7 +289,7 @@ public class SplitInput extends AbstractJob { if (fs.getFileStatus(inputDir) == null) { throw new IOException(inputDir + " does not exist"); } - if (!fs.getFileStatus(inputDir).isDir()) { + if (!fs.getFileStatus(inputDir).isDirectory()) { throw new IOException(inputDir + " is not a directory"); } @@ -300,7 +300,7 @@ public class SplitInput extends AbstractJob { // input dir contains one file per category. FileStatus[] fileStats = fs.listStatus(inputDir, PathFilters.logsCRCFilter()); for (FileStatus inputFile : fileStats) { - if (!inputFile.isDir()) { + if (!inputFile.isDirectory()) { splitFile(inputFile.getPath()); } } @@ -317,7 +317,7 @@ public class SplitInput extends AbstractJob { if (fs.getFileStatus(inputFile) == null) { throw new IOException(inputFile + " does not exist"); } - if (fs.getFileStatus(inputFile).isDir()) { + if (fs.getFileStatus(inputFile).isDirectory()) { throw new IOException(inputFile + " is a directory"); } @@ -419,7 +419,7 @@ public class SplitInput extends AbstractJob { } } else { SequenceFileIterator<Writable, Writable> iterator = - new SequenceFileIterator<Writable, Writable>(inputFile, false, fs.getConf()); + new SequenceFileIterator<>(inputFile, false, fs.getConf()); SequenceFile.Writer trainingWriter = SequenceFile.createWriter(fs, fs.getConf(), trainingOutputFile, iterator.getKeyClass(), iterator.getValueClass()); SequenceFile.Writer testWriter = SequenceFile.createWriter(fs, fs.getConf(), testOutputFile, @@ -650,10 +650,10 @@ public class SplitInput extends AbstractJob { Configuration conf = getConf(); FileSystem fs = trainingOutputDirectory.getFileSystem(conf); FileStatus trainingOutputDirStatus = fs.getFileStatus(trainingOutputDirectory); - Preconditions.checkArgument(trainingOutputDirStatus != null && trainingOutputDirStatus.isDir(), + Preconditions.checkArgument(trainingOutputDirStatus != null && trainingOutputDirStatus.isDirectory(), "%s is not a directory", trainingOutputDirectory); FileStatus testOutputDirStatus = fs.getFileStatus(testOutputDirectory); - Preconditions.checkArgument(testOutputDirStatus != null && testOutputDirStatus.isDir(), + Preconditions.checkArgument(testOutputDirStatus != null && testOutputDirStatus.isDirectory(), "%s is not a directory", testOutputDirectory); } } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/utils/SplitInputJob.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/utils/SplitInputJob.java b/integration/src/main/java/org/apache/mahout/utils/SplitInputJob.java index ad8c977..825f7a5 100644 --- a/integration/src/main/java/org/apache/mahout/utils/SplitInputJob.java +++ b/integration/src/main/java/org/apache/mahout/utils/SplitInputJob.java @@ -87,7 +87,7 @@ public final class SplitInputJob { FileSystem fs = FileSystem.get(initialConf); SequenceFileDirIterator<? extends WritableComparable, Writable> iterator = - new SequenceFileDirIterator<WritableComparable, Writable>(inputPath, + new SequenceFileDirIterator<>(inputPath, PathType.LIST, PathFilters.partFilter(), null, false, fs.getConf()); Class<? extends WritableComparable> keyClass; Class<? extends Writable> valueClass; http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/utils/clustering/AbstractClusterWriter.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/utils/clustering/AbstractClusterWriter.java b/integration/src/main/java/org/apache/mahout/utils/clustering/AbstractClusterWriter.java index e654891..ac884d0 100644 --- a/integration/src/main/java/org/apache/mahout/utils/clustering/AbstractClusterWriter.java +++ b/integration/src/main/java/org/apache/mahout/utils/clustering/AbstractClusterWriter.java @@ -142,7 +142,7 @@ public abstract class AbstractClusterWriter implements ClusterWriter { log.error("Dictionary entry missing for {}", index); continue; } - topTerms.add(new Pair<String, Double>(dictTerm, vectorTerms.get(i).weight)); + topTerms.add(new Pair<>(dictTerm, vectorTerms.get(i).weight)); } return topTerms; http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/utils/vectors/VectorDumper.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/utils/vectors/VectorDumper.java b/integration/src/main/java/org/apache/mahout/utils/vectors/VectorDumper.java index 51a22fb..2a8a42b 100644 --- a/integration/src/main/java/org/apache/mahout/utils/vectors/VectorDumper.java +++ b/integration/src/main/java/org/apache/mahout/utils/vectors/VectorDumper.java @@ -97,7 +97,7 @@ public final class VectorDumper extends AbstractJob { FileSystem fs = FileSystem.get(conf); Path input = getInputPath(); FileStatus fileStatus = fs.getFileStatus(input); - if (fileStatus.isDir()) { + if (fileStatus.isDirectory()) { pathArr = FileUtil.stat2Paths(fs.listStatus(input, PathFilters.logsCRCFilter())); } else { FileStatus[] inputPaths = fs.globStatus(input); @@ -186,8 +186,7 @@ public final class VectorDumper extends AbstractJob { if (quiet) { log.info("Processing file '{}' ({}/{})", path, ++fileCount, pathArr.length); } - SequenceFileIterable<Writable, Writable> iterable = - new SequenceFileIterable<Writable, Writable>(path, true, conf); + SequenceFileIterable<Writable, Writable> iterable = new SequenceFileIterable<>(path, true, conf); Iterator<Pair<Writable, Writable>> iterator = iterable.iterator(); long i = 0; while (iterator.hasNext() && (numItems == null || itemCount < numItems)) { http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/utils/vectors/VectorHelper.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/utils/vectors/VectorHelper.java b/integration/src/main/java/org/apache/mahout/utils/vectors/VectorHelper.java index 00fbf33..29b02aa 100644 --- a/integration/src/main/java/org/apache/mahout/utils/vectors/VectorHelper.java +++ b/integration/src/main/java/org/apache/mahout/utils/vectors/VectorHelper.java @@ -91,7 +91,7 @@ public final class VectorHelper { maxEntries = sizeOfNonZeroElementsInVector; } - PriorityQueue<Pair<Integer, Double>> queue = new TDoublePQ<Integer>(-1, maxEntries); + PriorityQueue<Pair<Integer, Double>> queue = new TDoublePQ<>(-1, maxEntries); for (Element e : vector.nonZeroes()) { queue.insertWithOverflow(Pair.of(e.index(), e.get())); } @@ -177,11 +177,8 @@ public final class VectorHelper { * </pre> */ public static String[] loadTermDictionary(File dictFile) throws IOException { - InputStream in = new FileInputStream(dictFile); - try { + try (InputStream in = new FileInputStream(dictFile)) { return loadTermDictionary(in); - } finally { - in.close(); } } @@ -192,7 +189,7 @@ public final class VectorHelper { * @param filePattern <PATH TO DICTIONARY>/dictionary.file-* */ public static String[] loadTermDictionary(Configuration conf, String filePattern) { - OpenObjectIntHashMap<String> dict = new OpenObjectIntHashMap<String>(); + OpenObjectIntHashMap<String> dict = new OpenObjectIntHashMap<>(); int maxIndexValue = 0; for (Pair<Text, IntWritable> record : new SequenceFileDirIterable<Text, IntWritable>(new Path(filePattern), PathType.GLOB, null, null, true, http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/main/java/org/apache/mahout/utils/vectors/lucene/ClusterLabels.java ---------------------------------------------------------------------- diff --git a/integration/src/main/java/org/apache/mahout/utils/vectors/lucene/ClusterLabels.java b/integration/src/main/java/org/apache/mahout/utils/vectors/lucene/ClusterLabels.java index 8fe4f65..06bec60 100644 --- a/integration/src/main/java/org/apache/mahout/utils/vectors/lucene/ClusterLabels.java +++ b/integration/src/main/java/org/apache/mahout/utils/vectors/lucene/ClusterLabels.java @@ -187,7 +187,7 @@ public class ClusterLabels { */ Terms t = MultiFields.getTerms(reader, contentField); TermsEnum te = t.iterator(null); - Map<String, TermEntry> termEntryMap = new LinkedHashMap<String, TermEntry>(); + Map<String, TermEntry> termEntryMap = new LinkedHashMap<>(); Bits liveDocs = MultiFields.getLiveDocs(reader); //WARNING: returns null if there are no deletions @@ -246,7 +246,7 @@ public class ClusterLabels { Set<String> idFieldSelector = null; if (idField != null) { - idFieldSelector = new TreeSet<String>(); + idFieldSelector = new TreeSet<>(); idFieldSelector.add(idField); } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/test/java/org/apache/mahout/text/LuceneSegmentRecordReaderTest.java ---------------------------------------------------------------------- diff --git a/integration/src/test/java/org/apache/mahout/text/LuceneSegmentRecordReaderTest.java b/integration/src/test/java/org/apache/mahout/text/LuceneSegmentRecordReaderTest.java index 6d0906b..28f2ac8 100644 --- a/integration/src/test/java/org/apache/mahout/text/LuceneSegmentRecordReaderTest.java +++ b/integration/src/test/java/org/apache/mahout/text/LuceneSegmentRecordReaderTest.java @@ -43,7 +43,8 @@ public class LuceneSegmentRecordReaderTest extends AbstractLuceneStorageTest { @Before public void before() throws IOException, InterruptedException { - LuceneStorageConfiguration lucene2SeqConf = new LuceneStorageConfiguration(getConfiguration(), asList(getIndexPath1()), new Path("output"), ID_FIELD, asList(FIELD)); + LuceneStorageConfiguration lucene2SeqConf = new LuceneStorageConfiguration(getConfiguration(), + asList(getIndexPath1()), new Path("output"), ID_FIELD, asList(FIELD)); configuration = lucene2SeqConf.serialize(); recordReader = new LuceneSegmentRecordReader(); commitDocuments(getDirectory(getIndexPath1AsFile()), docs.subList(0, 500)); @@ -61,13 +62,17 @@ public class LuceneSegmentRecordReaderTest extends AbstractLuceneStorageTest { public void testKey() throws Exception { for (SegmentCommitInfo segmentInfo : segmentInfos) { int docId = 0; - LuceneSegmentInputSplit inputSplit = new LuceneSegmentInputSplit(getIndexPath1(), segmentInfo.info.name, segmentInfo.sizeInBytes()); + LuceneSegmentInputSplit inputSplit = new LuceneSegmentInputSplit(getIndexPath1(), + segmentInfo.info.name, segmentInfo.sizeInBytes()); TaskAttemptContext context = getTaskAttemptContext(configuration, new TaskAttemptID()); recordReader.initialize(inputSplit, context); for (int i = 0; i < 500; i++){ recordReader.nextKeyValue(); - //we can't be sure of the order we are getting the segments, so we have to fudge here a bit on the id, but it is either id: i or i + 500 - assertTrue("i = " + i + " docId= " + docId, String.valueOf(docId).equals(recordReader.getCurrentKey().toString()) || String.valueOf(docId+500).equals(recordReader.getCurrentKey().toString())); + //we can't be sure of the order we are getting the segments, so we have to fudge here a bit on the id, + // but it is either id: i or i + 500 + assertTrue("i = " + i + " docId= " + + docId, String.valueOf(docId).equals(recordReader.getCurrentKey().toString()) || + String.valueOf(docId+500).equals(recordReader.getCurrentKey().toString())); assertEquals(NullWritable.get(), recordReader.getCurrentValue()); docId++; } @@ -76,18 +81,22 @@ public class LuceneSegmentRecordReaderTest extends AbstractLuceneStorageTest { @Test(expected = IllegalArgumentException.class) public void testNonExistingIdField() throws Exception { - configuration = new LuceneStorageConfiguration(getConfiguration(), asList(getIndexPath1()), new Path("output"), "nonExistingId", asList(FIELD)).serialize(); + configuration = new LuceneStorageConfiguration(getConfiguration(), + asList(getIndexPath1()), new Path("output"), "nonExistingId", asList(FIELD)).serialize(); SegmentCommitInfo segmentInfo = segmentInfos.iterator().next(); - LuceneSegmentInputSplit inputSplit = new LuceneSegmentInputSplit(getIndexPath1(), segmentInfo.info.name, segmentInfo.sizeInBytes()); + LuceneSegmentInputSplit inputSplit = new LuceneSegmentInputSplit(getIndexPath1(), + segmentInfo.info.name, segmentInfo.sizeInBytes()); TaskAttemptContext context = getTaskAttemptContext(configuration, new TaskAttemptID()); recordReader.initialize(inputSplit, context); } @Test(expected = IllegalArgumentException.class) public void testNonExistingField() throws Exception { - configuration = new LuceneStorageConfiguration(getConfiguration(), asList(getIndexPath1()), new Path("output"), ID_FIELD, asList("nonExistingField")).serialize(); + configuration = new LuceneStorageConfiguration(getConfiguration(), asList(getIndexPath1()), + new Path("output"), ID_FIELD, asList("nonExistingField")).serialize(); SegmentCommitInfo segmentInfo = segmentInfos.iterator().next(); - LuceneSegmentInputSplit inputSplit = new LuceneSegmentInputSplit(getIndexPath1(), segmentInfo.info.name, segmentInfo.sizeInBytes()); + LuceneSegmentInputSplit inputSplit = new LuceneSegmentInputSplit(getIndexPath1(), + segmentInfo.info.name, segmentInfo.sizeInBytes()); TaskAttemptContext context = getTaskAttemptContext(configuration, new TaskAttemptID()); recordReader.initialize(inputSplit, context); } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/test/java/org/apache/mahout/text/TestSequenceFilesFromDirectory.java ---------------------------------------------------------------------- diff --git a/integration/src/test/java/org/apache/mahout/text/TestSequenceFilesFromDirectory.java b/integration/src/test/java/org/apache/mahout/text/TestSequenceFilesFromDirectory.java index ab197c2..aec5e39 100644 --- a/integration/src/test/java/org/apache/mahout/text/TestSequenceFilesFromDirectory.java +++ b/integration/src/test/java/org/apache/mahout/text/TestSequenceFilesFromDirectory.java @@ -219,7 +219,7 @@ public final class TestSequenceFilesFromDirectory extends MahoutTestCase { // read a chunk to check content SequenceFileIterator<Text, Text> iterator = - new SequenceFileIterator<Text, Text>(fileStatuses[0].getPath(), true, configuration); + new SequenceFileIterator<>(fileStatuses[0].getPath(), true, configuration); try { while (iterator.hasNext()) { Pair<Text, Text> record = iterator.next(); @@ -254,7 +254,8 @@ public final class TestSequenceFilesFromDirectory extends MahoutTestCase { } // read a chunk to check content - SequenceFileIterator<Text, Text> iterator = new SequenceFileIterator<Text, Text>(fileStatuses[0].getPath(), true, configuration); + SequenceFileIterator<Text, Text> iterator = + new SequenceFileIterator<>(fileStatuses[0].getPath(), true, configuration); try { while (iterator.hasNext()) { Pair<Text, Text> record = iterator.next(); @@ -285,7 +286,7 @@ public final class TestSequenceFilesFromDirectory extends MahoutTestCase { } // read a chunk to check content - SequenceFileIterator<Text, Text> iterator = new SequenceFileIterator<Text, Text>( + SequenceFileIterator<Text, Text> iterator = new SequenceFileIterator<>( fileStatuses[0].getPath(), true, conf); try { while (iterator.hasNext()) { @@ -318,7 +319,7 @@ public final class TestSequenceFilesFromDirectory extends MahoutTestCase { } // read a chunk to check content - SequenceFileIterator<Text, Text> iterator = new SequenceFileIterator<Text, Text>( + SequenceFileIterator<Text, Text> iterator = new SequenceFileIterator<>( fileStatuses[0].getPath(), true, configuration); try { while (iterator.hasNext()) { http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/integration/src/test/java/org/apache/mahout/utils/TestConcatenateVectorsJob.java ---------------------------------------------------------------------- diff --git a/integration/src/test/java/org/apache/mahout/utils/TestConcatenateVectorsJob.java b/integration/src/test/java/org/apache/mahout/utils/TestConcatenateVectorsJob.java index c0fcb89..329b590 100644 --- a/integration/src/test/java/org/apache/mahout/utils/TestConcatenateVectorsJob.java +++ b/integration/src/test/java/org/apache/mahout/utils/TestConcatenateVectorsJob.java @@ -57,7 +57,7 @@ public class TestConcatenateVectorsJob extends MahoutTestCase { // Yes, all of this generic rigmarole is needed, and woe betide he who changes it ConcatenateVectorsReducer reducer = new ConcatenateVectorsReducer(); - DummyRecordWriter<IntWritable, VectorWritable> recordWriter = new DummyRecordWriter<IntWritable, VectorWritable>(); + DummyRecordWriter<IntWritable, VectorWritable> recordWriter = new DummyRecordWriter<>(); Reducer<IntWritable, VectorWritable, IntWritable, VectorWritable>.Context reduceContext = DummyRecordWriter.build(reducer, configuration, recordWriter, IntWritable.class, VectorWritable.class); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/common/Refreshable.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/common/Refreshable.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/common/Refreshable.java index b4fe9ca..9b26bee 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/common/Refreshable.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/common/Refreshable.java @@ -44,7 +44,7 @@ public interface Refreshable { * * @param alreadyRefreshed * {@link org.apache.mahout.cf.taste.common.Refreshable}s that are known to have already been - * refreshed as a result of an initial call to a {@link #refresh(Collection)} method on some + * refreshed as a result of an initial call to a {#refresh(Collection)} method on some * object. This ensure that objects in a refresh dependency graph aren't refreshed twice * needlessly. */ http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ALS.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ALS.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ALS.java index 3ad5c73..c5ccf38 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ALS.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ALS.java @@ -35,7 +35,6 @@ import org.apache.mahout.common.iterator.sequencefile.SequenceFileDirValueIterat import org.apache.mahout.math.Vector; import org.apache.mahout.math.VectorWritable; import org.apache.mahout.math.als.AlternatingLeastSquaresSolver; -import org.apache.mahout.math.function.Functions; import org.apache.mahout.math.map.OpenIntObjectHashMap; import java.io.IOException; @@ -47,7 +46,7 @@ final class ALS { private ALS() {} static Vector readFirstRow(Path dir, Configuration conf) throws IOException { - Iterator<VectorWritable> iterator = new SequenceFileDirValueIterator<VectorWritable>(dir, PathType.LIST, + Iterator<VectorWritable> iterator = new SequenceFileDirValueIterator<>(dir, PathType.LIST, PathFilters.partFilter(), null, true, conf); return iterator.hasNext() ? iterator.next().get() : null; } @@ -83,7 +82,7 @@ final class ALS { } public static OpenIntObjectHashMap<Vector> readMatrixByRows(Path dir, Configuration conf) { - OpenIntObjectHashMap<Vector> matrix = new OpenIntObjectHashMap<Vector>(); + OpenIntObjectHashMap<Vector> matrix = new OpenIntObjectHashMap<>(); for (Pair<IntWritable,VectorWritable> pair : new SequenceFileDirIterable<IntWritable,VectorWritable>(dir, PathType.LIST, PathFilters.partFilter(), conf)) { int rowIndex = pair.getFirst().get(); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ParallelALSFactorizationJob.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ParallelALSFactorizationJob.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ParallelALSFactorizationJob.java index 5b42353..624a8c4 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ParallelALSFactorizationJob.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/ParallelALSFactorizationJob.java @@ -102,10 +102,6 @@ public class ParallelALSFactorizationJob extends AbstractJob { private double lambda; private double alpha; private int numThreadsPerSolver; - private boolean usesLongIDs; - - private int numItems; - private int numUsers; enum Stats { NUM_USERS } @@ -138,7 +134,7 @@ public class ParallelALSFactorizationJob extends AbstractJob { implicitFeedback = Boolean.parseBoolean(getOption("implicitFeedback")); numThreadsPerSolver = Integer.parseInt(getOption("numThreadsPerSolver")); - usesLongIDs = Boolean.parseBoolean(getOption("usesLongIDs", String.valueOf(false))); + boolean usesLongIDs = Boolean.parseBoolean(getOption("usesLongIDs", String.valueOf(false))); /* * compute the factorization A = U M' @@ -196,8 +192,8 @@ public class ParallelALSFactorizationJob extends AbstractJob { Vector averageRatings = ALS.readFirstRow(getTempPath("averageRatings"), getConf()); - numItems = averageRatings.getNumNondefaultElements(); - numUsers = (int) userRatings.getCounters().findCounter(Stats.NUM_USERS).getValue(); + int numItems = averageRatings.getNumNondefaultElements(); + int numUsers = (int) userRatings.getCounters().findCounter(Stats.NUM_USERS).getValue(); log.info("Found {} users and {} items", numUsers, numItems); @@ -208,11 +204,11 @@ public class ParallelALSFactorizationJob extends AbstractJob { /* broadcast M, read A row-wise, recompute U row-wise */ log.info("Recomputing U (iteration {}/{})", currentIteration, numIterations); runSolver(pathToUserRatings(), pathToU(currentIteration), pathToM(currentIteration - 1), currentIteration, "U", - numItems); + numItems); /* broadcast U, read A' row-wise, recompute M row-wise */ log.info("Recomputing M (iteration {}/{})", currentIteration, numIterations); runSolver(pathToItemRatings(), pathToM(currentIteration), pathToU(currentIteration), currentIteration, "M", - numUsers); + numUsers); } return 0; http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/PredictionMapper.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/PredictionMapper.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/PredictionMapper.java index 6bcb500..6e7ea81 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/PredictionMapper.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/als/PredictionMapper.java @@ -63,7 +63,7 @@ public class PredictionMapper extends SharingMapper<IntWritable,VectorWritable,L OpenIntObjectHashMap<Vector> U = ALS.readMatrixByRows(pathToU, conf); OpenIntObjectHashMap<Vector> M = ALS.readMatrixByRows(pathToM, conf); - return new Pair<OpenIntObjectHashMap<Vector>, OpenIntObjectHashMap<Vector>>(U, M); + return new Pair<>(U, M); } @Override http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java index 13bfbeb..b44fd5b 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java @@ -212,10 +212,8 @@ public final class AggregateAndRecommendReducer extends return true; } else if (itemsForUser != null) { return itemsForUser.contains(itemID); - } else if (allItemsToRecommendFor != null) { - return allItemsToRecommendFor.contains(itemID); } else { - return false; + return allItemsToRecommendFor.contains(itemID); } } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/IDReader.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/IDReader.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/IDReader.java index 9dce435..b8cf6bb 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/IDReader.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/IDReader.java @@ -17,13 +17,12 @@ package org.apache.mahout.cf.taste.hadoop.item; +import com.google.common.collect.Maps; +import com.google.common.io.Closeables; import java.io.IOException; import java.io.InputStream; -import java.util.HashMap; import java.util.Map; import java.util.regex.Pattern; - -import com.google.common.collect.Maps; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.Path; import org.apache.mahout.cf.taste.impl.common.FastIDSet; @@ -32,8 +31,6 @@ import org.apache.mahout.common.iterator.FileLineIterable; import org.slf4j.Logger; import org.slf4j.LoggerFactory; -import com.google.common.io.Closeables; - /** * Reads user ids and item ids from files specified in usersFile, itemsFile or userItemFile options in item-based * recommender. Composes a list of users and a list of items which can be used by @@ -118,7 +115,7 @@ public class IDReader { } else if (isUserItemFileSpecified() && !isUsersFileSpecified()) { readUserItemFilterIfNeeded(); userIds = extractAllUserIdsFromUserItemFilter(userItemFilter); - } else if (!isUsersFileSpecified() && !isUserItemFileSpecified()) { + } else if (!isUsersFileSpecified()) { throw new IllegalStateException("Neither usersFile nor userItemFile options are specified"); } else { throw new IllegalStateException("usersFile and userItemFile options cannot be used simultaneously"); @@ -131,7 +128,7 @@ public class IDReader { } else if (isUserItemFileSpecified() && !isItemsFileSpecified()) { readUserItemFilterIfNeeded(); itemIds = extractAllItemIdsFromUserItemFilter(userItemFilter); - } else if (!isItemsFileSpecified() && !isUserItemFileSpecified()) { + } else if (!isItemsFileSpecified()) { throw new IllegalStateException("Neither itemsFile nor userItemFile options are specified"); } else { throw new IllegalStateException("itemsFile and userItemFile options cannot be specified simultaneously"); @@ -151,7 +148,7 @@ public class IDReader { try { for (String line : new FileLineIterable(in)) { try { - String[] tokens = SEPARATOR.split(line.toString()); + String[] tokens = SEPARATOR.split(line); Long userId = Long.parseLong(tokens[0]); Long itemId = Long.parseLong(tokens[1]); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/UserVectorSplitterMapper.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/UserVectorSplitterMapper.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/UserVectorSplitterMapper.java index 66a412d..2290d06 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/UserVectorSplitterMapper.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/hadoop/item/UserVectorSplitterMapper.java @@ -62,6 +62,9 @@ public final class UserVectorSplitterMapper extends VectorWritable value, Context context) throws IOException, InterruptedException { long userID = key.get(); + + log.info("UserID = {}", userID); + if (usersToRecommendFor != null && !usersToRecommendFor.contains(userID)) { return; } http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/AbstractDifferenceRecommenderEvaluator.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/AbstractDifferenceRecommenderEvaluator.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/AbstractDifferenceRecommenderEvaluator.java index b5d2f85..f6598f3 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/AbstractDifferenceRecommenderEvaluator.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/AbstractDifferenceRecommenderEvaluator.java @@ -106,9 +106,9 @@ public abstract class AbstractDifferenceRecommenderEvaluator implements Recommen log.info("Beginning evaluation using {} of {}", trainingPercentage, dataModel); int numUsers = dataModel.getNumUsers(); - FastByIDMap<PreferenceArray> trainingPrefs = new FastByIDMap<PreferenceArray>( + FastByIDMap<PreferenceArray> trainingPrefs = new FastByIDMap<>( 1 + (int) (evaluationPercentage * numUsers)); - FastByIDMap<PreferenceArray> testPrefs = new FastByIDMap<PreferenceArray>( + FastByIDMap<PreferenceArray> testPrefs = new FastByIDMap<>( 1 + (int) (evaluationPercentage * numUsers)); LongPrimitiveIterator it = dataModel.getUserIDs(); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/GenericRecommenderIRStatsEvaluator.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/GenericRecommenderIRStatsEvaluator.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/GenericRecommenderIRStatsEvaluator.java index e29125d..0e121d1 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/GenericRecommenderIRStatsEvaluator.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/eval/GenericRecommenderIRStatsEvaluator.java @@ -127,7 +127,7 @@ public final class GenericRecommenderIRStatsEvaluator implements RecommenderIRSt continue; } - FastByIDMap<PreferenceArray> trainingUsers = new FastByIDMap<PreferenceArray>(dataModel.getNumUsers()); + FastByIDMap<PreferenceArray> trainingUsers = new FastByIDMap<>(dataModel.getNumUsers()); LongPrimitiveIterator it2 = dataModel.getUserIDs(); while (it2.hasNext()) { dataSplitter.processOtherUser(userID, relevantItemIDs, trainingUsers, it2.nextLong(), dataModel); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericBooleanPrefDataModel.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericBooleanPrefDataModel.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericBooleanPrefDataModel.java index 43ccb4e..2c1ff4d 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericBooleanPrefDataModel.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericBooleanPrefDataModel.java @@ -74,7 +74,7 @@ public final class GenericBooleanPrefDataModel extends AbstractDataModel { Preconditions.checkArgument(userData != null, "userData is null"); this.preferenceFromUsers = userData; - this.preferenceForItems = new FastByIDMap<FastIDSet>(); + this.preferenceForItems = new FastByIDMap<>(); FastIDSet itemIDSet = new FastIDSet(); for (Map.Entry<Long, FastIDSet> entry : preferenceFromUsers.entrySet()) { long userID = entry.getKey(); @@ -132,7 +132,7 @@ public final class GenericBooleanPrefDataModel extends AbstractDataModel { * that user's associated items */ public static FastByIDMap<FastIDSet> toDataMap(DataModel dataModel) throws TasteException { - FastByIDMap<FastIDSet> data = new FastByIDMap<FastIDSet>(dataModel.getNumUsers()); + FastByIDMap<FastIDSet> data = new FastByIDMap<>(dataModel.getNumUsers()); LongPrimitiveIterator it = dataModel.getUserIDs(); while (it.hasNext()) { long userID = it.nextLong(); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericDataModel.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericDataModel.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericDataModel.java index a898d85..f58d349 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericDataModel.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/GenericDataModel.java @@ -81,7 +81,7 @@ public final class GenericDataModel extends AbstractDataModel { Preconditions.checkArgument(userData != null, "userData is null"); this.preferenceFromUsers = userData; - FastByIDMap<Collection<Preference>> prefsForItems = new FastByIDMap<Collection<Preference>>(); + FastByIDMap<Collection<Preference>> prefsForItems = new FastByIDMap<>(); FastIDSet itemIDSet = new FastIDSet(); int currentCount = 0; float maxPrefValue = Float.NEGATIVE_INFINITY; @@ -174,7 +174,7 @@ public final class GenericDataModel extends AbstractDataModel { * that user's preferences */ public static FastByIDMap<PreferenceArray> toDataMap(DataModel dataModel) throws TasteException { - FastByIDMap<PreferenceArray> data = new FastByIDMap<PreferenceArray>(dataModel.getNumUsers()); + FastByIDMap<PreferenceArray> data = new FastByIDMap<>(dataModel.getNumUsers()); LongPrimitiveIterator it = dataModel.getUserIDs(); while (it.hasNext()) { long userID = it.nextLong(); http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/MemoryIDMigrator.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/MemoryIDMigrator.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/MemoryIDMigrator.java index d511094..3463ff5 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/MemoryIDMigrator.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/MemoryIDMigrator.java @@ -28,7 +28,7 @@ public final class MemoryIDMigrator extends AbstractIDMigrator implements Updata private final FastByIDMap<String> longToString; public MemoryIDMigrator() { - this.longToString = new FastByIDMap<String>(100); + this.longToString = new FastByIDMap<>(100); } @Override http://git-wip-us.apache.org/repos/asf/mahout/blob/87c15bea/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/PlusAnonymousConcurrentUserDataModel.java ---------------------------------------------------------------------- diff --git a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/PlusAnonymousConcurrentUserDataModel.java b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/PlusAnonymousConcurrentUserDataModel.java index e7ccb40..c97a545 100644 --- a/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/PlusAnonymousConcurrentUserDataModel.java +++ b/mrlegacy/src/main/java/org/apache/mahout/cf/taste/impl/model/PlusAnonymousConcurrentUserDataModel.java @@ -102,8 +102,8 @@ public final class PlusAnonymousConcurrentUserDataModel extends PlusAnonymousUse public PlusAnonymousConcurrentUserDataModel(DataModel delegate, int maxConcurrentUsers) { super(delegate); - tempPrefs = new ConcurrentHashMap<Long, PreferenceArray>(); - prefItemIDs = new ConcurrentHashMap<Long, FastIDSet>(); + tempPrefs = new ConcurrentHashMap<>(); + prefItemIDs = new ConcurrentHashMap<>(); initializeUsersPools(maxConcurrentUsers); } @@ -114,7 +114,7 @@ public final class PlusAnonymousConcurrentUserDataModel extends PlusAnonymousUse * @param usersPoolSize Maximum allowed number of concurrent anonymous user. Depends on the consumer system. */ private void initializeUsersPools(int usersPoolSize) { - usersPool = new ConcurrentLinkedQueue<Long>(); + usersPool = new ConcurrentLinkedQueue<>(); for (int i = 0; i < usersPoolSize; i++) { usersPool.add(TEMP_USER_ID + i); }
