Examples with FileInputSplit org.apache.flink.core.fs.FileInputSplit used on opensource projects

Example 96 with FileInputSplit

use of org.apache.flink.core.fs.FileInputSplit in project flink by apache.

the class GenericCsvInputFormatTest method testReadNoPosFirstN.

@Test
public void testReadNoPosFirstN() throws IOException {
    try {
        final String fileContent = "111|222|333|444|555|\n666|777|888|999|000|";
        final FileInputSplit split = createTempFile(fileContent);
        final Configuration parameters = new Configuration();
        format.setFieldDelimiter("|");
        format.setFieldTypesGeneric(IntValue.class, IntValue.class);
        format.configure(parameters);
        format.open(split);
        Value[] values = createIntValues(2);
        // if this would parse all, we would get an index out of bounds exception
        values = format.nextRecord(values);
        assertNotNull(values);
        assertEquals(111, ((IntValue) values[0]).getValue());
        assertEquals(222, ((IntValue) values[1]).getValue());
        values = format.nextRecord(values);
        assertNotNull(values);
        assertEquals(666, ((IntValue) values[0]).getValue());
        assertEquals(777, ((IntValue) values[1]).getValue());
        assertNull(format.nextRecord(values));
        assertTrue(format.reachedEnd());
    } catch (Exception ex) {
        fail("Test failed due to a " + ex.getClass().getSimpleName() + ": " + ex.getMessage());
    }
}

Example 97 with FileInputSplit

use of org.apache.flink.core.fs.FileInputSplit in project flink by apache.

the class GenericCsvInputFormatTest method testReadTooShortInput.

@Test
public void testReadTooShortInput() throws IOException {
    try {
        final String fileContent = "111|222|333|444\n666|777|888|999";
        final FileInputSplit split = createTempFile(fileContent);
        final Configuration parameters = new Configuration();
        format.setFieldDelimiter("|");
        format.setFieldTypesGeneric(IntValue.class, IntValue.class, IntValue.class, IntValue.class, IntValue.class);
        format.configure(parameters);
        format.open(split);
        Value[] values = createIntValues(5);
        try {
            format.nextRecord(values);
            fail("Should have thrown a parse exception on too short input.");
        } catch (ParseException e) {
        // all is well
        }
    } catch (Exception ex) {
        fail("Test failed due to a " + ex.getClass().getSimpleName() + ": " + ex.getMessage());
    }
}

Example 98 with FileInputSplit

use of org.apache.flink.core.fs.FileInputSplit in project flink by apache.

the class ContinuousFileProcessingTest method testFileReadingOperatorWithIngestionTime.

@Test
public void testFileReadingOperatorWithIngestionTime() throws Exception {
    String testBasePath = hdfsURI + "/" + UUID.randomUUID() + "/";
    Set<org.apache.hadoop.fs.Path> filesCreated = new HashSet<>();
    Map<Integer, String> expectedFileContents = new HashMap<>();
    Map<String, Long> modTimes = new HashMap<>();
    for (int i = 0; i < NO_OF_FILES; i++) {
        Tuple2<org.apache.hadoop.fs.Path, String> file = createFileAndFillWithData(testBasePath, "file", i, "This is test line.");
        filesCreated.add(file.f0);
        modTimes.put(file.f0.getName(), hdfs.getFileStatus(file.f0).getModificationTime());
        expectedFileContents.put(i, file.f1);
    }
    TextInputFormat format = new TextInputFormat(new Path(testBasePath));
    TypeInformation<String> typeInfo = TypeExtractor.getInputFormatTypes(format);
    final long watermarkInterval = 10;
    ContinuousFileReaderOperator<String> reader = new ContinuousFileReaderOperator<>(format);
    final OneInputStreamOperatorTestHarness<TimestampedFileInputSplit, String> tester = new OneInputStreamOperatorTestHarness<>(reader);
    tester.getExecutionConfig().setAutoWatermarkInterval(watermarkInterval);
    tester.setTimeCharacteristic(TimeCharacteristic.IngestionTime);
    reader.setOutputType(typeInfo, tester.getExecutionConfig());
    tester.open();
    Assert.assertEquals(TimeCharacteristic.IngestionTime, tester.getTimeCharacteristic());
    // test that watermarks are correctly emitted
    ConcurrentLinkedQueue<Object> output = tester.getOutput();
    tester.setProcessingTime(201);
    Assert.assertTrue(output.peek() instanceof Watermark);
    Assert.assertEquals(200, ((Watermark) output.poll()).getTimestamp());
    tester.setProcessingTime(301);
    Assert.assertTrue(output.peek() instanceof Watermark);
    Assert.assertEquals(300, ((Watermark) output.poll()).getTimestamp());
    tester.setProcessingTime(401);
    Assert.assertTrue(output.peek() instanceof Watermark);
    Assert.assertEquals(400, ((Watermark) output.poll()).getTimestamp());
    tester.setProcessingTime(501);
    Assert.assertTrue(output.peek() instanceof Watermark);
    Assert.assertEquals(500, ((Watermark) output.poll()).getTimestamp());
    Assert.assertTrue(output.isEmpty());
    // create the necessary splits for the test
    FileInputSplit[] splits = format.createInputSplits(reader.getRuntimeContext().getNumberOfParallelSubtasks());
    // and feed them to the operator
    Map<Integer, List<String>> actualFileContents = new HashMap<>();
    long lastSeenWatermark = Long.MIN_VALUE;
    // counter for the lines read from the splits
    int lineCounter = 0;
    int watermarkCounter = 0;
    for (FileInputSplit split : splits) {
        // set the next "current processing time".
        long nextTimestamp = tester.getProcessingTime() + watermarkInterval;
        tester.setProcessingTime(nextTimestamp);
        // send the next split to be read and wait until it is fully read, the +1 is for the watermark.
        tester.processElement(new StreamRecord<>(new TimestampedFileInputSplit(modTimes.get(split.getPath().getName()), split.getSplitNumber(), split.getPath(), split.getStart(), split.getLength(), split.getHostnames())));
        // BUT THIS IS JUST FOR THIS TEST
        while (tester.getOutput().isEmpty() || tester.getOutput().size() != (LINES_PER_FILE + 1)) {
            Thread.sleep(10);
        }
        // verify that the results are the expected
        for (Object line : tester.getOutput()) {
            if (line instanceof StreamRecord) {
                @SuppressWarnings("unchecked") StreamRecord<String> element = (StreamRecord<String>) line;
                lineCounter++;
                Assert.assertEquals(nextTimestamp, element.getTimestamp());
                int fileIdx = Character.getNumericValue(element.getValue().charAt(0));
                List<String> content = actualFileContents.get(fileIdx);
                if (content == null) {
                    content = new ArrayList<>();
                    actualFileContents.put(fileIdx, content);
                }
                content.add(element.getValue() + "\n");
            } else if (line instanceof Watermark) {
                long watermark = ((Watermark) line).getTimestamp();
                Assert.assertEquals(nextTimestamp - (nextTimestamp % watermarkInterval), watermark);
                Assert.assertTrue(watermark > lastSeenWatermark);
                watermarkCounter++;
                lastSeenWatermark = watermark;
            } else {
                Assert.fail("Unknown element in the list.");
            }
        }
        // clean the output to be ready for the next split
        tester.getOutput().clear();
    }
    // now we are processing one split after the other,
    // so all the elements must be here by now.
    Assert.assertEquals(NO_OF_FILES * LINES_PER_FILE, lineCounter);
    // because we expect one watermark per split.
    Assert.assertEquals(splits.length, watermarkCounter);
    // then close the reader gracefully so that the Long.MAX watermark is emitted
    synchronized (tester.getCheckpointLock()) {
        tester.close();
    }
    for (org.apache.hadoop.fs.Path file : filesCreated) {
        hdfs.delete(file, false);
    }
    // check if the last element is the LongMax watermark (by now this must be the only element)
    Assert.assertEquals(1, tester.getOutput().size());
    Assert.assertTrue(tester.getOutput().peek() instanceof Watermark);
    Assert.assertEquals(Long.MAX_VALUE, ((Watermark) tester.getOutput().poll()).getTimestamp());
    // check if the elements are the expected ones.
    Assert.assertEquals(expectedFileContents.size(), actualFileContents.size());
    for (Integer fileIdx : expectedFileContents.keySet()) {
        Assert.assertTrue("file" + fileIdx + " not found", actualFileContents.keySet().contains(fileIdx));
        List<String> cntnt = actualFileContents.get(fileIdx);
        Collections.sort(cntnt, new Comparator<String>() {

            @Override
            public int compare(String o1, String o2) {
                return getLineNo(o1) - getLineNo(o2);
            }
        });
        StringBuilder cntntStr = new StringBuilder();
        for (String line : cntnt) {
            cntntStr.append(line);
        }
        Assert.assertEquals(expectedFileContents.get(fileIdx), cntntStr.toString());
    }
}

Example 99 with FileInputSplit

use of org.apache.flink.core.fs.FileInputSplit in project flink by apache.

the class ContinuousFileProcessingTest method testReaderSnapshotRestore.

@Test
public void testReaderSnapshotRestore() throws Exception {
    String testBasePath = hdfsURI + "/" + UUID.randomUUID() + "/";
    TimestampedFileInputSplit split1 = new TimestampedFileInputSplit(0, 3, new Path("test/test1"), 0, 100, null);
    TimestampedFileInputSplit split2 = new TimestampedFileInputSplit(10, 2, new Path("test/test2"), 101, 200, null);
    TimestampedFileInputSplit split3 = new TimestampedFileInputSplit(10, 1, new Path("test/test2"), 0, 100, null);
    TimestampedFileInputSplit split4 = new TimestampedFileInputSplit(11, 0, new Path("test/test3"), 0, 100, null);
    final OneShotLatch latch = new OneShotLatch();
    BlockingFileInputFormat format = new BlockingFileInputFormat(latch, new Path(testBasePath));
    TypeInformation<FileInputSplit> typeInfo = TypeExtractor.getInputFormatTypes(format);
    ContinuousFileReaderOperator<FileInputSplit> initReader = new ContinuousFileReaderOperator<>(format);
    initReader.setOutputType(typeInfo, new ExecutionConfig());
    OneInputStreamOperatorTestHarness<TimestampedFileInputSplit, FileInputSplit> initTestInstance = new OneInputStreamOperatorTestHarness<>(initReader);
    initTestInstance.setTimeCharacteristic(TimeCharacteristic.EventTime);
    initTestInstance.open();
    // create some state in the reader
    initTestInstance.processElement(new StreamRecord<>(split1));
    initTestInstance.processElement(new StreamRecord<>(split2));
    initTestInstance.processElement(new StreamRecord<>(split3));
    initTestInstance.processElement(new StreamRecord<>(split4));
    // take a snapshot of the operator's state. This will be used
    // to initialize another reader and compare the results of the
    // two operators.
    final OperatorStateHandles snapshot;
    synchronized (initTestInstance.getCheckpointLock()) {
        snapshot = initTestInstance.snapshot(0L, 0L);
    }
    ContinuousFileReaderOperator<FileInputSplit> restoredReader = new ContinuousFileReaderOperator<>(new BlockingFileInputFormat(latch, new Path(testBasePath)));
    restoredReader.setOutputType(typeInfo, new ExecutionConfig());
    OneInputStreamOperatorTestHarness<TimestampedFileInputSplit, FileInputSplit> restoredTestInstance = new OneInputStreamOperatorTestHarness<>(restoredReader);
    restoredTestInstance.setTimeCharacteristic(TimeCharacteristic.EventTime);
    restoredTestInstance.initializeState(snapshot);
    restoredTestInstance.open();
    // now let computation start
    latch.trigger();
    synchronized (initTestInstance.getCheckpointLock()) {
        initTestInstance.close();
    }
    synchronized (restoredTestInstance.getCheckpointLock()) {
        restoredTestInstance.close();
    }
    FileInputSplit fsSplit1 = createSplitFromTimestampedSplit(split1);
    FileInputSplit fsSplit2 = createSplitFromTimestampedSplit(split2);
    FileInputSplit fsSplit3 = createSplitFromTimestampedSplit(split3);
    FileInputSplit fsSplit4 = createSplitFromTimestampedSplit(split4);
    // compare if the results contain what they should contain and also if
    // they are the same, as they should.
    Assert.assertTrue(initTestInstance.getOutput().contains(new StreamRecord<>(fsSplit1)));
    Assert.assertTrue(initTestInstance.getOutput().contains(new StreamRecord<>(fsSplit2)));
    Assert.assertTrue(initTestInstance.getOutput().contains(new StreamRecord<>(fsSplit3)));
    Assert.assertTrue(initTestInstance.getOutput().contains(new StreamRecord<>(fsSplit4)));
    Assert.assertArrayEquals(initTestInstance.getOutput().toArray(), restoredTestInstance.getOutput().toArray());
}

Example 100 with FileInputSplit

use of org.apache.flink.core.fs.FileInputSplit in project flink by apache.

the class CsvInputFormatTest method testPojoSubclassType.

/**
	 * Tests that the CSV input format can deal with POJOs which are subclasses.
	 *
	 * @throws Exception
	 */
@Test
public void testPojoSubclassType() throws Exception {
    final String fileContent = "t1,foobar,tweet2\nt2,barfoo,tweet2";
    final File tempFile = File.createTempFile("CsvReaderPOJOSubclass", "tmp");
    tempFile.deleteOnExit();
    OutputStreamWriter writer = new OutputStreamWriter(new FileOutputStream(tempFile));
    writer.write(fileContent);
    writer.close();
    @SuppressWarnings("unchecked") PojoTypeInfo<TwitterPOJO> typeInfo = (PojoTypeInfo<TwitterPOJO>) TypeExtractor.createTypeInfo(TwitterPOJO.class);
    CsvInputFormat<TwitterPOJO> inputFormat = new PojoCsvInputFormat<>(new Path(tempFile.toURI().toString()), typeInfo);
    inputFormat.configure(new Configuration());
    FileInputSplit[] splits = inputFormat.createInputSplits(1);
    inputFormat.open(splits[0]);
    List<TwitterPOJO> expected = new ArrayList<>();
    for (String line : fileContent.split("\n")) {
        String[] elements = line.split(",");
        expected.add(new TwitterPOJO(elements[0], elements[1], elements[2]));
    }
    List<TwitterPOJO> actual = new ArrayList<>();
    TwitterPOJO pojo;
    while ((pojo = inputFormat.nextRecord(new TwitterPOJO())) != null) {
        actual.add(pojo);
    }
    assertEquals(expected, actual);
}