@Test
  public void testAvroScheme() {

    byte[] byteArray = {
      12, 109, 121, 112, 105, 112, 101, 14, 116, 101, 115, 116, 105, 110, 103, -96, 1, 4, 4, 105,
      100, 18, 10, 115, 99, 111, 114, 101, -56, 1, 0, 8, 14, 115, 117, 98, 106, 101, 99, 116, 16,
      114, 101, 108, 105, 103, 105, 111, 110, 16, 108, 97, 115, 116, 110, 97, 109, 101, 10, 115,
      116, 97, 109, 109, 18, 102, 105, 114, 115, 116, 110, 97, 109, 101, 14, 114, 111, 115, 97, 110,
      110, 101, 8, 100, 97, 116, 101, 20, 50, 48, 49, 52, 45, 49, 48, 45, 50, 50, 0, 0
    };

    try {
      SpecificDatumReader<InsertMutation> reader =
          new SpecificDatumReader<>(InsertMutation.getClassSchema());
      Decoder decoder = DecoderFactory.get().binaryDecoder(byteArray, null);
      InsertMutation insertMutation = reader.read(null, decoder);

      LOG.info("ENTRY: " + insertMutation.toString());

      LOG.info("DATABASE NAME: " + insertMutation.getDatabase());
      assertEquals("mypipe", insertMutation.getDatabase().toString());

      LOG.info("TABLE NAME: " + insertMutation.getTable());
      assertEquals("testing", insertMutation.getTable().toString());

      LOG.info("ID: " + AvroSchemaUtils.getIntegerValueByKey(insertMutation.getIntegers(), "id"));
      assertEquals(
          9, (int) AvroSchemaUtils.getIntegerValueByKey(insertMutation.getIntegers(), "id"));

      LOG.info(
          "FIRST NAME: "
              + AvroSchemaUtils.getStringValueByKey(insertMutation.getStrings(), "firstname"));
      assertEquals(
          "rosanne", AvroSchemaUtils.getStringValueByKey(insertMutation.getStrings(), "firstname"));

      LOG.info(
          "LAST NAME: "
              + AvroSchemaUtils.getStringValueByKey(insertMutation.getStrings(), "lastname"));
      assertEquals(
          "stamm", AvroSchemaUtils.getStringValueByKey(insertMutation.getStrings(), "lastname"));

      LOG.info(
          "SUBJECT: "
              + AvroSchemaUtils.getStringValueByKey(insertMutation.getStrings(), "subject"));
      assertEquals(
          "religion", AvroSchemaUtils.getStringValueByKey(insertMutation.getStrings(), "subject"));

      LOG.info(
          "SCORE: " + AvroSchemaUtils.getIntegerValueByKey(insertMutation.getIntegers(), "score"));
      assertEquals(
          100, (int) AvroSchemaUtils.getIntegerValueByKey(insertMutation.getIntegers(), "score"));

      LOG.info("DATE: " + AvroSchemaUtils.getStringValueByKey(insertMutation.getStrings(), "date"));
      assertEquals(
          "2014-10-22", AvroSchemaUtils.getStringValueByKey(insertMutation.getStrings(), "date"));

    } catch (IOException e) {
      e.printStackTrace();
    }
  }
Exemplo n.º 2
0
 @Override
 public int run(InputStream stdin, PrintStream out, PrintStream err, List<String> args)
     throws Exception {
   if (args.size() != 2) {
     err.println("Expected 2 arguments: schema binary_data_file");
     err.println("Use '-' as binary_data_file for stdin.");
     return 1;
   }
   Schema schema = Schema.parse(args.get(0));
   InputStream input;
   boolean needsClosing;
   if (args.get(1).equals("-")) {
     input = stdin;
     needsClosing = false;
   } else {
     input = new FileInputStream(args.get(1));
     needsClosing = true;
   }
   try {
     DatumReader<Object> reader = new GenericDatumReader<Object>(schema);
     Object datum = reader.read(null, DecoderFactory.get().binaryDecoder(input, null));
     DatumWriter<Object> writer = new GenericDatumWriter<Object>(schema);
     JsonGenerator g = new JsonFactory().createJsonGenerator(out, JsonEncoding.UTF8);
     g.useDefaultPrettyPrinter();
     writer.write(datum, EncoderFactory.get().jsonEncoder(schema, g));
     g.flush();
     out.println();
     out.flush();
   } finally {
     if (needsClosing) {
       input.close();
     }
   }
   return 0;
 }
Exemplo n.º 3
0
  /**
   * Create a deep copy of an Avro value.
   *
   * @param source The value to be copied
   * @return The deep copy of the value
   */
  @Override
  public T deepCopy(T source) {

    if (source == null) {
      return null;
    }

    if (datumReader == null) {
      datumReader = createDatumReader(conf);
    }
    if (datumWriter == null) {
      datumWriter = createDatumWriter(conf);
    }
    ByteArrayOutputStream byteOutStream = new ByteArrayOutputStream();
    binaryEncoder = EncoderFactory.get().binaryEncoder(byteOutStream, binaryEncoder);
    T target = createCopyTarget();
    try {
      datumWriter.write(source, binaryEncoder);
      binaryEncoder.flush();
      binaryDecoder =
          DecoderFactory.get().binaryDecoder(byteOutStream.toByteArray(), binaryDecoder);
      return datumReader.read(target, binaryDecoder);
    } catch (Exception e) {
      throw new CrunchRuntimeException("Error while deep copying avro value " + source, e);
    }
  }
  // It is assumed that record has following format.
  // byte 1 : version, static 0x1
  // byte 2-5 : int schemaId
  // remaining bytes would have avro data
  @Override
  public GenericRecord parse(ByteBuffer bytes) {
    if (bytes.remaining() < 5) {
      throw new ParseException("record must have at least 5 bytes carrying version and schemaId");
    }

    byte version = bytes.get();
    if (version != V1) {
      throw new ParseException("found record of arbitrary version [%s]", version);
    }

    int schemaId = bytes.getInt();
    Schema schemaObj = schemaObjs.get(schemaId);
    if (schemaObj == null) {
      throw new ParseException("Failed to find schema for id [%s]", schemaId);
    }

    try {
      DatumReader<GenericRecord> reader = new GenericDatumReader<GenericRecord>(schemaObj);
      ByteBufferInputStream inputStream =
          new ByteBufferInputStream(Collections.singletonList(bytes));

      return reader.read(null, DecoderFactory.get().binaryDecoder(inputStream, null));
    } catch (Exception e) {
      throw new ParseException(e, "Fail to decode avro message with schemaId [%s].", schemaId);
    }
  }
Exemplo n.º 5
0
 public Record fromStream(
     final InputStream stream, @Nullable final Set<URI> propertiesToDeserialize)
     throws IOException {
   final Decoder decoder = DecoderFactory.get().directBinaryDecoder(stream, null);
   final DatumReader<GenericRecord> reader = new GenericDatumReader<GenericRecord>(Schemas.NODE);
   final GenericRecord generic = reader.read(null, decoder);
   return decodeRecord(generic, propertiesToDeserialize);
 }
Exemplo n.º 6
0
  public static void checkBinaryJson(String json) throws Exception {
    Object node = Json.parseJson(json);
    ByteArrayOutputStream out = new ByteArrayOutputStream();
    DatumWriter<Object> writer = new Json.ObjectWriter();
    Encoder encoder = EncoderFactory.get().binaryEncoder(out, null);
    encoder = EncoderFactory.get().validatingEncoder(Json.SCHEMA, encoder);
    writer.write(node, encoder);
    encoder.flush();
    byte[] bytes = out.toByteArray();

    DatumReader<Object> reader = new Json.ObjectReader();
    Decoder decoder = DecoderFactory.get().binaryDecoder(bytes, null);
    decoder = DecoderFactory.get().validatingDecoder(Json.SCHEMA, decoder);
    Object decoded = reader.read(null, decoder);

    assertEquals("Decoded json does not match.", Json.toString(node), Json.toString(decoded));
  }
Exemplo n.º 7
0
  @Override
  public void deserialize(byte[] data, LuceneWorksBuilder hydrator) {
    final ByteArrayInputStream inputStream = new ByteArrayInputStream(data);
    final int majorVersion = inputStream.read();
    final int minorVersion = inputStream.read();
    final Protocol protocol = protocols.getProtocol(majorVersion, minorVersion);

    Decoder decoder = DecoderFactory.get().binaryDecoder(inputStream, null);
    GenericDatumReader<GenericRecord> reader =
        new GenericDatumReader<>(protocol.getType("Message"));
    GenericRecord result;
    try {
      result = reader.read(null, decoder);
    } catch (IOException e) {
      throw log.unableToDeserializeAvroStream(e);
    }

    classReferences = asListOfString(result, "classReferences");
    final List<GenericRecord> operations = asListOfGenericRecords(result, "operations");
    final ConversionContext conversionContext = new ContextualExceptionBridgeHelper();
    for (GenericRecord operation : operations) {
      String schema = operation.getSchema().getName();
      if ("OptimizeAll".equals(schema)) {
        hydrator.addOptimizeAll();
      } else if ("PurgeAll".equals(schema)) {
        hydrator.addPurgeAllLuceneWork(asClass(operation, "class"));
      } else if ("Flush".equals(schema)) {
        hydrator.addFlush();
      } else if ("Delete".equals(schema)) {
        processId(operation, hydrator);
        hydrator.addDeleteLuceneWork(asClass(operation, "class"), conversionContext);
      } else if ("DeleteByQuery".equals(schema)) {
        String entityClassName = asClass(operation, "class");
        int queryKey = asInt(operation, "key");
        DeleteByQuerySupport.StringToQueryMapper mapper =
            DeleteByQuerySupport.getStringToQueryMapper(queryKey);
        List<Utf8> stringList = asListOfString(operation, "query");
        String[] query = new String[stringList.size()];
        for (int i = 0; i < stringList.size(); ++i) {
          query[i] = stringList.get(i).toString();
        }
        hydrator.addDeleteByQueryLuceneWork(entityClassName, mapper.fromString(query));
      } else if ("Add".equals(schema)) {
        buildLuceneDocument(asGenericRecord(operation, "document"), hydrator);
        Map<String, String> analyzers = getAnalyzers(operation);
        processId(operation, hydrator);
        hydrator.addAddLuceneWork(asClass(operation, "class"), analyzers, conversionContext);
      } else if ("Update".equals(schema)) {
        buildLuceneDocument(asGenericRecord(operation, "document"), hydrator);
        Map<String, String> analyzers = getAnalyzers(operation);
        processId(operation, hydrator);
        hydrator.addUpdateLuceneWork(asClass(operation, "class"), analyzers, conversionContext);
      } else {
        throw log.cannotDeserializeOperation(schema);
      }
    }
  }
Exemplo n.º 8
0
 @Test(expected = AvroTypeException.class)
 public void testNoDefaultField() throws Exception {
   Schema expected =
       Schema.parse(
           "{\"type\":\"record\", \"name\":\"Foo\", \"fields\":"
               + "[{\"name\":\"f\", \"type\": \"string\"}]}");
   DatumReader<Object> in = new GenericDatumReader<Object>(ACTUAL, expected);
   in.read(null, DecoderFactory.get().binaryDecoder(new ByteArrayInputStream(new byte[0]), null));
 }
Exemplo n.º 9
0
  /**
   * Create a new Event Reader
   *
   * @param in
   * @throws IOException
   */
  @SuppressWarnings("deprecation")
  public EventReader(DataInputStream in) throws IOException {
    this.in = in;
    this.version = in.readLine();

    if (!EventWriter.VERSION.equals(version)) {
      throw new IOException("Incompatible event log version: " + version);
    }

    this.schema = Schema.parse(in.readLine());
    this.reader = new SpecificDatumReader(schema);
    this.decoder = DecoderFactory.get().jsonDecoder(schema, in);
  }
Exemplo n.º 10
0
  public URI expandURI(final byte[] bytes) {
    Preconditions.checkNotNull(bytes);
    try {
      final InputStream stream = new ByteArrayInputStream(bytes);
      final Decoder decoder = DecoderFactory.get().directBinaryDecoder(stream, null);
      final DatumReader<Object> reader =
          new GenericDatumReader<Object>(Schemas.COMPRESSED_IDENTIFIER);
      final Object generic = reader.read(null, decoder);
      return (URI) decodeNode(generic);

    } catch (final IOException ex) {
      throw new Error("Unexpected exception (!): " + ex.getMessage(), ex);
    }
  }
Exemplo n.º 11
0
  public static void checkBlockingBinary(
      Schema schema, Object datum, DatumWriter<Object> writer, DatumReader<Object> reader)
      throws IOException {
    ByteArrayOutputStream out = new ByteArrayOutputStream();
    writer.setSchema(schema);
    Encoder encoder = EncoderFactory.get().blockingBinaryEncoder(out, null);
    writer.write(datum, encoder);
    encoder.flush();
    byte[] data = out.toByteArray();

    reader.setSchema(schema);

    Object decoded = reader.read(null, DecoderFactory.get().binaryDecoder(data, null));

    assertEquals("Decoded data does not match.", datum, decoded);
  }
Exemplo n.º 12
0
 private static void checkDefault(String schemaJson, String defaultJson, Object defaultValue)
     throws Exception {
   String recordJson =
       "{\"type\":\"record\", \"name\":\"Foo\", \"fields\":[{\"name\":\"f\", "
           + "\"type\":"
           + schemaJson
           + ", "
           + "\"default\":"
           + defaultJson
           + "}]}";
   Schema expected = Schema.parse(recordJson);
   DatumReader<Object> in = new GenericDatumReader<Object>(ACTUAL, expected);
   GenericData.Record record =
       (GenericData.Record) in.read(null, DecoderFactory.get().binaryDecoder(new byte[0], null));
   assertEquals("Wrong default.", defaultValue, record.get("f"));
   assertEquals("Wrong toString", expected, Schema.parse(expected.toString()));
 }
Exemplo n.º 13
0
  private static void checkJson(Schema schema, Object datum, String json) throws Exception {
    ByteArrayOutputStream out = new ByteArrayOutputStream();
    Encoder encoder = EncoderFactory.get().jsonEncoder(schema, out);
    DatumWriter<Object> writer = new GenericDatumWriter<Object>();
    writer.setSchema(schema);
    writer.write(datum, encoder);
    encoder.flush();
    byte[] data = out.toByteArray();

    String encoded = new String(data, "UTF-8");
    assertEquals("Encoded data does not match.", json, encoded);

    DatumReader<Object> reader = new GenericDatumReader<Object>();
    reader.setSchema(schema);
    Object decoded =
        reader.read(null, DecoderFactory.get().jsonDecoder(schema, new ByteArrayInputStream(data)));

    assertEquals("Decoded data does not match.", datum, decoded);
  }
Exemplo n.º 14
0
 private void parseTopics() {
   if (state.getProperty(TOPIC_LIST) != null) {
     byte[] data = base64.decodeBase64(state.getProperty(TOPIC_LIST));
     BinaryDecoder decoder = DecoderFactory.get().binaryDecoder(data, null);
     SpecificDatumReader<Topic> avroReader = new SpecificDatumReader<>(Topic.class);
     try { // NOSONAR
       Topic decodedTopic;
       while (!decoder.isEnd()) {
         decodedTopic = avroReader.read(null, decoder);
         LOG.debug("Loaded {}", decodedTopic);
         topicMap.put(decodedTopic.getId(), decodedTopic);
       }
     } catch (Exception e) {
       LOG.error("Unexpected exception occurred while reading information from decoder", e);
     }
   } else {
     LOG.info("No topic list found in state");
   }
 }
Exemplo n.º 15
0
  private static void checkJson(
      Schema schema, Object datum, DatumWriter<Object> writer, DatumReader<Object> reader)
      throws IOException {
    ByteArrayOutputStream out = new ByteArrayOutputStream();
    Encoder encoder = EncoderFactory.get().jsonEncoder(schema, out);
    writer.setSchema(schema);
    writer.write(datum, encoder);
    writer.write(datum, encoder);
    encoder.flush();
    byte[] data = out.toByteArray();

    reader.setSchema(schema);
    Decoder decoder = DecoderFactory.get().jsonDecoder(schema, new ByteArrayInputStream(data));
    Object decoded = reader.read(null, decoder);
    assertEquals("Decoded data does not match.", datum, decoded);

    decoded = reader.read(decoded, decoder);
    assertEquals("Decoded data does not match.", datum, decoded);
  }
Exemplo n.º 16
0
 @Test
 public void testEnumMismatch() throws Exception {
   Schema actual = Schema.parse("{\"type\":\"enum\",\"name\":\"E\",\"symbols\":[\"X\",\"Y\"]}");
   Schema expected = Schema.parse("{\"type\":\"enum\",\"name\":\"E\",\"symbols\":[\"Y\",\"Z\"]}");
   ByteArrayOutputStream out = new ByteArrayOutputStream();
   DatumWriter<Object> writer = new GenericDatumWriter<Object>(actual);
   Encoder encoder = EncoderFactory.get().directBinaryEncoder(out, null);
   writer.write(new GenericData.EnumSymbol(actual, "Y"), encoder);
   writer.write(new GenericData.EnumSymbol(actual, "X"), encoder);
   encoder.flush();
   byte[] data = out.toByteArray();
   Decoder decoder = DecoderFactory.get().binaryDecoder(data, null);
   DatumReader<String> in = new GenericDatumReader<String>(actual, expected);
   assertEquals("Wrong value", new GenericData.EnumSymbol(expected, "Y"), in.read(null, decoder));
   try {
     in.read(null, decoder);
     fail("Should have thrown exception.");
   } catch (AvroTypeException e) {
     // expected
   }
 }
Exemplo n.º 17
0
  public void serializeGeneric() throws IOException {
    // Create a datum to serialize.
    Schema schema = new Schema.Parser().parse(getClass().getResourceAsStream("/MyPair.avsc"));
    GenericRecord datum = new GenericData.Record(schema);
    datum.put("left", new Utf8("dog"));
    datum.put("right", new Utf8("cat"));

    // Serialize it.
    ByteArrayOutputStream out = new ByteArrayOutputStream();
    DatumWriter<GenericRecord> writer = new GenericDatumWriter<GenericRecord>(schema);
    Encoder encoder = EncoderFactory.get().binaryEncoder(out, null);
    writer.write(datum, encoder);
    encoder.flush();
    out.close();
    System.out.println("Serialization: " + out);

    // Deserialize it.
    DatumReader<GenericRecord> reader = new GenericDatumReader<GenericRecord>(schema);
    BinaryDecoder decoder = DecoderFactory.get().binaryDecoder(out.toByteArray(), null);
    GenericRecord result = reader.read(null, decoder);
    System.out.printf("Left: %s, Right: %s\n", result.get("left"), result.get("right"));
  }
Exemplo n.º 18
0
  @Test
  public void experimentWithAvro() throws Exception {
    String root = "org/hibernate/search/remote/codex/avro/v1_1/";
    parseSchema(root + "attribute/TokenTrackingAttribute.avro", "attribute/TokenTrackingAttribute");
    parseSchema(root + "attribute/CharTermAttribute.avro", "attribute/CharTermAttribute");
    parseSchema(root + "attribute/PayloadAttribute.avro", "attribute/PayloadAttribute");
    parseSchema(root + "attribute/KeywordAttribute.avro", "attribute/KeywordAttribute");
    parseSchema(
        root + "attribute/PositionIncrementAttribute.avro", "attribute/PositionIncrementAttribute");
    parseSchema(root + "attribute/FlagsAttribute.avro", "attribute/FlagsAttribute");
    parseSchema(root + "attribute/TypeAttribute.avro", "attribute/TypeAttribute");
    parseSchema(root + "attribute/OffsetAttribute.avro", "attribute/OffsetAttribute");
    parseSchema(root + "field/TermVector.avro", "field/TermVector");
    parseSchema(root + "field/Index.avro", "field/Index");
    parseSchema(root + "field/Store.avro", "field/Store");
    parseSchema(root + "field/TokenStreamField.avro", "field/TokenStreamField");
    parseSchema(root + "field/ReaderField.avro", "field/ReaderField");
    parseSchema(root + "field/StringField.avro", "field/StringField");
    parseSchema(root + "field/BinaryField.avro", "field/BinaryField");
    parseSchema(root + "field/NumericIntField.avro", "field/NumericIntField");
    parseSchema(root + "field/NumericLongField.avro", "field/NumericLongField");
    parseSchema(root + "field/NumericFloatField.avro", "field/NumericFloatField");
    parseSchema(root + "field/NumericDoubleField.avro", "field/NumericDoubleField");
    parseSchema(root + "field/CustomFieldable.avro", "field/CustomFieldable");
    parseSchema(root + "Document.avro", "Document");
    parseSchema(root + "operation/Id.avro", "operation/Id");
    parseSchema(root + "operation/OptimizeAll.avro", "operation/OptimizeAll");
    parseSchema(root + "operation/PurgeAll.avro", "operation/PurgeAll");
    parseSchema(root + "operation/Flush.avro", "operation/Flush");
    parseSchema(root + "operation/Delete.avro", "operation/Delete");
    parseSchema(root + "operation/Add.avro", "operation/Add");
    parseSchema(root + "operation/Update.avro", "operation/Update");
    parseSchema(root + "Message.avro", "Message");

    String filename = root + "Works.avpr";
    Protocol protocol = parseProtocol(filename, "Works");
    final Schema termVectorSchema = protocol.getType("TermVector");
    final Schema indexSchema = protocol.getType("Index");
    final Schema storeSchema = protocol.getType("Store");
    final Schema tokenTrackingAttribute = protocol.getType("TokenTrackingAttribute");
    final Schema tokenStreamSchema = protocol.getType("TokenStreamField");
    final Schema readerSchema = protocol.getType("ReaderField");
    final Schema stringSchema = protocol.getType("StringField");
    final Schema binarySchema = protocol.getType("BinaryField");
    final Schema intFieldSchema = protocol.getType("NumericIntField");
    final Schema longFieldSchema = protocol.getType("NumericLongField");
    final Schema floatFieldSchema = protocol.getType("NumericFloatField");
    final Schema doubleFieldSchema = protocol.getType("NumericDoubleField");
    final Schema custonFieldableSchema = protocol.getType("CustomFieldable");
    final Schema documentSchema = protocol.getType("Document");
    final Schema idSchema = protocol.getType("Id");
    final Schema optimizeAllSchema = protocol.getType("OptimizeAll");
    final Schema purgeAllSchema = protocol.getType("PurgeAll");
    final Schema flushSchema = protocol.getType("Flush");
    final Schema deleteSchema = protocol.getType("Delete");
    final Schema addSchema = protocol.getType("Add");
    final Schema updateSchema = protocol.getType("Update");
    Schema messageSchema = protocol.getType("Message");

    final ByteArrayOutputStream out = new ByteArrayOutputStream();
    GenericDatumWriter<GenericRecord> writer = new GenericDatumWriter<GenericRecord>(messageSchema);
    Encoder encoder = EncoderFactory.get().directBinaryEncoder(out, null);

    byte[] serializableSample = new byte[10];
    for (int i = 0; i < 10; i++) {
      serializableSample[i] = (byte) i;
    }

    List<String> classReferences = new ArrayList<String>();
    classReferences.add(AvroTest.class.getName());

    List<GenericRecord> fieldables = new ArrayList<GenericRecord>(1);
    // custom fieldable
    GenericRecord customFieldable = new GenericData.Record(custonFieldableSchema);
    customFieldable.put("instance", ByteBuffer.wrap(serializableSample));
    fieldables.add(customFieldable);

    // numeric fields
    GenericRecord numericField = createNumeric(intFieldSchema);
    numericField.put("value", 3);
    fieldables.add(numericField);
    numericField = createNumeric(longFieldSchema);
    numericField.put("value", 3l);
    fieldables.add(numericField);
    numericField = createNumeric(floatFieldSchema);
    numericField.put("value", 2.3f);
    fieldables.add(numericField);
    numericField = createNumeric(doubleFieldSchema);
    numericField.put("value", 2.3d);
    fieldables.add(numericField);

    // fields
    GenericRecord field = createField(binarySchema);
    field.put("offset", 0);
    field.put("length", 10);
    field.put("value", ByteBuffer.wrap(serializableSample));
    fieldables.add(field);
    field = createField(stringSchema);
    field.put("value", stringSchema.getName());
    field.put("store", "YES");
    field.put("index", "ANALYZED");
    field.put("termVector", "WITH_OFFSETS");
    fieldables.add(field);
    field = createField(tokenStreamSchema);

    List<List<Object>> tokens = new ArrayList<List<Object>>();
    List<Object> attrs = new ArrayList<Object>();
    tokens.add(attrs);
    GenericData.Record attr = new GenericData.Record(tokenTrackingAttribute);
    List<Integer> positions = new ArrayList<Integer>();
    positions.add(1);
    positions.add(2);
    positions.add(3);
    positions.add(4);
    attr.put("positions", positions);
    attrs.add(attr);
    attrs.add(ByteBuffer.wrap(serializableSample));

    field.put("value", tokens);
    field.put("termVector", "WITH_OFFSETS");
    fieldables.add(field);
    field = createField(readerSchema);
    field.put("value", ByteBuffer.wrap(serializableSample));
    field.put("termVector", "WITH_OFFSETS");
    fieldables.add(field);

    GenericRecord doc = new GenericData.Record(documentSchema);
    doc.put("boost", 2.3f);
    doc.put("fieldables", fieldables);

    GenericRecord add = new GenericData.Record(addSchema);
    add.put("class", classReferences.indexOf(AvroTest.class.getName()));
    GenericRecord id = new GenericData.Record(idSchema);
    id.put("value", ByteBuffer.wrap(serializableSample));
    add.put("id", id);
    add.put("document", doc);
    Map<String, String> analyzers = new HashMap<String, String>();
    analyzers.put("name", "ngram");
    analyzers.put("description", "porter");
    add.put("fieldToAnalyzerMap", analyzers);

    GenericRecord delete = new GenericData.Record(deleteSchema);
    delete.put("class", classReferences.indexOf(AvroTest.class.getName()));
    id = new GenericData.Record(idSchema);
    id.put("value", new Long(30));
    delete.put("id", id);

    GenericRecord purgeAll = new GenericData.Record(purgeAllSchema);
    purgeAll.put("class", classReferences.indexOf(AvroTest.class.getName()));
    GenericRecord optimizeAll = new GenericData.Record(optimizeAllSchema);

    GenericRecord flush = new GenericData.Record(flushSchema);

    List<GenericRecord> operations = new ArrayList<GenericRecord>(1);
    operations.add(purgeAll);
    operations.add(optimizeAll);
    operations.add(flush);
    operations.add(delete);
    operations.add(add);

    GenericRecord message = new GenericData.Record(messageSchema);
    message.put("classReferences", classReferences);
    message.put("operations", operations);

    writer.write(message, encoder);
    encoder.flush();

    ByteArrayInputStream inputStream = new ByteArrayInputStream(out.toByteArray());
    Decoder decoder = DecoderFactory.get().binaryDecoder(inputStream, null);
    GenericDatumReader<GenericRecord> reader = new GenericDatumReader<GenericRecord>(messageSchema);
    while (true) {
      try {
        GenericRecord result = reader.read(null, decoder);
        System.out.println(result);

        assertThat(result).isNotNull();
        // operations
        assertThat(result.get("operations")).isNotNull().isInstanceOf(List.class);
        List<?> ops = (List<?>) result.get("operations");
        assertThat(ops).hasSize(5);

        // Flush
        assertThat(ops.get(2)).isInstanceOf(GenericRecord.class);
        GenericRecord flushOp = (GenericRecord) ops.get(2);
        assertThat(flushOp.getSchema().getName()).isEqualTo("Flush");

        // Delete
        assertThat(ops.get(3)).isInstanceOf(GenericRecord.class);
        GenericRecord deleteOp = (GenericRecord) ops.get(3);
        assertThat(deleteOp.getSchema().getName()).isEqualTo("Delete");
        Object actual = ((GenericRecord) deleteOp.get("id")).get("value");
        assertThat(actual).isInstanceOf(Long.class);
        assertThat(actual).isEqualTo(Long.valueOf(30));

        // Add
        assertThat(ops.get(4)).isInstanceOf(GenericRecord.class);
        GenericRecord addOp = (GenericRecord) ops.get(4);
        assertThat(addOp.getSchema().getName()).isEqualTo("Add");
        actual = ((GenericRecord) addOp.get("id")).get("value");
        assertThat(actual).isInstanceOf(ByteBuffer.class);
        ByteBuffer bb = (ByteBuffer) actual;
        assertThat(bb.hasArray()).isTrue();
        byte[] copy = new byte[bb.remaining()];
        bb.get(copy);
        assertThat(serializableSample).isEqualTo(copy);

        // fieldToAnalyzerMap
        assertThat(addOp.get("fieldToAnalyzerMap")).isInstanceOf(Map.class);
        assertThat((Map) addOp.get("fieldToAnalyzerMap")).hasSize(2);

        // document
        assertThat(addOp.get("document")).isNotNull();
        GenericRecord document = (GenericRecord) addOp.get("document");
        assertThat(document.get("boost")).isEqualTo(2.3f);

        // numeric fields
        assertThat(document.get("fieldables")).isNotNull().isInstanceOf(List.class);
        List<?> fields = (List<?>) document.get("fieldables");

        assertThat(fields).hasSize(9); // custom + 4 numerics + 4 fields

        field = (GenericRecord) fields.get(0);
        assertThat(field.getSchema().getName()).isEqualTo("CustomFieldable");
        field = (GenericRecord) fields.get(1);
        assertThat(field.getSchema().getName()).isEqualTo("NumericIntField");
        assertThat(field.get("value")).isEqualTo(3);
        assertNumericField(field);
        field = (GenericRecord) fields.get(2);
        assertThat(field.getSchema().getName()).isEqualTo("NumericLongField");
        assertThat(field.get("value")).isEqualTo(3l);
        assertNumericField(field);
        field = (GenericRecord) fields.get(3);
        assertThat(field.getSchema().getName()).isEqualTo("NumericFloatField");
        assertThat(field.get("value")).isEqualTo(2.3f);
        assertNumericField(field);
        field = (GenericRecord) fields.get(4);
        assertThat(field.getSchema().getName()).isEqualTo("NumericDoubleField");
        assertThat(field.get("value")).isEqualTo(2.3d);
        assertNumericField(field);

        // fields
        field = (GenericRecord) fields.get(5);
        assertThat(field.getSchema().getName()).isEqualTo("BinaryField");
        assertThat(field.get("value")).isInstanceOf(ByteBuffer.class);
        assertField(field);

        field = (GenericRecord) fields.get(6);
        assertThat(field.getSchema().getName()).isEqualTo("StringField");
        assertThat(field.get("value")).isInstanceOf(Utf8.class);
        assertTermVector(field);
        assertIndexAndStore(field);
        assertField(field);

        field = (GenericRecord) fields.get(7);
        assertThat(field.getSchema().getName()).isEqualTo("TokenStreamField");
        assertThat(field.get("value")).isInstanceOf(List.class);
        List<List<Object>> l1 = (List<List<Object>>) field.get("value");
        assertThat(l1.get(0)).as("Wrong attribute impl list").hasSize(2);
        Object object = l1.get(0).get(0);
        assertThat(object).isNotNull();
        assertTermVector(field);
        assertField(field);

        field = (GenericRecord) fields.get(8);
        assertThat(field.getSchema().getName()).isEqualTo("ReaderField");
        assertThat(field.get("value")).isInstanceOf(ByteBuffer.class);
        assertTermVector(field);
        assertField(field);
      } catch (EOFException eof) {
        break;
      } catch (Exception ex) {
        ex.printStackTrace();
        throw ex;
      }
    }
  }
Exemplo n.º 19
0
 public Object fromStream(final InputStream stream) throws IOException {
   final Decoder decoder = DecoderFactory.get().directBinaryDecoder(stream, null);
   final DatumReader<Object> reader = new GenericDatumReader<Object>(Schemas.NODE);
   final Object generic = reader.read(null, decoder);
   return decodeNode(generic);
 }