Java DataFileReader类代码示例

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
/**
 * Reads in binary Avro-encoded entities using the schema stored in the file
 * and prints them out.
 */
public static void readAvroFile(File file) throws IOException {
	GenericDatumReader datum = new GenericDatumReader();
	DataFileReader reader = new DataFileReader(file, datum);

	GenericData.Record record = new GenericData.Record(reader.getSchema());
	while (reader.hasNext()) {
		reader.next(record);
		System.out.println("Name " + record.get("name") + " on "
				+ record.get("Meetup_date") + " attending "
				+ record.get("going") + " organized by  "
				+ record.get("organizer") + " on  " + record.get("topics"));
	}

	reader.close();
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
public AvroFileInputStream(FileStatus status) throws IOException {
  pos = 0;
  buffer = new byte[0];
  GenericDatumReader<Object> reader = new GenericDatumReader<Object>();
  FileContext fc = FileContext.getFileContext(new Configuration());
  fileReader =
    DataFileReader.openReader(new AvroFSInput(fc, status.getPath()),reader);
  Schema schema = fileReader.getSchema();
  writer = new GenericDatumWriter<Object>(schema);
  output = new ByteArrayOutputStream();
  JsonGenerator generator =
    new JsonFactory().createJsonGenerator(output, JsonEncoding.UTF8);
  MinimalPrettyPrinter prettyPrinter = new MinimalPrettyPrinter();
  prettyPrinter.setRootValueSeparator(System.getProperty("line.separator"));
  generator.setPrettyPrinter(prettyPrinter);
  encoder = EncoderFactory.get().jsonEncoder(schema, generator);
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
@Test
public void testGenericRecord() throws IOException {
	final Path outputPath = new Path(File.createTempFile("avro-output-file", "generic.avro").getAbsolutePath());
	final AvroOutputFormat<GenericRecord> outputFormat = new AvroOutputFormat<>(outputPath, GenericRecord.class);
	Schema schema = new Schema.Parser().parse("{\"type\":\"record\", \"name\":\"user\", \"fields\": [{\"name\":\"user_name\", \"type\":\"string\"}, {\"name\":\"favorite_number\", \"type\":\"int\"}, {\"name\":\"favorite_color\", \"type\":\"string\"}]}");
	outputFormat.setWriteMode(FileSystem.WriteMode.OVERWRITE);
	outputFormat.setSchema(schema);
	output(outputFormat, schema);

	GenericDatumReader<GenericRecord> reader = new GenericDatumReader<>(schema);
	DataFileReader<GenericRecord> dataFileReader = new DataFileReader<>(new File(outputPath.getPath()), reader);

	while (dataFileReader.hasNext()) {
		GenericRecord record = dataFileReader.next();
		assertEquals(record.get("user_name").toString(), "testUser");
		assertEquals(record.get("favorite_number"), 1);
		assertEquals(record.get("favorite_color").toString(), "blue");
	}

	//cleanup
	FileSystem fs = FileSystem.getLocalFileSystem();
	fs.delete(outputPath, false);

}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
public static Schema getSchema(SeekableInput input) throws IOException
{
    DatumReader<GenericRecord> datumReader = new GenericDatumReader<GenericRecord>();
    DataFileReader<GenericRecord> dataFileReader =
            new DataFileReader<GenericRecord>(input, datumReader);
    Schema schema = dataFileReader.getSchema();

    if (PadDefaultNullsToSchema)
    {
        // a list of "cloned" fields, with optional default value set to null
        ArrayList<Field> paddedFields = new ArrayList<Field>();

        for (Field field: schema.getFields())
        {
            // should this field be padded?
            boolean needsNullPadding = (field.schema() != null) // the field has nested schema
                && (field.schema().getType().equals(Type.UNION)) // the nested schema is UNION
                && (field.schema().getTypes().get(0).getType().equals(Type.NULL)); // the first element of union is NULL type

            JsonNode defValue = needsNullPadding ? NullNode.getInstance() : field.defaultValue();

            Field f = new Field(field.name(), field.schema(), field.doc(), defValue);
            paddedFields.add(f);
        }

        schema = Schema.createRecord(schema.getName(), schema.getDoc(), schema.getNamespace(), schema.isError());
        schema.setFields(paddedFields);
    }

    return schema;
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
private void initialize() throws IOException, NoSuchAlgorithmException {
  SeekableResettableInputBridge in = new SeekableResettableInputBridge(ris);
  long pos = in.tell();
  in.seek(0L);
  fileReader = new DataFileReader<GenericRecord>(in,
      new GenericDatumReader<GenericRecord>());
  fileReader.sync(pos);

  schema = fileReader.getSchema();
  datumWriter = new GenericDatumWriter(schema);
  out = new ByteArrayOutputStream();
  encoder = EncoderFactory.get().binaryEncoder(out, encoder);

  schemaHash = SchemaNormalization.parsingFingerprint("CRC-64-AVRO", schema);
  schemaHashString = Hex.encodeHexString(schemaHash);
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
public void testFirstUnderscoreInColumnName() throws IOException {
  String [] names = { "_NAME" };
  String [] types = { "INT" };
  String [] vals = { "1987" };
  createTableWithColTypesAndNames(names, types, vals);

  runImport(getOutputArgv(true, null));

  Path outputFile = new Path(getTablePath(), "part-m-00000.avro");
  DataFileReader<GenericRecord> reader = read(outputFile);
  Schema schema = reader.getSchema();
  assertEquals(Schema.Type.RECORD, schema.getType());
  List<Field> fields = schema.getFields();
  assertEquals(types.length, fields.size());

  checkField(fields.get(0), "__NAME", Type.INT);

  GenericRecord record1 = reader.next();
  assertEquals("__NAME", 1987, record1.get("__NAME"));
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
public void testNonstandardCharactersInColumnName() throws IOException {
  String [] names = { "avro\uC3A11" };
  String [] types = { "INT" };
  String [] vals = { "1987" };
  createTableWithColTypesAndNames(names, types, vals);

  runImport(getOutputArgv(true, null));

  Path outputFile = new Path(getTablePath(), "part-m-00000.avro");
  DataFileReader<GenericRecord> reader = read(outputFile);
  Schema schema = reader.getSchema();
  assertEquals(Schema.Type.RECORD, schema.getType());
  List<Field> fields = schema.getFields();
  assertEquals(types.length, fields.size());

  checkField(fields.get(0), "AVRO1", Type.INT);

  GenericRecord record1 = reader.next();
  assertEquals("AVRO1", 1987, record1.get("AVRO1"));
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
public void testNonIdentCharactersInColumnName() throws IOException {
  String [] names = { "test_a-v+r/o" };
  String [] types = { "INT" };
  String [] vals = { "2015" };
  createTableWithColTypesAndNames(names, types, vals);

  runImport(getOutputArgv(true, null));

  Path outputFile = new Path(getTablePath(), "part-m-00000.avro");
  DataFileReader<GenericRecord> reader = read(outputFile);
  Schema schema = reader.getSchema();
  assertEquals(Schema.Type.RECORD, schema.getType());
  List<Field> fields = schema.getFields();
  assertEquals(types.length, fields.size());

  checkField(fields.get(0), "TEST_A_V_R_O", Type.INT);

  GenericRecord record1 = reader.next();
  assertEquals("TEST_A_V_R_O", 2015, record1.get("TEST_A_V_R_O"));
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
/** Import blob data that is smaller than inline lob limit. Blob data
 * should be saved as Avro bytes.
 * @throws IOException
 * @throws SQLException
 */
public void testBlobAvroImportInline() throws IOException, SQLException {
  String [] types = { getBlobType() };
  String expectedVal = "This is short BLOB data";
  String [] vals = { getBlobInsertStr(expectedVal) };

  createTableWithColTypes(types, vals);

  runImport(getArgv());

  Path outputFile = new Path(getTablePath(), "part-m-00000.avro");
  DataFileReader<GenericRecord> reader = read(outputFile);
  GenericRecord record = reader.next();

  // Verify that blob data is imported as Avro bytes.
  ByteBuffer buf = (ByteBuffer) record.get(getColName(0));
  String returnVal = new String(buf.array());

  assertEquals(getColName(0), expectedVal, returnVal);
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
/**
 * Import blob data that is smaller than inline lob limit and compress with
 * deflate codec. Blob data should be encoded and saved as Avro bytes.
 * @throws IOException
 * @throws SQLException
 */
public void testBlobCompressedAvroImportInline()
    throws IOException, SQLException {
  String [] types = { getBlobType() };
  String expectedVal = "This is short BLOB data";
  String [] vals = { getBlobInsertStr(expectedVal) };

  createTableWithColTypes(types, vals);

  runImport(getArgv("--compression-codec", CodecMap.DEFLATE));

  Path outputFile = new Path(getTablePath(), "part-m-00000.avro");
  DataFileReader<GenericRecord> reader = read(outputFile);
  GenericRecord record = reader.next();

  // Verify that the data block of the Avro file is compressed with deflate
  // codec.
  assertEquals(CodecMap.DEFLATE,
      reader.getMetaString(DataFileConstants.CODEC));

  // Verify that all columns are imported correctly.
  ByteBuffer buf = (ByteBuffer) record.get(getColName(0));
  String returnVal = new String(buf.array());

  assertEquals(getColName(0), expectedVal, returnVal);
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
public void deserUserCompile(){
	// Deserialize Users from disk
	DatumReader<User> userDatumReader = new SpecificDatumReader<User>(User.class);
	DataFileReader<User> dataFileReader = null;
	User user = null;
	try {
		dataFileReader = new DataFileReader<User>(
										new File("/Users/a/Desktop/tmp/users.avro"),
										userDatumReader);
		while (dataFileReader.hasNext()) {
			// Reuse user object by passing it to next(). This saves us from
			// allocating and garbage collecting many objects for files with
			// many items.
			user = dataFileReader.next(user);
			System.out.println(user);
		}
	} catch (IOException e) {
		e.printStackTrace();
	}
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
/**
 * 动态反序列：通过Schema文件进行动态反序列化操作
 * 
 * @throws IOException
 */
public void MemberInfoDynDeser() throws IOException {
    // 1.schema文件解析
    Parser parser = new Parser();
    Schema mSchema = parser.parse(this.getClass().getResourceAsStream("/Members.avsc"));

    // 2.构建数据读对象
    DatumReader<GenericRecord> mGr = new SpecificDatumReader<GenericRecord>(mSchema);
    DataFileReader<GenericRecord> mDfr = new DataFileReader<GenericRecord>(new File("/Users/a/Desktop/tmp/members.avro"), mGr);
    // 3.从序列化文件中进行数据反序列化取出数据
    GenericRecord gr = null;
    while (mDfr.hasNext()) {
        gr = mDfr.next();
        System.err.println("deser data:" + gr.toString());
    }
    mDfr.close();
    System.out.println("Dyn Builder Ser Start Complete.");
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
public static void main(String[] args) throws IOException
{
    String filename = args[0] ;
    
    File file=new File(filename) ;
    DatumReader<GenericRecord> reader= new GenericDatumReader<GenericRecord>();
    DataFileReader<GenericRecord> dataFileReader=new DataFileReader<GenericRecord>(file,reader);
    
    while (dataFileReader.hasNext())
    {
        GenericRecord result=dataFileReader.next();
        String output = String.format("%s %s %s %f",
            result.get("sighting_date"), result.get("city"), result.get("shape"), result.get("duration")) ;
        System.out.println(output) ;
    }
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
public static void main(String[] args) throws IOException
{
    String filename = args[0] ;
    
    File file=new File(filename) ;
    DatumReader<GenericRecord> reader= new GenericDatumReader<GenericRecord>();
    DataFileReader<GenericRecord> dataFileReader=new DataFileReader<GenericRecord>(file,reader);
    
    while (dataFileReader.hasNext())
    {
        GenericRecord result=dataFileReader.next();
        String output = String.format("%s %d",
            result.get("shape"), result.get("count")) ;
        System.out.println(output) ;
    }
}
 

import org.apache.avro.file.DataFileReader; //导入依赖的package包/类
private DataFileReader<E> initReader(FileInputSplit split) throws IOException {
	DatumReader<E> datumReader;

	if (org.apache.avro.generic.GenericRecord.class == avroValueType) {
		datumReader = new GenericDatumReader<E>();
	} else {
		datumReader = org.apache.avro.specific.SpecificRecordBase.class.isAssignableFrom(avroValueType)
			? new SpecificDatumReader<E>(avroValueType) : new ReflectDatumReader<E>(avroValueType);
	}
	if (LOG.isInfoEnabled()) {
		LOG.info("Opening split {}", split);
	}

	SeekableInput in = new FSDataInputStreamWrapper(stream, split.getPath().getFileSystem().getFileStatus(split.getPath()).getLen());
	DataFileReader<E> dataFileReader = (DataFileReader) DataFileReader.openReader(in, datumReader);

	if (LOG.isDebugEnabled()) {
		LOG.debug("Loaded SCHEMA: {}", dataFileReader.getSchema());
	}

	end = split.getStart() + split.getLength();
	recordsReadSinceLastSync = 0;
	return dataFileReader;
}