本文整理汇总了Java中org.apache.hadoop.hive.ql.io.RCFileOutputFormat类的典型用法代码示例。如果您正苦于以下问题:Java RCFileOutputFormat类的具体用法?Java RCFileOutputFormat怎么用?Java RCFileOutputFormat使用的例子?那么恭喜您, 这里精选的类代码示例或许可以为您提供帮助。
RCFileOutputFormat类属于org.apache.hadoop.hive.ql.io包,在下文中一共展示了RCFileOutputFormat类的11个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Java代码示例。
示例1: testRCText
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
@Test
public void testRCText()
throws Exception
{
List<TestColumn> testColumns = ImmutableList.copyOf(filter(TEST_COLUMNS, testColumn -> {
return !testColumn.getName().equals("t_struct_null") // TODO: This is a bug in the RC text reader
&& !testColumn.getName().equals("t_map_null_key_complex_key_value"); // RC file does not support complex type as key of a map
}));
HiveOutputFormat<?, ?> outputFormat = new RCFileOutputFormat();
InputFormat<?, ?> inputFormat = new RCFileInputFormat<>();
@SuppressWarnings("deprecation")
SerDe serde = new ColumnarSerDe();
File file = File.createTempFile("presto_test", "rc-text");
try {
FileSplit split = createTestFile(file.getAbsolutePath(), outputFormat, serde, null, testColumns, NUM_ROWS);
testCursorProvider(new ColumnarTextHiveRecordCursorProvider(), split, inputFormat, serde, testColumns, NUM_ROWS);
testCursorProvider(new GenericHiveRecordCursorProvider(), split, inputFormat, serde, testColumns, NUM_ROWS);
}
finally {
//noinspection ResultOfMethodCallIgnored
file.delete();
}
}
示例2: testRcTextPageSource
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
@Test(enabled = false)
public void testRcTextPageSource()
throws Exception
{
HiveOutputFormat<?, ?> outputFormat = new RCFileOutputFormat();
InputFormat<?, ?> inputFormat = new RCFileInputFormat<>();
@SuppressWarnings("deprecation")
SerDe serde = new ColumnarSerDe();
File file = File.createTempFile("presto_test", "rc-binary");
file.delete();
try {
FileSplit split = createTestFile(file.getAbsolutePath(), outputFormat, serde, null, TEST_COLUMNS, NUM_ROWS);
testPageSourceFactory(new RcFilePageSourceFactory(TYPE_MANAGER), split, inputFormat, serde, TEST_COLUMNS);
}
finally {
//noinspection ResultOfMethodCallIgnored
file.delete();
}
}
示例3: testRCBinary
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
@Test
public void testRCBinary()
throws Exception
{
List<TestColumn> testColumns = ImmutableList.copyOf(filter(TEST_COLUMNS, testColumn -> {
// RC file does not support complex type as key of a map
return !testColumn.getName().equals("t_map_null_key_complex_key_value");
}));
HiveOutputFormat<?, ?> outputFormat = new RCFileOutputFormat();
InputFormat<?, ?> inputFormat = new RCFileInputFormat<>();
@SuppressWarnings("deprecation")
SerDe serde = new LazyBinaryColumnarSerDe();
File file = File.createTempFile("presto_test", "rc-binary");
try {
FileSplit split = createTestFile(file.getAbsolutePath(), outputFormat, serde, null, testColumns, NUM_ROWS);
testCursorProvider(new ColumnarBinaryHiveRecordCursorProvider(), split, inputFormat, serde, testColumns, NUM_ROWS);
testCursorProvider(new GenericHiveRecordCursorProvider(), split, inputFormat, serde, testColumns, NUM_ROWS);
}
finally {
//noinspection ResultOfMethodCallIgnored
file.delete();
}
}
示例4: testRcBinaryPageSource
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
@Test(enabled = false)
public void testRcBinaryPageSource()
throws Exception
{
HiveOutputFormat<?, ?> outputFormat = new RCFileOutputFormat();
InputFormat<?, ?> inputFormat = new RCFileInputFormat<>();
@SuppressWarnings("deprecation")
SerDe serde = new LazyBinaryColumnarSerDe();
File file = File.createTempFile("presto_test", "rc-binary");
file.delete();
try {
FileSplit split = createTestFile(file.getAbsolutePath(), outputFormat, serde, null, TEST_COLUMNS, NUM_ROWS);
testPageSourceFactory(new RcFilePageSourceFactory(TYPE_MANAGER), split, inputFormat, serde, TEST_COLUMNS);
}
finally {
//noinspection ResultOfMethodCallIgnored
file.delete();
}
}
示例5: getStoreType
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
public static String getStoreType(String fileFormat) {
Preconditions.checkNotNull(fileFormat);
String[] fileFormatArrary = fileFormat.split("\\.");
if(fileFormatArrary.length < 1) {
throw new CatalogException("Hive file output format is wrong. - file output format:" + fileFormat);
}
String outputFormatClass = fileFormatArrary[fileFormatArrary.length-1];
if(outputFormatClass.equals(HiveIgnoreKeyTextOutputFormat.class.getSimpleName())) {
return CatalogProtos.StoreType.CSV.name();
} else if(outputFormatClass.equals(RCFileOutputFormat.class.getSimpleName())) {
return CatalogProtos.StoreType.RCFILE.name();
} else {
throw new CatalogException("Not supported file output format. - file output format:" + fileFormat);
}
}
示例6: getStoreType
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
public static String getStoreType(String fileFormat) {
Preconditions.checkNotNull(fileFormat);
String[] fileFormatArrary = fileFormat.split("\\.");
if(fileFormatArrary.length < 1) {
throw new CatalogException("Hive file output format is wrong. - file output format:" + fileFormat);
}
String outputFormatClass = fileFormatArrary[fileFormatArrary.length-1];
if(outputFormatClass.equals(HiveIgnoreKeyTextOutputFormat.class.getSimpleName())) {
return CatalogProtos.StoreType.CSV.name();
} else if(outputFormatClass.equals(HiveSequenceFileOutputFormat.class.getSimpleName())) {
return CatalogProtos.StoreType.SEQUENCEFILE.name();
} else if(outputFormatClass.equals(RCFileOutputFormat.class.getSimpleName())) {
return CatalogProtos.StoreType.RCFILE.name();
} else {
throw new CatalogException("Not supported file output format. - file output format:" + fileFormat);
}
}
示例7: setStoreLocation
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
@Override
public void setStoreLocation(String location, Job job) throws IOException {
super.setStoreLocation(location, job);
// set number of columns if this is set in context.
Properties p = getUDFProperties();
if (p != null) {
numColumns = Integer.parseInt(p.getProperty("numColumns", "-1"));
}
if (numColumns > 0) {
RCFileOutputFormat.setColumnNumber(job.getConfiguration(), numColumns);
}
}
示例8: writeRCFileTest
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
private static int writeRCFileTest(FileSystem fs, int rowCount, Path file, int columnNum,
CompressionCodec codec, int columnCount) throws IOException {
fs.delete(file, true);
int rowsWritten = 0;
resetRandomGenerators();
RCFileOutputFormat.setColumnNumber(conf, columnNum);
RCFile.Writer writer = new RCFile.Writer(fs, conf, file, null, codec);
byte[][] columnRandom;
BytesRefArrayWritable bytes = new BytesRefArrayWritable(columnNum);
columnRandom = new byte[columnNum][];
for (int i = 0; i < columnNum; i++) {
BytesRefWritable cu = new BytesRefWritable();
bytes.set(i, cu);
}
for (int i = 0; i < rowCount; i++) {
nextRandomRow(columnRandom, bytes, columnCount);
rowsWritten++;
writer.append(bytes);
}
writer.close();
return rowsWritten;
}
示例9: writeRCFileTest
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
private static int writeRCFileTest(FileSystem fs, int rowCount, Path file, int columnNum,
CompressionCodec codec, int columnCount) throws IOException {
fs.delete(file, true);
int rowsWritten = 0;
RCFileOutputFormat.setColumnNumber(conf, columnNum);
RCFile.Writer writer = new RCFile.Writer(fs, conf, file, null, codec);
byte[][] columnRandom;
BytesRefArrayWritable bytes = new BytesRefArrayWritable(columnNum);
columnRandom = new byte[columnNum][];
for (int i = 0; i < columnNum; i++) {
BytesRefWritable cu = new BytesRefWritable();
bytes.set(i, cu);
}
for (int i = 0; i < rowCount; i++) {
bytes.resetValid(columnRandom.length);
for (int j = 0; j < columnRandom.length; j++) {
columnRandom[j]= "Sample value".getBytes();
bytes.get(j).set(columnRandom[j], 0, columnRandom[j].length);
}
rowsWritten++;
writer.append(bytes);
}
writer.close();
return rowsWritten;
}
示例10: writeRCFileTest
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
private static int writeRCFileTest(FileSystem fs, int rowCount, Path file,
int columnNum, CompressionCodec codec, int columnCount)
throws IOException {
fs.delete(file, true);
int rowsWritten = 0;
resetRandomGenerators();
RCFileOutputFormat.setColumnNumber(conf, columnNum);
RCFile.Writer writer = new RCFile.Writer(fs, conf, file, null, codec);
byte[][] columnRandom;
BytesRefArrayWritable bytes = new BytesRefArrayWritable(columnNum);
columnRandom = new byte[columnNum][];
for (int i = 0; i < columnNum; i++) {
BytesRefWritable cu = new BytesRefWritable();
bytes.set(i, cu);
}
for (int i = 0; i < rowCount; i++) {
nextRandomRow(columnRandom, bytes, columnCount);
rowsWritten++;
writer.append(bytes);
}
writer.close();
return rowsWritten;
}
示例11: writeTestData
import org.apache.hadoop.hive.ql.io.RCFileOutputFormat; //导入依赖的package包/类
@Override
public void writeTestData(File file, int recordCounts, int columnCount,
String colSeparator) throws IOException {
// write random test data
Configuration conf = new Configuration();
FileSystem fs = FileSystem.getLocal(conf);
RCFileOutputFormat.setColumnNumber(conf, columnCount);
RCFile.Writer writer = new RCFile.Writer(fs, conf, new Path(
file.getAbsolutePath()));
BytesRefArrayWritable bytes = new BytesRefArrayWritable(columnCount);
for (int c = 0; c < columnCount; c++) {
bytes.set(c, new BytesRefWritable());
}
try {
for (int r = 0; r < recordCounts; r++) {
// foreach row write n columns
for (int c = 0; c < columnCount; c++) {
byte[] stringbytes = String.valueOf(Math.random())
.getBytes();
bytes.get(c).set(stringbytes, 0, stringbytes.length);
}
writer.append(bytes);
}
} finally {
writer.close();
}
}