Search in sources :

Example 76 with PigServer

use of org.apache.pig.PigServer in project hive by apache.

the class AbstractHCatStorerTest method testDynamicPartitioningMultiPartColsNoDataInDataNoSpec.

@Test
public void testDynamicPartitioningMultiPartColsNoDataInDataNoSpec() throws Exception {
    AbstractHCatLoaderTest.dropTable("employee", driver);
    AbstractHCatLoaderTest.createTableDefaultDB("employee", "emp_id INT, emp_name STRING, emp_start_date STRING , emp_gender STRING", "emp_country STRING , emp_state STRING", driver, storageFormat);
    String[] inputData = {};
    HcatTestUtils.createTestDataFile(INPUT_FILE_NAME, inputData);
    PigServer pig = createPigServer(false);
    pig.setBatchOn();
    pig.registerQuery("A = LOAD '" + INPUT_FILE_NAME + "' USING PigStorage() AS (emp_id:int,emp_name:chararray,emp_start_date:chararray," + "emp_gender:chararray,emp_country:chararray,emp_state:chararray);");
    pig.registerQuery("IN = FILTER A BY emp_country == 'IN';");
    pig.registerQuery("STORE IN INTO 'employee' USING " + HCatStorer.class.getName() + "();");
    pig.executeBatch();
    driver.run("select * from employee");
    ArrayList<String> results = new ArrayList<String>();
    driver.getResults(results);
    assertEquals(0, results.size());
    driver.run("drop table employee");
}
Also used : PigServer(org.apache.pig.PigServer) ArrayList(java.util.ArrayList) Test(org.junit.Test) HCatBaseTest(org.apache.hive.hcatalog.mapreduce.HCatBaseTest)

Example 77 with PigServer

use of org.apache.pig.PigServer in project hive by apache.

the class AbstractHCatStorerTest method testStoreWithNoCtorArgs.

@Test
public void testStoreWithNoCtorArgs() throws Exception {
    AbstractHCatLoaderTest.dropTable("junit_unparted", driver);
    AbstractHCatLoaderTest.createTableDefaultDB("junit_unparted", "a int, b string", null, driver, storageFormat);
    int LOOP_SIZE = 3;
    String[] input = new String[LOOP_SIZE * LOOP_SIZE];
    int k = 0;
    for (int i = 1; i <= LOOP_SIZE; i++) {
        String si = i + "";
        for (int j = 1; j <= LOOP_SIZE; j++) {
            input[k++] = si + "\t" + j;
        }
    }
    HcatTestUtils.createTestDataFile(INPUT_FILE_NAME, input);
    PigServer server = createPigServer(false);
    server.setBatchOn();
    server.registerQuery("A = load '" + INPUT_FILE_NAME + "' as (a:int, b:chararray);");
    server.registerQuery("store A into 'junit_unparted' using " + HCatStorer.class.getName() + "();");
    server.executeBatch();
    driver.run("select * from junit_unparted");
    ArrayList<String> res = new ArrayList<String>();
    driver.getResults(res);
    driver.run("drop table junit_unparted");
    Iterator<String> itr = res.iterator();
    for (int i = 0; i < LOOP_SIZE * LOOP_SIZE; i++) {
        assertEquals(input[i], itr.next());
    }
    assertFalse(itr.hasNext());
}
Also used : PigServer(org.apache.pig.PigServer) ArrayList(java.util.ArrayList) Test(org.junit.Test) HCatBaseTest(org.apache.hive.hcatalog.mapreduce.HCatBaseTest)

Example 78 with PigServer

use of org.apache.pig.PigServer in project hive by apache.

the class AbstractHCatStorerTest method testDateCharTypes.

/**
 * Create a data file with datatypes added in 0.13. Read it with Pig and use Pig + HCatStorer to
 * write to a Hive table. Then read it using Pig and Hive and make sure results match.
 */
@Test
public void testDateCharTypes() throws Exception {
    final String tblName = "junit_date_char";
    AbstractHCatLoaderTest.dropTable(tblName, driver);
    AbstractHCatLoaderTest.createTableDefaultDB(tblName, "id int, char5 char(5), varchar10 varchar(10), dec52 decimal(5,2)", null, driver, storageFormat);
    int NUM_ROWS = 5;
    String[] rows = new String[NUM_ROWS];
    for (int i = 0; i < NUM_ROWS; i++) {
        // since the file is read by Pig, we need to make sure the values are in format that Pig
        // understands
        // otherwise it will turn the value to NULL on read
        rows[i] = i + "\txxxxx\tyyy\t" + 5.2;
    }
    HcatTestUtils.createTestDataFile(INPUT_FILE_NAME, rows);
    LOG.debug("File=" + INPUT_FILE_NAME);
    // dumpFile(INPUT_FILE_NAME);
    PigServer server = createPigServer(true);
    int queryNumber = 1;
    logAndRegister(server, "A = load '" + INPUT_FILE_NAME + "' as (id:int, char5:chararray, varchar10:chararray, dec52:bigdecimal);", queryNumber++);
    logAndRegister(server, "store A into '" + tblName + "' using " + HCatStorer.class.getName() + "();", queryNumber++);
    logAndRegister(server, "B = load '" + tblName + "' using " + HCatLoader.class.getName() + "();", queryNumber);
    try {
        driver.run("select * from " + tblName);
    } catch (CommandProcessorException e) {
        LOG.debug("cpr.respCode=" + e.getResponseCode() + " cpr.errMsg=" + e.getMessage());
    }
    List l = new ArrayList();
    driver.getResults(l);
    LOG.debug("Dumping rows via SQL from " + tblName);
    /*
     * Unfortunately Timestamp.toString() adjusts the value for local TZ and 't' is a String thus
     * the timestamp in 't' doesn't match rawData
     */
    for (Object t : l) {
        LOG.debug(t == null ? null : t.toString());
    }
    Iterator<Tuple> itr = server.openIterator("B");
    int numRowsRead = 0;
    while (itr.hasNext()) {
        Tuple t = itr.next();
        StringBuilder rowFromPig = new StringBuilder();
        for (int i = 0; i < t.size(); i++) {
            rowFromPig.append(t.get(i)).append("\t");
        }
        rowFromPig.setLength(rowFromPig.length() - 1);
        assertEquals("Comparing Pig to Raw data", rows[numRowsRead], rowFromPig.toString());
        // see comment at "Dumping rows via SQL..." for why this doesn't work (for all types)
        // assertEquals("Comparing Pig to Hive", rowFromPig.toString(), l.get(numRowsRead));
        numRowsRead++;
    }
    assertEquals("Expected " + NUM_ROWS + " rows; got " + numRowsRead + " file=" + INPUT_FILE_NAME, NUM_ROWS, numRowsRead);
}
Also used : CommandProcessorException(org.apache.hadoop.hive.ql.processors.CommandProcessorException) ArrayList(java.util.ArrayList) PigServer(org.apache.pig.PigServer) ArrayList(java.util.ArrayList) List(java.util.List) Tuple(org.apache.pig.data.Tuple) Test(org.junit.Test) HCatBaseTest(org.apache.hive.hcatalog.mapreduce.HCatBaseTest)

Example 79 with PigServer

use of org.apache.pig.PigServer in project hive by apache.

the class AbstractHCatStorerTest method testStaticPartitioningMultiPartCols.

@Test
public void testStaticPartitioningMultiPartCols() throws Exception {
    AbstractHCatLoaderTest.dropTable("employee", driver);
    AbstractHCatLoaderTest.createTableDefaultDB("employee", "emp_id INT, emp_name STRING, emp_start_date STRING , emp_gender STRING", "emp_country STRING , emp_state STRING", driver, storageFormat);
    String[] inputData = { "111237\tKrishna\t01/01/1990\tM\tIN\tKA", "111238\tKalpana\t01/01/2000\tF\tIN\tKA", "111239\tSatya\t01/01/2001\tM\tIN\tKA", "111240\tKavya\t01/01/2002\tF\tIN\tKA" };
    HcatTestUtils.createTestDataFile(INPUT_FILE_NAME, inputData);
    PigServer pig = createPigServer(false);
    pig.setBatchOn();
    pig.registerQuery("A = LOAD '" + INPUT_FILE_NAME + "' USING PigStorage() AS (emp_id:int,emp_name:chararray,emp_start_date:chararray," + "emp_gender:chararray,emp_country:chararray,emp_state:chararray);");
    pig.registerQuery("IN = FILTER A BY emp_country == 'IN' AND emp_state== 'KA';");
    pig.registerQuery("STORE IN INTO 'employee' USING " + HCatStorer.class.getName() + "('emp_country=IN, emp_state=KA');");
    pig.executeBatch();
    driver.run("select * from employee");
    ArrayList<String> results = new ArrayList<String>();
    driver.getResults(results);
    assertEquals(4, results.size());
    Collections.sort(results);
    assertEquals(inputData[0], results.get(0));
    assertEquals(inputData[1], results.get(1));
    assertEquals(inputData[2], results.get(2));
    assertEquals(inputData[3], results.get(3));
    driver.run("drop table employee");
}
Also used : PigServer(org.apache.pig.PigServer) ArrayList(java.util.ArrayList) Test(org.junit.Test) HCatBaseTest(org.apache.hive.hcatalog.mapreduce.HCatBaseTest)

Example 80 with PigServer

use of org.apache.pig.PigServer in project hive by apache.

the class AbstractHCatStorerTest method testStoreInPartiitonedTbl.

@Test
public void testStoreInPartiitonedTbl() throws Exception {
    AbstractHCatLoaderTest.dropTable("junit_unparted", driver);
    AbstractHCatLoaderTest.createTableDefaultDB("junit_unparted", "a int", "b string", driver, storageFormat);
    int LOOP_SIZE = 11;
    String[] input = new String[LOOP_SIZE];
    for (int i = 0; i < LOOP_SIZE; i++) {
        input[i] = i + "";
    }
    HcatTestUtils.createTestDataFile(INPUT_FILE_NAME, input);
    PigServer server = createPigServer(false);
    server.registerQuery("A = load '" + INPUT_FILE_NAME + "' as (a:int);");
    server.registerQuery("store A into 'default.junit_unparted' using " + HCatStorer.class.getName() + "('b=1');");
    server.registerQuery("B = load 'default.junit_unparted' using " + HCatLoader.class.getName() + "();");
    Iterator<Tuple> itr = server.openIterator("B");
    int i = 0;
    while (itr.hasNext()) {
        Tuple t = itr.next();
        assertEquals(2, t.size());
        assertEquals(t.get(0), i);
        assertEquals(t.get(1), "1");
        i++;
    }
    assertFalse(itr.hasNext());
    assertEquals(11, i);
    // verify the scratch directories has been cleaned up
    Path path = new Path(client.getTable("default", "junit_unparted").getSd().getLocation());
    FileSystem fs = path.getFileSystem(hiveConf);
    assertEquals(1, fs.listStatus(path).length);
}
Also used : Path(org.apache.hadoop.fs.Path) PigServer(org.apache.pig.PigServer) FileSystem(org.apache.hadoop.fs.FileSystem) Tuple(org.apache.pig.data.Tuple) Test(org.junit.Test) HCatBaseTest(org.apache.hive.hcatalog.mapreduce.HCatBaseTest)

Aggregations

PigServer (org.apache.pig.PigServer)115 Tuple (org.apache.pig.data.Tuple)74 ArrayList (java.util.ArrayList)70 Test (org.junit.Test)59 HCatBaseTest (org.apache.hive.hcatalog.mapreduce.HCatBaseTest)37 Data (org.apache.pig.builtin.mock.Storage.Data)15 File (java.io.File)14 Schema (org.apache.pig.impl.logicalLayer.schema.Schema)14 FieldSchema (org.apache.pig.impl.logicalLayer.schema.Schema.FieldSchema)9 Properties (java.util.Properties)8 Vector (java.util.Vector)8 HCatFieldSchema (org.apache.hive.hcatalog.data.schema.HCatFieldSchema)6 Path (org.apache.hadoop.fs.Path)4 FileWriter (java.io.FileWriter)3 List (java.util.List)3 Map (java.util.Map)3 Admin (org.apache.hadoop.hbase.client.Admin)3 Connection (org.apache.hadoop.hbase.client.Connection)3 Pair (org.apache.hive.hcatalog.data.Pair)3 ExecJob (org.apache.pig.backend.executionengine.ExecJob)3