Examples with FileSinkDesc org.apache.hadoop.hive.ql.plan.FileSinkDesc used on opensource projects

Example 16 with FileSinkDesc

use of org.apache.hadoop.hive.ql.plan.FileSinkDesc in project hive by apache.

the class TestGenMapRedUtilsCreateConditionalTask method createFileSinkOperator.

private FileSinkOperator createFileSinkOperator(Path finalDirName) {
    FileSinkOperator fileSinkOperator = mock(FileSinkOperator.class);
    TableDesc tableDesc = new TableDesc(HiveInputFormat.class, HiveOutputFormat.class, new Properties());
    FileSinkDesc fileSinkDesc = new FileSinkDesc(finalDirName, tableDesc, false);
    fileSinkDesc.setDirName(finalDirName);
    when(fileSinkOperator.getConf()).thenReturn(fileSinkDesc);
    when(fileSinkOperator.getSchema()).thenReturn(mock(RowSchema.class));
    fileSinkDesc.setTableInfo(tableDesc);
    when(fileSinkOperator.getCompilationOpContext()).thenReturn(mock(CompilationOpContext.class));
    return fileSinkOperator;
}

Example 17 with FileSinkDesc

use of org.apache.hadoop.hive.ql.plan.FileSinkDesc in project hive by apache.

the class TestGenTezWork method setUp.

/**
 * @throws java.lang.Exception
 */
@SuppressWarnings("unchecked")
@Before
public void setUp() throws Exception {
    // Init conf
    final HiveConf conf = new HiveConf(SemanticAnalyzer.class);
    SessionState.start(conf);
    // Init parse context
    final ParseContext pctx = new ParseContext();
    pctx.setContext(new Context(conf));
    ctx = new GenTezProcContext(conf, pctx, Collections.EMPTY_LIST, new ArrayList<Task<? extends Serializable>>(), Collections.EMPTY_SET, Collections.EMPTY_SET);
    proc = new GenTezWork(new GenTezUtils() {

        @Override
        protected void setupMapWork(MapWork mapWork, GenTezProcContext context, PrunedPartitionList partitions, TableScanOperator root, String alias) throws SemanticException {
            LinkedHashMap<String, Operator<? extends OperatorDesc>> map = new LinkedHashMap<String, Operator<? extends OperatorDesc>>();
            map.put("foo", root);
            mapWork.setAliasToWork(map);
            return;
        }
    });
    CompilationOpContext cCtx = new CompilationOpContext();
    fs = new FileSinkOperator(cCtx);
    fs.setConf(new FileSinkDesc());
    rs = new ReduceSinkOperator(cCtx);
    rs.setConf(new ReduceSinkDesc());
    TableDesc tableDesc = new TableDesc();
    tableDesc.setProperties(new Properties());
    rs.getConf().setKeySerializeInfo(tableDesc);
    ts = new TableScanOperator(cCtx);
    ts.setConf(new TableScanDesc(null));
    ts.getChildOperators().add(rs);
    rs.getParentOperators().add(ts);
    rs.getChildOperators().add(fs);
    fs.getParentOperators().add(rs);
    ctx.preceedingWork = null;
    ctx.currentRootOperator = ts;
}

Example 18 with FileSinkDesc

use of org.apache.hadoop.hive.ql.plan.FileSinkDesc in project hive by apache.

the class Vectorizer method vectorizeOperator.

public Operator<? extends OperatorDesc> vectorizeOperator(Operator<? extends OperatorDesc> op, VectorizationContext vContext, boolean isTezOrSpark, VectorTaskColumnInfo vectorTaskColumnInfo) throws HiveException {
    Operator<? extends OperatorDesc> vectorOp = null;
    boolean isNative;
    switch(op.getType()) {
        case TABLESCAN:
            vectorOp = vectorizeTableScanOperator(op, vContext);
            isNative = true;
            break;
        case MAPJOIN:
            {
                if (op instanceof MapJoinOperator) {
                    VectorMapJoinInfo vectorMapJoinInfo = new VectorMapJoinInfo();
                    MapJoinDesc desc = (MapJoinDesc) op.getConf();
                    boolean specialize = canSpecializeMapJoin(op, desc, isTezOrSpark, vContext, vectorMapJoinInfo);
                    if (!specialize) {
                        Class<? extends Operator<?>> opClass = null;
                        // *NON-NATIVE* vector map differences for LEFT OUTER JOIN and Filtered...
                        List<ExprNodeDesc> bigTableFilters = desc.getFilters().get((byte) desc.getPosBigTable());
                        boolean isOuterAndFiltered = (!desc.isNoOuterJoin() && bigTableFilters.size() > 0);
                        if (!isOuterAndFiltered) {
                            opClass = VectorMapJoinOperator.class;
                        } else {
                            opClass = VectorMapJoinOuterFilteredOperator.class;
                        }
                        vectorOp = OperatorFactory.getVectorOperator(opClass, op.getCompilationOpContext(), op.getConf(), vContext);
                        isNative = false;
                    } else {
                        // TEMPORARY Until Native Vector Map Join with Hybrid passes tests...
                        // HiveConf.setBoolVar(physicalContext.getConf(),
                        //    HiveConf.ConfVars.HIVEUSEHYBRIDGRACEHASHJOIN, false);
                        vectorOp = specializeMapJoinOperator(op, vContext, desc, vectorMapJoinInfo);
                        isNative = true;
                        if (vectorTaskColumnInfo != null) {
                            if (usesVectorUDFAdaptor(vectorMapJoinInfo.getBigTableKeyExpressions())) {
                                vectorTaskColumnInfo.setUsesVectorUDFAdaptor(true);
                            }
                            if (usesVectorUDFAdaptor(vectorMapJoinInfo.getBigTableValueExpressions())) {
                                vectorTaskColumnInfo.setUsesVectorUDFAdaptor(true);
                            }
                        }
                    }
                } else {
                    Preconditions.checkState(op instanceof SMBMapJoinOperator);
                    SMBJoinDesc smbJoinSinkDesc = (SMBJoinDesc) op.getConf();
                    VectorSMBJoinDesc vectorSMBJoinDesc = new VectorSMBJoinDesc();
                    smbJoinSinkDesc.setVectorDesc(vectorSMBJoinDesc);
                    vectorOp = OperatorFactory.getVectorOperator(op.getCompilationOpContext(), smbJoinSinkDesc, vContext);
                    isNative = false;
                }
            }
            break;
        case REDUCESINK:
            {
                VectorReduceSinkInfo vectorReduceSinkInfo = new VectorReduceSinkInfo();
                ReduceSinkDesc desc = (ReduceSinkDesc) op.getConf();
                boolean specialize = canSpecializeReduceSink(desc, isTezOrSpark, vContext, vectorReduceSinkInfo);
                if (!specialize) {
                    vectorOp = OperatorFactory.getVectorOperator(op.getCompilationOpContext(), op.getConf(), vContext);
                    isNative = false;
                } else {
                    vectorOp = specializeReduceSinkOperator(op, vContext, desc, vectorReduceSinkInfo);
                    isNative = true;
                    if (vectorTaskColumnInfo != null) {
                        if (usesVectorUDFAdaptor(vectorReduceSinkInfo.getReduceSinkKeyExpressions())) {
                            vectorTaskColumnInfo.setUsesVectorUDFAdaptor(true);
                        }
                        if (usesVectorUDFAdaptor(vectorReduceSinkInfo.getReduceSinkValueExpressions())) {
                            vectorTaskColumnInfo.setUsesVectorUDFAdaptor(true);
                        }
                    }
                }
            }
            break;
        case FILTER:
            {
                vectorOp = vectorizeFilterOperator(op, vContext);
                isNative = true;
                if (vectorTaskColumnInfo != null) {
                    VectorFilterDesc vectorFilterDesc = (VectorFilterDesc) ((AbstractOperatorDesc) vectorOp.getConf()).getVectorDesc();
                    VectorExpression vectorPredicateExpr = vectorFilterDesc.getPredicateExpression();
                    if (usesVectorUDFAdaptor(vectorPredicateExpr)) {
                        vectorTaskColumnInfo.setUsesVectorUDFAdaptor(true);
                    }
                }
            }
            break;
        case SELECT:
            {
                vectorOp = vectorizeSelectOperator(op, vContext);
                isNative = true;
                if (vectorTaskColumnInfo != null) {
                    VectorSelectDesc vectorSelectDesc = (VectorSelectDesc) ((AbstractOperatorDesc) vectorOp.getConf()).getVectorDesc();
                    VectorExpression[] vectorSelectExprs = vectorSelectDesc.getSelectExpressions();
                    if (usesVectorUDFAdaptor(vectorSelectExprs)) {
                        vectorTaskColumnInfo.setUsesVectorUDFAdaptor(true);
                    }
                }
            }
            break;
        case GROUPBY:
            {
                vectorOp = vectorizeGroupByOperator(op, vContext);
                isNative = false;
                if (vectorTaskColumnInfo != null) {
                    VectorGroupByDesc vectorGroupByDesc = (VectorGroupByDesc) ((AbstractOperatorDesc) vectorOp.getConf()).getVectorDesc();
                    if (!vectorGroupByDesc.isVectorOutput()) {
                        vectorTaskColumnInfo.setGroupByVectorOutput(false);
                    }
                    VectorExpression[] vecKeyExpressions = vectorGroupByDesc.getKeyExpressions();
                    if (usesVectorUDFAdaptor(vecKeyExpressions)) {
                        vectorTaskColumnInfo.setUsesVectorUDFAdaptor(true);
                    }
                    VectorAggregateExpression[] vecAggregators = vectorGroupByDesc.getAggregators();
                    for (VectorAggregateExpression vecAggr : vecAggregators) {
                        if (usesVectorUDFAdaptor(vecAggr.inputExpression())) {
                            vectorTaskColumnInfo.setUsesVectorUDFAdaptor(true);
                        }
                    }
                }
            }
            break;
        case FILESINK:
            {
                FileSinkDesc fileSinkDesc = (FileSinkDesc) op.getConf();
                VectorFileSinkDesc vectorFileSinkDesc = new VectorFileSinkDesc();
                fileSinkDesc.setVectorDesc(vectorFileSinkDesc);
                vectorOp = OperatorFactory.getVectorOperator(op.getCompilationOpContext(), fileSinkDesc, vContext);
                isNative = false;
            }
            break;
        case LIMIT:
            {
                LimitDesc limitDesc = (LimitDesc) op.getConf();
                VectorLimitDesc vectorLimitDesc = new VectorLimitDesc();
                limitDesc.setVectorDesc(vectorLimitDesc);
                vectorOp = OperatorFactory.getVectorOperator(op.getCompilationOpContext(), limitDesc, vContext);
                isNative = true;
            }
            break;
        case EVENT:
            {
                AppMasterEventDesc eventDesc = (AppMasterEventDesc) op.getConf();
                VectorAppMasterEventDesc vectorEventDesc = new VectorAppMasterEventDesc();
                eventDesc.setVectorDesc(vectorEventDesc);
                vectorOp = OperatorFactory.getVectorOperator(op.getCompilationOpContext(), eventDesc, vContext);
                isNative = true;
            }
            break;
        case HASHTABLESINK:
            {
                SparkHashTableSinkDesc sparkHashTableSinkDesc = (SparkHashTableSinkDesc) op.getConf();
                VectorSparkHashTableSinkDesc vectorSparkHashTableSinkDesc = new VectorSparkHashTableSinkDesc();
                sparkHashTableSinkDesc.setVectorDesc(vectorSparkHashTableSinkDesc);
                vectorOp = OperatorFactory.getVectorOperator(op.getCompilationOpContext(), sparkHashTableSinkDesc, vContext);
                isNative = true;
            }
            break;
        case SPARKPRUNINGSINK:
            {
                SparkPartitionPruningSinkDesc sparkPartitionPruningSinkDesc = (SparkPartitionPruningSinkDesc) op.getConf();
                VectorSparkPartitionPruningSinkDesc vectorSparkPartitionPruningSinkDesc = new VectorSparkPartitionPruningSinkDesc();
                sparkPartitionPruningSinkDesc.setVectorDesc(vectorSparkPartitionPruningSinkDesc);
                vectorOp = OperatorFactory.getVectorOperator(op.getCompilationOpContext(), sparkPartitionPruningSinkDesc, vContext);
                isNative = true;
            }
            break;
        default:
            // These are children of GROUP BY operators with non-vector outputs.
            isNative = false;
            vectorOp = op;
            break;
    }
    Preconditions.checkState(vectorOp != null);
    if (vectorTaskColumnInfo != null && !isNative) {
        vectorTaskColumnInfo.setAllNative(false);
    }
    LOG.debug("vectorizeOperator " + vectorOp.getClass().getName());
    LOG.debug("vectorizeOperator " + vectorOp.getConf().getClass().getName());
    if (vectorOp != op) {
        fixupParentChildOperators(op, vectorOp);
        ((AbstractOperatorDesc) vectorOp.getConf()).setVectorMode(true);
    }
    return vectorOp;
}

Example 19 with FileSinkDesc

use of org.apache.hadoop.hive.ql.plan.FileSinkDesc in project hive by apache.

the class Driver method acquireLocksAndOpenTxn.

/**
   * Acquire read and write locks needed by the statement. The list of objects to be locked are
   * obtained from the inputs and outputs populated by the compiler. The lock acquisition scheme is
   * pretty simple. If all the locks cannot be obtained, error out. Deadlock is avoided by making
   * sure that the locks are lexicographically sorted.
   *
   * This method also records the list of valid transactions.  This must be done after any
   * transactions have been opened and locks acquired.
   * @param startTxnImplicitly in AC=false, the 1st DML starts a txn
   **/
private int acquireLocksAndOpenTxn(boolean startTxnImplicitly) {
    PerfLogger perfLogger = SessionState.getPerfLogger();
    perfLogger.PerfLogBegin(CLASS_NAME, PerfLogger.ACQUIRE_READ_WRITE_LOCKS);
    SessionState ss = SessionState.get();
    HiveTxnManager txnMgr = ss.getTxnMgr();
    if (startTxnImplicitly) {
        assert !txnMgr.getAutoCommit();
    }
    try {
        // Don't use the userName member, as it may or may not have been set.  Get the value from
        // conf, which calls into getUGI to figure out who the process is running as.
        String userFromUGI;
        try {
            userFromUGI = conf.getUser();
        } catch (IOException e) {
            errorMessage = "FAILED: Error in determining user while acquiring locks: " + e.getMessage();
            SQLState = ErrorMsg.findSQLState(e.getMessage());
            downstreamError = e;
            console.printError(errorMessage, "\n" + org.apache.hadoop.util.StringUtils.stringifyException(e));
            return 10;
        }
        boolean initiatingTransaction = false;
        boolean readOnlyQueryInAutoCommit = false;
        if ((txnMgr.getAutoCommit() && haveAcidWrite()) || plan.getOperation() == HiveOperation.START_TRANSACTION || (!txnMgr.getAutoCommit() && startTxnImplicitly)) {
            if (txnMgr.isTxnOpen()) {
                throw new RuntimeException("Already have an open transaction txnid:" + txnMgr.getCurrentTxnId());
            }
            // We are writing to tables in an ACID compliant way, so we need to open a transaction
            txnMgr.openTxn(ctx, userFromUGI);
            initiatingTransaction = true;
        } else {
            readOnlyQueryInAutoCommit = txnMgr.getAutoCommit() && plan.getOperation() == HiveOperation.QUERY && !haveAcidWrite();
        }
        // Set the transaction id in all of the acid file sinks
        if (haveAcidWrite()) {
            for (FileSinkDesc desc : acidSinks) {
                desc.setTransactionId(txnMgr.getCurrentTxnId());
                //it's possible to have > 1 FileSink writing to the same table/partition
                //e.g. Merge stmt, multi-insert stmt when mixing DP and SP writes
                desc.setStatementId(txnMgr.getWriteIdAndIncrement());
            }
        }
        /*Note, we have to record snapshot after lock acquisition to prevent lost update problem
      consider 2 concurrent "update table T set x = x + 1".  1st will get the locks and the
      2nd will block until 1st one commits and only then lock in the snapshot, i.e. it will
      see the changes made by 1st one.  This takes care of autoCommit=true case.
      For multi-stmt txns this is not sufficient and will be managed via WriteSet tracking
      in the lock manager.*/
        txnMgr.acquireLocks(plan, ctx, userFromUGI, lDrvState);
        if (initiatingTransaction || (readOnlyQueryInAutoCommit && acidInQuery)) {
            //For multi-stmt txns we should record the snapshot when txn starts but
            // don't update it after that until txn completes.  Thus the check for {@code initiatingTransaction}
            //For autoCommit=true, Read-only statements, txn is implicit, i.e. lock in the snapshot
            //for each statement.
            recordValidTxns();
        }
        return 0;
    } catch (Exception e) {
        errorMessage = "FAILED: Error in acquiring locks: " + e.getMessage();
        SQLState = ErrorMsg.findSQLState(e.getMessage());
        downstreamError = e;
        console.printError(errorMessage, "\n" + org.apache.hadoop.util.StringUtils.stringifyException(e));
        return 10;
    } finally {
        perfLogger.PerfLogEnd(CLASS_NAME, PerfLogger.ACQUIRE_READ_WRITE_LOCKS);
    }
}

Example 20 with FileSinkDesc

use of org.apache.hadoop.hive.ql.plan.FileSinkDesc in project hive by apache.

the class TestExecDriver method populateMapRedPlan6.

@SuppressWarnings("unchecked")
private void populateMapRedPlan6(Table src) throws Exception {
    // map-side work
    ArrayList<String> outputColumns = new ArrayList<String>();
    for (int i = 0; i < 2; i++) {
        outputColumns.add("_col" + i);
    }
    Operator<ReduceSinkDesc> op1 = OperatorFactory.get(ctx, PlanUtils.getReduceSinkDesc(Utilities.makeList(getStringColumn("tkey")), Utilities.makeList(getStringColumn("tkey"), getStringColumn("tvalue")), outputColumns, false, -1, 1, -1, AcidUtils.Operation.NOT_ACID));
    Operator<ScriptDesc> op0 = OperatorFactory.get(new ScriptDesc("\'cat\'", PlanUtils.getDefaultTableDesc("" + Utilities.tabCode, "tkey,tvalue"), TextRecordWriter.class, PlanUtils.getDefaultTableDesc("" + Utilities.tabCode, "tkey,tvalue"), TextRecordReader.class, TextRecordReader.class, PlanUtils.getDefaultTableDesc("" + Utilities.tabCode, "key")), op1);
    Operator<SelectDesc> op4 = OperatorFactory.get(new SelectDesc(Utilities.makeList(getStringColumn("key"), getStringColumn("value")), outputColumns), op0);
    addMapWork(mr, src, "a", op4);
    ReduceWork rWork = new ReduceWork();
    mr.setReduceWork(rWork);
    rWork.setNumReduceTasks(Integer.valueOf(1));
    rWork.setKeyDesc(op1.getConf().getKeySerializeInfo());
    rWork.getTagToValueDesc().add(op1.getConf().getValueSerializeInfo());
    // reduce side work
    Operator<FileSinkDesc> op3 = OperatorFactory.get(ctx, new FileSinkDesc(new Path(tmpdir + File.separator + "mapredplan6.out"), Utilities.defaultTd, false));
    Operator<FilterDesc> op2 = OperatorFactory.get(getTestFilterDesc("0"), op3);
    List<ExprNodeDesc> cols = new ArrayList<ExprNodeDesc>();
    cols.add(getStringColumn(Utilities.ReduceField.KEY + ".reducesinkkey" + 0));
    cols.add(getStringColumn(Utilities.ReduceField.VALUE.toString() + "." + outputColumns.get(1)));
    Operator<SelectDesc> op5 = OperatorFactory.get(new SelectDesc(cols, outputColumns), op2);
    rWork.setReducer(op5);
}