Search in sources :

Example 1 with HiveStoragePredicateHandler

use of org.apache.hadoop.hive.ql.metadata.HiveStoragePredicateHandler in project hive by apache.

the class OpProcFactory method pushFilterToStorageHandler.

/**
 * Attempts to push a predicate down into a storage handler.  For
 * native tables, this is a no-op.
 *
 * @param tableScanOp table scan against which predicate applies
 *
 * @param originalPredicate predicate to be pushed down
 *
 * @param owi object walk info
 *
 * @param hiveConf Hive configuration
 *
 * @return portion of predicate which needs to be evaluated
 * by Hive as a post-filter, or null if it was possible
 * to push down the entire predicate
 */
private static ExprNodeGenericFuncDesc pushFilterToStorageHandler(TableScanOperator tableScanOp, ExprNodeGenericFuncDesc originalPredicate, OpWalkerInfo owi, HiveConf hiveConf) throws SemanticException {
    TableScanDesc tableScanDesc = tableScanOp.getConf();
    Table tbl = tableScanDesc.getTableMetadata();
    if (HiveConf.getBoolVar(hiveConf, HiveConf.ConfVars.HIVEOPTINDEXFILTER)) {
        // attach the original predicate to the table scan operator for index
        // optimizations that require the pushed predicate before pcr & later
        // optimizations are applied
        tableScanDesc.setFilterExpr(originalPredicate);
    }
    if (!tbl.isNonNative()) {
        return originalPredicate;
    }
    HiveStorageHandler storageHandler = tbl.getStorageHandler();
    if (!(storageHandler instanceof HiveStoragePredicateHandler)) {
        // The storage handler does not provide predicate decomposition
        // support, so we'll implement the entire filter in Hive.  However,
        // we still provide the full predicate to the storage handler in
        // case it wants to do any of its own prefiltering.
        tableScanDesc.setFilterExpr(originalPredicate);
        return originalPredicate;
    }
    HiveStoragePredicateHandler predicateHandler = (HiveStoragePredicateHandler) storageHandler;
    JobConf jobConf = new JobConf(owi.getParseContext().getConf());
    Utilities.setColumnNameList(jobConf, tableScanOp);
    Utilities.setColumnTypeList(jobConf, tableScanOp);
    try {
        Utilities.copyTableJobPropertiesToConf(Utilities.getTableDesc(tbl), jobConf);
    } catch (Exception e) {
        throw new SemanticException(e);
    }
    Deserializer deserializer = tbl.getDeserializer();
    HiveStoragePredicateHandler.DecomposedPredicate decomposed = predicateHandler.decomposePredicate(jobConf, deserializer, originalPredicate);
    if (decomposed == null) {
        // not able to push anything down
        if (LOG.isDebugEnabled()) {
            LOG.debug("No pushdown possible for predicate:  " + originalPredicate.getExprString());
        }
        return originalPredicate;
    }
    if (LOG.isDebugEnabled()) {
        LOG.debug("Original predicate:  " + originalPredicate.getExprString());
        if (decomposed.pushedPredicate != null) {
            LOG.debug("Pushed predicate:  " + decomposed.pushedPredicate.getExprString());
        }
        if (decomposed.residualPredicate != null) {
            LOG.debug("Residual predicate:  " + decomposed.residualPredicate.getExprString());
        }
    }
    tableScanDesc.setFilterExpr(decomposed.pushedPredicate);
    tableScanDesc.setFilterObject(decomposed.pushedPredicateObject);
    return decomposed.residualPredicate;
}
Also used : HiveStoragePredicateHandler(org.apache.hadoop.hive.ql.metadata.HiveStoragePredicateHandler) HiveStorageHandler(org.apache.hadoop.hive.ql.metadata.HiveStorageHandler) Table(org.apache.hadoop.hive.ql.metadata.Table) Deserializer(org.apache.hadoop.hive.serde2.Deserializer) TableScanDesc(org.apache.hadoop.hive.ql.plan.TableScanDesc) JobConf(org.apache.hadoop.mapred.JobConf) SemanticException(org.apache.hadoop.hive.ql.parse.SemanticException) SemanticException(org.apache.hadoop.hive.ql.parse.SemanticException)

Example 2 with HiveStoragePredicateHandler

use of org.apache.hadoop.hive.ql.metadata.HiveStoragePredicateHandler in project hive by apache.

the class HiveInputFormat method pushFiltersAndAsOf.

public static void pushFiltersAndAsOf(JobConf jobConf, TableScanOperator tableScan, final MapWork mrwork) {
    // Push as of information
    pushAsOf(jobConf, tableScan);
    // ensure filters are not set from previous pushFilters
    jobConf.unset(TableScanDesc.FILTER_TEXT_CONF_STR);
    jobConf.unset(TableScanDesc.FILTER_EXPR_CONF_STR);
    Utilities.unsetSchemaEvolution(jobConf);
    TableScanDesc scanDesc = tableScan.getConf();
    if (scanDesc == null) {
        return;
    }
    Utilities.addTableSchemaToConf(jobConf, tableScan);
    // construct column name list and types for reference by filter push down
    Utilities.setColumnNameList(jobConf, tableScan);
    Utilities.setColumnTypeList(jobConf, tableScan);
    // push down filters
    ExprNodeGenericFuncDesc filterExpr = scanDesc.getFilterExpr();
    String pruningFilter = jobConf.get(TableScanDesc.PARTITION_PRUNING_FILTER);
    // If we have a pruning filter then combine it with the original
    if (pruningFilter != null) {
        ExprNodeGenericFuncDesc pruningExpr = SerializationUtilities.deserializeExpression(pruningFilter);
        if (filterExpr != null) {
            // Combine the 2 filters with AND
            filterExpr = ExprNodeDescUtils.and(filterExpr, pruningExpr);
        } else {
            // Use the pruning filter if there was no filter before
            filterExpr = pruningExpr;
        }
        // Set the combined filter in the TableScanDesc and remove the pruning filter
        scanDesc.setFilterExpr(filterExpr);
        scanDesc.setSerializedFilterExpr(SerializationUtilities.serializeExpression(filterExpr));
        jobConf.unset(TableScanDesc.PARTITION_PRUNING_FILTER);
    }
    if (filterExpr == null) {
        return;
    }
    // since we don't clone jobConf per alias
    if (mrwork != null && mrwork.getAliases() != null && mrwork.getAliases().size() > 1 && jobConf.get(ConfVars.HIVE_EXECUTION_ENGINE.varname).equals("mr") && (scanDesc.getTableMetadata() == null || !(scanDesc.getTableMetadata().getStorageHandler() instanceof HiveStoragePredicateHandler))) {
        return;
    }
    String serializedFilterObj = scanDesc.getSerializedFilterObject();
    String serializedFilterExpr = scanDesc.getSerializedFilterExpr();
    boolean hasObj = serializedFilterObj != null, hasExpr = serializedFilterExpr != null;
    if (!hasObj) {
        Serializable filterObject = scanDesc.getFilterObject();
        if (filterObject != null) {
            serializedFilterObj = SerializationUtilities.serializeObject(filterObject);
        }
    }
    if (serializedFilterObj != null) {
        jobConf.set(TableScanDesc.FILTER_OBJECT_CONF_STR, serializedFilterObj);
    }
    if (!hasExpr) {
        serializedFilterExpr = SerializationUtilities.serializeExpression(filterExpr);
    }
    String filterText = filterExpr.getExprString();
    if (LOG.isDebugEnabled()) {
        LOG.debug("Pushdown initiated with filterText = " + filterText + ", filterExpr = " + filterExpr + ", serializedFilterExpr = " + serializedFilterExpr + " (" + (hasExpr ? "desc" : "new") + ")" + (serializedFilterObj == null ? "" : (", serializedFilterObj = " + serializedFilterObj + " (" + (hasObj ? "desc" : "new") + ")")));
    }
    jobConf.set(TableScanDesc.FILTER_TEXT_CONF_STR, filterText);
    jobConf.set(TableScanDesc.FILTER_EXPR_CONF_STR, serializedFilterExpr);
}
Also used : HiveStoragePredicateHandler(org.apache.hadoop.hive.ql.metadata.HiveStoragePredicateHandler) Serializable(java.io.Serializable) TableScanDesc(org.apache.hadoop.hive.ql.plan.TableScanDesc) ExprNodeGenericFuncDesc(org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc)

Aggregations

HiveStoragePredicateHandler (org.apache.hadoop.hive.ql.metadata.HiveStoragePredicateHandler)2 TableScanDesc (org.apache.hadoop.hive.ql.plan.TableScanDesc)2 Serializable (java.io.Serializable)1 HiveStorageHandler (org.apache.hadoop.hive.ql.metadata.HiveStorageHandler)1 Table (org.apache.hadoop.hive.ql.metadata.Table)1 SemanticException (org.apache.hadoop.hive.ql.parse.SemanticException)1 ExprNodeGenericFuncDesc (org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc)1 Deserializer (org.apache.hadoop.hive.serde2.Deserializer)1 JobConf (org.apache.hadoop.mapred.JobConf)1