Package org.apache.hadoop.hive.ql.exec

Examples of org.apache.hadoop.hive.ql.exec.TableScanOperator

124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
    }

    @Override
    public Object process(Node nd, Stack<Node> stack, NodeProcessorCtx procCtx,
        Object... nodeOutputs) throws SemanticException {
      TableScanOperator node = (TableScanOperator) nd;
      WalkerCtx walkerCtx = (WalkerCtx) procCtx;
      if (((node.getNeededColumnIDs() == null) || (node.getNeededColumnIDs().size() == 0))
          && ((node.getConf() == null) ||
              (node.getConf().getVirtualCols() == null) ||
              (node.getConf().getVirtualCols().isEmpty()))) {
        walkerCtx.setMayBeMetadataOnly(node);
      }
      return nd;
    }
View Full Code Here
297
298
299
300
301
302
303
304
305
306
307
308
          walkerCtx.getMetadataOnlyTableScans().size()));
      Iterator<TableScanOperator> iterator
        = walkerCtx.getMetadataOnlyTableScans().iterator();

      while (iterator.hasNext()) {
        TableScanOperator tso = iterator.next();
        LOG.info("Metadata only table scan for " + tso.getConf().getAlias());
        convertToMetadataOnlyQuery((MapredWork) task.getWork(), tso);
      }

      return null;
    }
View Full Code Here
236
237
238
239
240
241
242
243
244
245
246
      Map<String, Operator<? extends OperatorDesc>> topOps = this.pGraphContext
          .getTopOps();
      Map<TableScanOperator, Table> topToTable = this.pGraphContext
          .getTopToTable();
      TableScanOperator tso = (TableScanOperator) topOps.get(alias);
      if (tso == null) {
        return false;
      }

      List<ExprNodeDesc> keys = op.getConf().getKeys().get((byte) pos);
View Full Code Here
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
   * The Node Processor for Column Pruning on Table Scan Operators. It will store
   * needed columns in tableScanDesc.
   */
  public static class ColumnPrunerTableScanProc implements NodeProcessor {
    public Object process(Node nd, Stack<Node> stack, NodeProcessorCtx ctx, Object... nodeOutputs) throws SemanticException {
      TableScanOperator scanOp = (TableScanOperator)nd;
      ColumnPrunerProcCtx cppCtx = (ColumnPrunerProcCtx)ctx;
      List<String> cols = cppCtx.genColLists((Operator<? extends Serializable>)nd);
      cppCtx.getPrunedColLists().put((Operator<? extends Serializable>)nd, cols);
      ArrayList<Integer> needed_columns = new ArrayList<Integer>();
      RowResolver inputRR = cppCtx.getOpToParseCtxMap().get(scanOp).getRR();
      for (int i = 0; i < cols.size(); i++) {
        int position = inputRR.getPosition(cols.get(i));
        needed_columns.add(position);
      }
      scanOp.setNeededColumnIDs(needed_columns);
      return null;
    }
View Full Code Here
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
      // symbol. So we just pop out the two elements from the top and if the
      // second one of them is not a table scan then the operator on the top of
      // the stack is the Table scan operator.
      Node tmp = stack.pop();
      Node tmp2 = stack.pop();
      TableScanOperator top = null;
      if (tmp2 instanceof TableScanOperator) {
        top = (TableScanOperator)tmp2;
      }
      else {
        top = (TableScanOperator) stack.peek();
        fop2 = (FilterOperator) tmp2;
      }
      stack.push(tmp2);
      stack.push(tmp);

      // If fop2 exists (i.e this is not the top level filter and fop2 is not
      // a sampling filter then we ignore the current filter
      if (fop2 != null && !fop2.getConf().getIsSamplingPred())
        return null;
     
      // ignore the predicate in case it is not a sampling predicate
      if (fop.getConf().getIsSamplingPred()) {
        return null;
      }
     
      // Otherwise this is not a sampling predicate and we need to
      exprNodeDesc predicate = fop.getConf().getPredicate();
      String alias = top.getConf().getAlias();
     
      // Generate the partition pruning predicate
      boolean hasNonPartCols = false;
      exprNodeDesc ppr_pred = ExprProcFactory.genPruner(alias, predicate, hasNonPartCols);
      owc.addHasNonPartCols(hasNonPartCols);
View Full Code Here
223
224
225
226
227
228
229
230
231
232
233
234
    }
    for (String alias : aliases) {
      Operator<? extends Serializable> op = this.mrwork.getAliasToWork().get(
          alias);
      if (op instanceof TableScanOperator) {
        TableScanOperator tableScan = (TableScanOperator) op;
        ArrayList<Integer> list = tableScan.getNeededColumnIDs();
        if (list != null)
          HiveFileFormatUtils.setReadColumnIDs(jobConf, list);
        else
          HiveFileFormatUtils.setFullyReadColumns(jobConf);
      }
View Full Code Here
92
93
94
95
96
97
98
99
100
101
    public Object process(Node nd, Stack<Node> stack, NodeProcessorCtx procCtx,
        Object... nodeOutputs) throws SemanticException {
      LOG.info("Processing for " +  nd.getName() + "(" + ((Operator)nd).getIdentifier() + ")");
      OpWalkerInfo owi = (OpWalkerInfo)procCtx;
      RowResolver inputRR = owi.getRowResolver(nd);
      TableScanOperator tsOp = (TableScanOperator)nd;
      mergeWithChildrenPred(tsOp, owi, null, null, false);
      ExprWalkerInfo pushDownPreds = owi.getPrunedPreds(tsOp);
      return createFilter(tsOp, pushDownPreds, owi);
    }
View Full Code Here
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
      // LineageCtx
      LineageCtx lCtx = (LineageCtx) procCtx;
      ParseContext pctx = lCtx.getParseCtx();

      // Table scan operator.
      TableScanOperator top = (TableScanOperator)nd;
      org.apache.hadoop.hive.ql.metadata.Table t = pctx.getTopToTable().get(top);
      Table tab = t.getTTable();

      // Generate the mappings
      RowSchema rs = top.getSchema();
      List<FieldSchema> cols = t.getAllCols();
      Map<String, FieldSchema> fieldSchemaMap = new HashMap<String, FieldSchema>();
      for(FieldSchema col : cols) {
        fieldSchemaMap.put(col.getName(), col);
      }

      Iterator<VirtualColumn> vcs = VirtualColumn.getRegistry(pctx.getConf()).iterator();
      while (vcs.hasNext()) {
        VirtualColumn vc = vcs.next();
        fieldSchemaMap.put(vc.getName(), new FieldSchema(vc.getName(),
            vc.getTypeInfo().getTypeName(), ""));
      }

      TableAliasInfo tai = new TableAliasInfo();
      tai.setAlias(top.getConf().getAlias());
      tai.setTable(tab);
      for(ColumnInfo ci : rs.getSignature()) {
        // Create a dependency
        Dependency dep = new Dependency();
        BaseColumnInfo bci = new BaseColumnInfo();
View Full Code Here
192
193
194
195
196
197
198
199
200
201
202
      LinkedHashMap<Partition, List<String>> bigTblPartsToBucketFileNames = new LinkedHashMap<Partition, List<String>>();
      LinkedHashMap<Partition, Integer> bigTblPartsToBucketNumber = new LinkedHashMap<Partition, Integer>();

      for (int index = 0; index < joinAliases.size(); index++) {
        String alias = joinAliases.get(index);
        TableScanOperator tso = (TableScanOperator) topOps.get(alias);
        if (tso == null) {
          return null;
        }
        Table tbl = topToTable.get(tso);
        if(tbl.isPartitioned()) {
View Full Code Here
137
138
139
140
141
142
143
144
145
      }

      assert (stack.size() == 3 && stack.get(1) instanceof FilterOperator) ||
          stack.size() == 2;

      TableScanOperator tsOp = (TableScanOperator) stack.get(0);
      ((SamplePrunerCtx) procCtx).getOpToSamplePruner().put(tsOp, sampleDescr);
      return null;
    }
View Full Code Here
TOP

Related Classes of org.apache.hadoop.hive.ql.exec.TableScanOperator

  • org.apache.hadoop.fs.Path
  • org.apache.hadoop.hive.ql.Driver
  • org.apache.hadoop.hive.ql.exec.mr.ExecDriver
  • org.apache.hadoop.hive.ql.exec.mr.MapredLocalTask
  • org.apache.hadoop.hive.ql.io.HiveInputFormat
  • org.apache.hadoop.hive.ql.io.parquet.ProjectionPusher
  • org.apache.hadoop.hive.ql.optimizer.AbstractBucketJoinProc
  • org.apache.hadoop.hive.ql.optimizer.AbstractSMBJoinProc
  • org.apache.hadoop.hive.ql.optimizer.BucketingSortingReduceSinkOptimizer$BucketSortReduceSinkProcessor
  • org.apache.hadoop.hive.ql.optimizer.BucketMapJoinOptimizer$BucketMapjoinOptProc
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.