Package org.apache.hadoop.hive.ql.plan

Examples of org.apache.hadoop.hive.ql.plan.FetchWork

282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
  protected void genColumnStatsTask(QB qb, List<LoadTableDesc> loadTableWork,
      List<LoadFileDesc> loadFileWork, List<Task<? extends Serializable>> rootTasks) {
    QBParseInfo qbParseInfo = qb.getParseInfo();
    ColumnStatsTask cStatsTask = null;
    ColumnStatsWork cStatsWork = null;
    FetchWork fetch = null;
    String tableName = qbParseInfo.getTableName();
    String partName = qbParseInfo.getPartName();
    List<String> colName = qbParseInfo.getColName();
    List<String> colType = qbParseInfo.getColType();
    boolean isTblLevel = qbParseInfo.isTblLvl();

    String cols = loadFileWork.get(0).getColumns();
    String colTypes = loadFileWork.get(0).getColumnTypes();

    String resFileFormat = HiveConf.getVar(conf, HiveConf.ConfVars.HIVEQUERYRESULTFILEFORMAT);
    TableDesc resultTab = PlanUtils.getDefaultQueryOutputTableDesc(cols, colTypes, resFileFormat);

    fetch = new FetchWork(loadFileWork.get(0).getSourcePath(),
        resultTab, qb.getParseInfo().getOuterQueryLimit());

    ColumnStatsDesc cStatsDesc = new ColumnStatsDesc(tableName, partName,
        colName, colType, isTblLevel);
    cStatsWork = new ColumnStatsWork(fetch, cStatsDesc);
View Full Code Here
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
      for (String path : emptyPath) {
        pathToAliases.remove(path);
      }

      // create fetch work
      FetchWork fetchWork = null;
      List<Path> partDir = new ArrayList<Path>();
      List<PartitionDesc> partDesc = new ArrayList<PartitionDesc>();

      for (String tablePath : pathSet) {
        PartitionDesc partitionDesc = newWork.getMapWork().getPathToPartitionInfo().get(tablePath);
        // create fetchwork for non partitioned table
        if (partitionDesc.getPartSpec() == null || partitionDesc.getPartSpec().size() == 0) {
          fetchWork = new FetchWork(new Path(tablePath), partitionDesc.getTableDesc());
          break;
        }
        // if table is partitioned,add partDir and partitionDesc
        partDir.add(new Path(tablePath));
        partDesc.add(partitionDesc);
      }
      // create fetchwork for partitioned table
      if (fetchWork == null) {
        TableDesc table = newWork.getMapWork().getAliasToPartnInfo().get(alias).getTableDesc();
        fetchWork = new FetchWork(partDir, partDesc, table);
      }
      // set alias to fetch work
      newLocalWork.getAliasToFetchWork().put(alias, fetchWork);
    }
    // remove small table ailias from aliasToWork;Avoid concurrent modification
View Full Code Here
493
494
495
496
497
498
499
500
501
502
503
504
505
        for (ColumnInfo colInfo: gbyOp.getSchema().getSignature()) {
          colNames.add(colInfo.getInternalName());
        }
        StandardStructObjectInspector sOI = ObjectInspectorFactory.
            getStandardStructObjectInspector(colNames, ois);
        FetchWork fWork = new FetchWork(allRows, sOI);
        FetchTask fTask = (FetchTask)TaskFactory.get(fWork, pctx.getConf());
        fWork.setLimit(allRows.size());
        pctx.setFetchTask(fTask);

        return null;
      } catch (Exception e) {
        // this is best effort optimization, bail out in error conditions and
View Full Code Here
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
    private FetchWork convertToWork() throws HiveException {
      inputs.clear();
      if (!table.isPartitioned()) {
        inputs.add(new ReadEntity(table, parent));
        FetchWork work = new FetchWork(table.getPath(), Utilities.getTableDesc(table));
        PlanUtils.configureInputJobPropertiesForStorageHandler(work.getTblDesc());
        work.setSplitSample(splitSample);
        return work;
      }
      List<Path> listP = new ArrayList<Path>();
      List<PartitionDesc> partP = new ArrayList<PartitionDesc>();

      for (Partition partition : partsList.getNotDeniedPartns()) {
        inputs.add(new ReadEntity(partition, parent));
        listP.add(partition.getDataLocation());
        partP.add(Utilities.getPartitionDesc(partition));
      }
      Table sourceTable = partsList.getSourceTable();
      inputs.add(new ReadEntity(sourceTable, parent));
      TableDesc table = Utilities.getTableDesc(sourceTable);
      FetchWork work = new FetchWork(listP, partP, table);
      if (!work.getPartDesc().isEmpty()) {
        PartitionDesc part0 = work.getPartDesc().get(0);
        PlanUtils.configureInputJobPropertiesForStorageHandler(part0.getTableDesc());
        work.setSplitSample(splitSample);
      }
      return work;
    }
View Full Code Here
107
108
109
110
111
112
113
114
115
116
117
118
119
120
    boolean aggressive = "more".equals(mode);
    FetchData fetch = checkTree(aggressive, pctx, alias, source);
    if (fetch != null && checkThreshold(fetch, pctx)) {
      int limit = pctx.getQB().getParseInfo().getOuterQueryLimit();
      FetchWork fetchWork = fetch.convertToWork();
      FetchTask fetchTask = (FetchTask) TaskFactory.get(fetchWork, pctx.getConf());
      fetchWork.setSink(fetch.completed(pctx, fetchWork));
      fetchWork.setSource(source);
      fetchWork.setLimit(limit);
      return fetchTask;
    }
    return null;
  }
View Full Code Here
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
    prop.setProperty(serdeConstants.SERIALIZATION_NULL_FORMAT, " ");
    String[] colTypes = schema.split("#");
    prop.setProperty("columns", colTypes[0]);
    prop.setProperty("columns.types", colTypes[1]);
    prop.setProperty(serdeConstants.SERIALIZATION_LIB, LazySimpleSerDe.class.getName());
    FetchWork fetch = new FetchWork(ctx.getResFile(), new TableDesc(
        TextInputFormat.class,IgnoreKeyTextOutputFormat.class, prop), -1);
    fetch.setSerializationNullFormat(" ");
    return (FetchTask) TaskFactory.get(fetch, conf);
  }
View Full Code Here
183
184
185
186
187
188
189
190
191
192
193
    // multiple files in the small table given a file in the big table). The remaining
    // tree will be processed while processing the join.
    // Look at comments in DummyStoreOperator for additional explanation.
    for (Map.Entry<String, FetchWork> entry : aliasToFetchWork.entrySet()) {
      String alias = entry.getKey();
      FetchWork fetchWork = entry.getValue();

      JobConf jobClone = new JobConf(hconf);

      TableScanOperator ts = (TableScanOperator)aliasToWork.get(alias);
      // push down projections
View Full Code Here
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
    } else if (mWork.getSamplingType() == MapWork.SAMPLING_ON_START) {
      console.printInfo("Creating sampling data..");
      assert topOp instanceof TableScanOperator;
      TableScanOperator ts = (TableScanOperator) topOp;

      FetchWork fetchWork;
      if (!partDesc.isPartitioned()) {
        assert paths.size() == 1;
        fetchWork = new FetchWork(inputPaths.get(0), partDesc.getTableDesc());
      } else {
        fetchWork = new FetchWork(inputPaths, parts, partDesc.getTableDesc());
      }
      fetchWork.setSource(ts);

      // random sampling
      FetchOperator fetcher = PartitionKeySampler.createSampler(fetchWork, conf, job, ts);
      try {
        ts.initialize(conf, new ObjectInspector[]{fetcher.getOutputObjectInspector()});
View Full Code Here
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
    prop.setProperty(serdeConstants.SERIALIZATION_NULL_FORMAT, " ");
    String[] colTypes = schema.split("#");
    prop.setProperty("columns", colTypes[0]);
    prop.setProperty("columns.types", colTypes[1]);
    prop.setProperty(serdeConstants.SERIALIZATION_LIB, LazySimpleSerDe.class.getName());
    FetchWork fetch = new FetchWork(ctx.getResFile(), new TableDesc(
        TextInputFormat.class,IgnoreKeyTextOutputFormat.class, prop), -1);
    fetch.setSerializationNullFormat(" ");
    return (FetchTask) TaskFactory.get(fetch, conf);
  }
View Full Code Here
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
    } else if (mWork.getSamplingType() == MapWork.SAMPLING_ON_START) {
      console.printInfo("Creating sampling data..");
      assert topOp instanceof TableScanOperator;
      TableScanOperator ts = (TableScanOperator) topOp;

      FetchWork fetchWork;
      if (!partDesc.isPartitioned()) {
        assert paths.size() == 1;
        fetchWork = new FetchWork(paths.get(0), partDesc.getTableDesc());
      } else {
        fetchWork = new FetchWork(paths, parts, partDesc.getTableDesc());
      }
      fetchWork.setSource(ts);

      // random sampling
      FetchOperator fetcher = PartitionKeySampler.createSampler(fetchWork, conf, job, ts);
      try {
        ts.initialize(conf, new ObjectInspector[]{fetcher.getOutputObjectInspector()});
View Full Code Here
TOP

Related Classes of org.apache.hadoop.hive.ql.plan.FetchWork

  • com.sap.hadoop.windowing.runtime2.QueryOutputPrinter
  • org.apache.hadoop.fs.Path
  • org.apache.hadoop.hive.ql.exec.ColumnStatsTask
  • org.apache.hadoop.hive.ql.exec.ListSinkOperator
  • org.apache.hadoop.hive.ql.exec.mr.ExecDriver
  • org.apache.hadoop.hive.ql.exec.SMBMapJoinOperator
  • org.apache.hadoop.hive.ql.optimizer.GenMapRedUtils
  • org.apache.hadoop.hive.ql.optimizer.MapJoinProcessor
  • org.apache.hadoop.hive.ql.optimizer.physical.GenMRSkewJoinProcessor
  • org.apache.hadoop.hive.ql.optimizer.physical.SortMergeJoinTaskDispatcher
    • 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.