use of org.apache.hadoop.hive.ql.plan.ColumnStatsDesc in project hive by apache.
the class TaskCompiler method genColumnStatsTask.
/**
* A helper function to generate a column stats task on top of map-red task. The column stats
* task fetches from the output of the map-red task, constructs the column stats object and
* persists it to the metastore.
*
* This method generates a plan with a column stats task on top of map-red task and sets up the
* appropriate metadata to be used during execution.
*
* @param analyzeRewrite
* @param loadTableWork
* @param loadFileWork
* @param rootTasks
* @param outerQueryLimit
*/
@SuppressWarnings("unchecked")
protected void genColumnStatsTask(AnalyzeRewriteContext analyzeRewrite, List<LoadFileDesc> loadFileWork, Set<Task<? extends Serializable>> leafTasks, int outerQueryLimit, int numBitVector) {
ColumnStatsTask cStatsTask = null;
ColumnStatsWork cStatsWork = null;
FetchWork fetch = null;
String tableName = analyzeRewrite.getTableName();
List<String> colName = analyzeRewrite.getColName();
List<String> colType = analyzeRewrite.getColType();
boolean isTblLevel = analyzeRewrite.isTblLvl();
String cols = loadFileWork.get(0).getColumns();
String colTypes = loadFileWork.get(0).getColumnTypes();
String resFileFormat;
TableDesc resultTab;
if (SessionState.get().isHiveServerQuery() && conf.getBoolVar(HiveConf.ConfVars.HIVE_SERVER2_THRIFT_RESULTSET_SERIALIZE_IN_TASKS)) {
resFileFormat = "SequenceFile";
resultTab = PlanUtils.getDefaultQueryOutputTableDesc(cols, colTypes, resFileFormat, ThriftJDBCBinarySerDe.class);
} else {
resFileFormat = HiveConf.getVar(conf, HiveConf.ConfVars.HIVEQUERYRESULTFILEFORMAT);
resultTab = PlanUtils.getDefaultQueryOutputTableDesc(cols, colTypes, resFileFormat, LazySimpleSerDe.class);
}
fetch = new FetchWork(loadFileWork.get(0).getSourcePath(), resultTab, outerQueryLimit);
ColumnStatsDesc cStatsDesc = new ColumnStatsDesc(tableName, colName, colType, isTblLevel, numBitVector);
cStatsWork = new ColumnStatsWork(fetch, cStatsDesc);
cStatsTask = (ColumnStatsTask) TaskFactory.get(cStatsWork, conf);
for (Task<? extends Serializable> tsk : leafTasks) {
tsk.addDependentTask(cStatsTask);
}
}
use of org.apache.hadoop.hive.ql.plan.ColumnStatsDesc in project hive by apache.
the class TaskCompiler method genColumnStatsTask.
/**
* A helper function to generate a column stats task on top of map-red task. The column stats
* task fetches from the output of the map-red task, constructs the column stats object and
* persists it to the metastore.
*
* This method generates a plan with a column stats task on top of map-red task and sets up the
* appropriate metadata to be used during execution.
*/
protected void genColumnStatsTask(AnalyzeRewriteContext analyzeRewrite, List<LoadFileDesc> loadFileWork, Map<String, StatsTask> map, int outerQueryLimit, int numBitVector) throws SemanticException {
FetchWork fetch;
String tableName = analyzeRewrite.getTableName();
List<String> colName = analyzeRewrite.getColName();
List<String> colType = analyzeRewrite.getColType();
boolean isTblLevel = analyzeRewrite.isTblLvl();
String cols = loadFileWork.get(0).getColumns();
String colTypes = loadFileWork.get(0).getColumnTypes();
TableDesc resultTab;
if (SessionState.get().isHiveServerQuery() && conf.getBoolVar(HiveConf.ConfVars.HIVE_SERVER2_THRIFT_RESULTSET_SERIALIZE_IN_TASKS)) {
resultTab = PlanUtils.getDefaultQueryOutputTableDesc(cols, colTypes, ResultFileFormat.SEQUENCEFILE.toString(), ThriftJDBCBinarySerDe.class);
} else {
resultTab = PlanUtils.getDefaultQueryOutputTableDesc(cols, colTypes, conf.getResultFileFormat().toString(), LazySimpleSerDe.class);
}
fetch = new FetchWork(loadFileWork.get(0).getSourcePath(), resultTab, outerQueryLimit);
ColumnStatsDesc cStatsDesc = new ColumnStatsDesc(tableName, colName, colType, isTblLevel, numBitVector, fetch);
StatsTask columnStatsTask = map.get(tableName);
if (columnStatsTask == null) {
throw new SemanticException("Can not find " + tableName + " in genColumnStatsTask");
} else {
columnStatsTask.getWork().setColStats(cStatsDesc);
}
}
Aggregations