You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hive.apache.org by na...@apache.org on 2012/07/31 07:34:04 UTC

svn commit: r1367411 - in /hive/trunk/ql/src: java/org/apache/hadoop/hive/ql/exec/ java/org/apache/hadoop/hive/ql/udf/generic/ test/queries/clientpositive/ test/results/clientpositive/

Author: namit
Date: Tue Jul 31 05:34:03 2012
New Revision: 1367411

URL: http://svn.apache.org/viewvc?rev=1367411&view=rev
Log:
HIVE-3238 A table generating, table generating function
(Edward via namit)


Added:
    hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDTFInline.java
    hive/trunk/ql/src/test/queries/clientpositive/udf_inline.q
    hive/trunk/ql/src/test/results/clientpositive/udf_inline.q.out
Modified:
    hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java
    hive/trunk/ql/src/test/results/clientpositive/show_functions.q.out

Modified: hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java
URL: http://svn.apache.org/viewvc/hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java?rev=1367411&r1=1367410&r2=1367411&view=diff
==============================================================================
--- hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java (original)
+++ hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java Tue Jul 31 05:34:03 2012
@@ -204,6 +204,7 @@ import org.apache.hadoop.hive.ql.udf.gen
 import org.apache.hadoop.hive.ql.udf.generic.GenericUDFWhen;
 import org.apache.hadoop.hive.ql.udf.generic.GenericUDTF;
 import org.apache.hadoop.hive.ql.udf.generic.GenericUDTFExplode;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDTFInline;
 import org.apache.hadoop.hive.ql.udf.generic.GenericUDTFJSONTuple;
 import org.apache.hadoop.hive.ql.udf.generic.GenericUDTFParseUrlTuple;
 import org.apache.hadoop.hive.ql.udf.generic.GenericUDTFStack;
@@ -467,6 +468,7 @@ public final class FunctionRegistry {
 
     // Generic UDTF's
     registerGenericUDTF("explode", GenericUDTFExplode.class);
+    registerGenericUDTF("inline", GenericUDTFInline.class);
     registerGenericUDTF("json_tuple", GenericUDTFJSONTuple.class);
     registerGenericUDTF("parse_url_tuple", GenericUDTFParseUrlTuple.class);
     registerGenericUDTF("stack", GenericUDTFStack.class);

Added: hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDTFInline.java
URL: http://svn.apache.org/viewvc/hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDTFInline.java?rev=1367411&view=auto
==============================================================================
--- hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDTFInline.java (added)
+++ hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDTFInline.java Tue Jul 31 05:34:03 2012
@@ -0,0 +1,76 @@
+package org.apache.hadoop.hive.ql.udf.generic;
+
+import java.util.List;
+import org.apache.hadoop.hive.ql.exec.Description;
+import org.apache.hadoop.hive.ql.exec.UDFArgumentException;
+import org.apache.hadoop.hive.ql.metadata.HiveException;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFUtils;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDTF;
+import org.apache.hadoop.hive.serde2.objectinspector.ListObjectInspector;
+import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector;
+import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector.Category;
+import org.apache.hadoop.hive.serde2.objectinspector.StructField;
+import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector;
+
+@Description(name ="inline", value= "_FUNC_( ARRAY( STRUCT()[,STRUCT()] "
++ "- explodes and array and struct into a table")
+public class GenericUDTFInline extends GenericUDTF {
+
+  private Object[] forwardObj;
+  private ListObjectInspector li;
+  private StructObjectInspector daStruct;
+
+  public GenericUDTFInline(){
+
+  }
+
+  @Override
+  public StructObjectInspector initialize(ObjectInspector[] ois) throws UDFArgumentException {
+    //There should be one argument that is a array of struct
+    if (ois.length!=1){
+      throw new UDFArgumentException("UDF tables only one argument");
+    }
+    if (ois[0].getCategory()!= Category.LIST){
+      throw new UDFArgumentException("Top level object must be an array but "
+              + "was "+ois[0].getTypeName());
+    }
+    li = (ListObjectInspector) ois[0];
+    ObjectInspector sub=li.getListElementObjectInspector();
+    if (sub.getCategory() != Category.STRUCT){
+      throw new UDFArgumentException("The sub element must be struct, but was "+sub.getTypeName());
+    }
+    daStruct = (StructObjectInspector) sub;
+    forwardObj = new Object[daStruct.getAllStructFieldRefs().size()];
+    return daStruct;
+  }
+
+  @Override
+  public void process(Object[] os) throws HiveException {
+    //list is always one item
+    List l = li.getList(os);
+    List<? extends StructField> fields = this.daStruct.getAllStructFieldRefs();
+    for (Object linner: l ){
+      List<List> innerList = (List) linner;
+      for (List rowList : innerList){
+        int i=0;
+        for (StructField f: fields){
+          GenericUDFUtils.ReturnObjectInspectorResolver res
+            = new GenericUDFUtils.ReturnObjectInspectorResolver();
+          res.update(f.getFieldObjectInspector());
+          this.forwardObj[i]=res.convertIfNecessary(rowList.get(i), f.getFieldObjectInspector());
+          i++;
+        }
+        forward(this.forwardObj);
+      }
+    }
+  }
+
+  @Override
+  public void close() throws HiveException {
+  }
+
+  @Override
+  public String toString() {
+    return "inline";
+  }
+}

Added: hive/trunk/ql/src/test/queries/clientpositive/udf_inline.q
URL: http://svn.apache.org/viewvc/hive/trunk/ql/src/test/queries/clientpositive/udf_inline.q?rev=1367411&view=auto
==============================================================================
--- hive/trunk/ql/src/test/queries/clientpositive/udf_inline.q (added)
+++ hive/trunk/ql/src/test/queries/clientpositive/udf_inline.q Tue Jul 31 05:34:03 2012
@@ -0,0 +1,18 @@
+describe function inline;
+
+explain SELECT inline( 
+  ARRAY(
+    STRUCT (1,'dude!'),
+    STRUCT (2,'Wheres'),
+    STRUCT (3,'my car?')
+  )
+)  as (id, text) FROM SRC limit 2;
+
+SELECT inline( 
+  ARRAY(
+    STRUCT (1,'dude!'),
+    STRUCT (2,'Wheres'),
+    STRUCT (3,'my car?')
+  )
+)  as (id, text) FROM SRC limit 2;
+

Modified: hive/trunk/ql/src/test/results/clientpositive/show_functions.q.out
URL: http://svn.apache.org/viewvc/hive/trunk/ql/src/test/results/clientpositive/show_functions.q.out?rev=1367411&r1=1367410&r2=1367411&view=diff
==============================================================================
--- hive/trunk/ql/src/test/results/clientpositive/show_functions.q.out (original)
+++ hive/trunk/ql/src/test/results/clientpositive/show_functions.q.out Tue Jul 31 05:34:03 2012
@@ -76,6 +76,7 @@ if
 in
 in_file
 index
+inline
 instr
 isnotnull
 isnull
@@ -202,6 +203,7 @@ e
 explode
 from_unixtime
 in_file
+inline
 json_tuple
 lcase
 like

Added: hive/trunk/ql/src/test/results/clientpositive/udf_inline.q.out
URL: http://svn.apache.org/viewvc/hive/trunk/ql/src/test/results/clientpositive/udf_inline.q.out?rev=1367411&view=auto
==============================================================================
--- hive/trunk/ql/src/test/results/clientpositive/udf_inline.q.out (added)
+++ hive/trunk/ql/src/test/results/clientpositive/udf_inline.q.out Tue Jul 31 05:34:03 2012
@@ -0,0 +1,77 @@
+PREHOOK: query: describe function inline
+PREHOOK: type: DESCFUNCTION
+POSTHOOK: query: describe function inline
+POSTHOOK: type: DESCFUNCTION
+inline( ARRAY( STRUCT()[,STRUCT()] - explodes and array and struct into a table
+PREHOOK: query: explain SELECT inline( 
+  ARRAY(
+    STRUCT (1,'dude!'),
+    STRUCT (2,'Wheres'),
+    STRUCT (3,'my car?')
+  )
+)  as (id, text) FROM SRC limit 2
+PREHOOK: type: QUERY
+POSTHOOK: query: explain SELECT inline( 
+  ARRAY(
+    STRUCT (1,'dude!'),
+    STRUCT (2,'Wheres'),
+    STRUCT (3,'my car?')
+  )
+)  as (id, text) FROM SRC limit 2
+POSTHOOK: type: QUERY
+ABSTRACT SYNTAX TREE:
+  (TOK_QUERY (TOK_FROM (TOK_TABREF (TOK_TABNAME SRC))) (TOK_INSERT (TOK_DESTINATION (TOK_DIR TOK_TMP_FILE)) (TOK_SELECT (TOK_SELEXPR (TOK_FUNCTION inline (TOK_FUNCTION ARRAY (TOK_FUNCTION STRUCT 1 'dude!') (TOK_FUNCTION STRUCT 2 'Wheres') (TOK_FUNCTION STRUCT 3 'my car?'))) id text)) (TOK_LIMIT 2)))
+
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 is a root stage
+
+STAGE PLANS:
+  Stage: Stage-1
+    Map Reduce
+      Alias -> Map Operator Tree:
+        src 
+          TableScan
+            alias: src
+            Select Operator
+              expressions:
+                    expr: array(struct(1,'dude!'),struct(2,'Wheres'),struct(3,'my car?'))
+                    type: array<struct<col1:int,col2:string>>
+              outputColumnNames: _col0
+              UDTF Operator
+                function name: inline
+                Limit
+                  File Output Operator
+                    compressed: false
+                    GlobalTableId: 0
+                    table:
+                        input format: org.apache.hadoop.mapred.TextInputFormat
+                        output format: org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: 2
+
+
+PREHOOK: query: SELECT inline( 
+  ARRAY(
+    STRUCT (1,'dude!'),
+    STRUCT (2,'Wheres'),
+    STRUCT (3,'my car?')
+  )
+)  as (id, text) FROM SRC limit 2
+PREHOOK: type: QUERY
+PREHOOK: Input: default@src
+#### A masked pattern was here ####
+POSTHOOK: query: SELECT inline( 
+  ARRAY(
+    STRUCT (1,'dude!'),
+    STRUCT (2,'Wheres'),
+    STRUCT (3,'my car?')
+  )
+)  as (id, text) FROM SRC limit 2
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@src
+#### A masked pattern was here ####
+1	dude!
+2	Wheres