You are viewing a plain text version of this content. The canonical link for it is here.
Posted to gitbox@hive.apache.org by GitBox <gi...@apache.org> on 2020/11/03 15:14:59 UTC
[GitHub] [hive] abstractdog opened a new pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
abstractdog opened a new pull request #1649:
URL: https://github.com/apache/hive/pull/1649
<!--
Thanks for sending a pull request! Here are some tips for you:
1. If this is your first time, please read our contributor guidelines: https://cwiki.apache.org/confluence/display/Hive/HowToContribute
2. Ensure that you have created an issue on the Hive project JIRA: https://issues.apache.org/jira/projects/HIVE/summary
3. Ensure you have added or run the appropriate tests for your PR:
4. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP]HIVE-XXXXX: Your PR title ...'.
5. Be sure to keep the PR description updated to reflect all changes.
6. Please write your PR title to summarize what this PR proposes.
7. If possible, provide a concise example to reproduce the issue for a faster review.
-->
### What changes were proposed in this pull request?
<!--
Please clarify what changes you are proposing. The purpose of this section is to outline the changes and how this PR fixes the issue.
If possible, please consider writing useful notes for better and faster reviews in your PR. See the examples below.
1. If you refactor some codes with changing classes, showing the class hierarchy will help reviewers.
2. If you fix some SQL features, you can provide some references of other DBMSes.
3. If there is design documentation, please add the link.
4. If there is a discussion in the mailing list, please add the link.
-->
### Why are the changes needed?
<!--
Please clarify why the changes are needed. For instance,
1. If you propose a new API, clarify the use case for a new API.
2. If you fix a bug, you can clarify why it is a bug.
-->
### Does this PR introduce _any_ user-facing change?
<!--
Note that it means *any* user-facing change including all aspects such as the documentation fix.
If yes, please clarify the previous behavior and the change this PR proposes - provide the console output, description, screenshot and/or a reproducable example to show the behavior difference if possible.
If possible, please also clarify if this is a user-facing change compared to the released Hive versions or within the unreleased branches such as master.
If no, write 'No'.
-->
### How was this patch tested?
<!--
If tests were added, say they were added here. Please make sure to add some test cases that check the changes thoroughly including negative and positive cases if possible.
If it was tested in a way different from regular unit tests, please clarify how you tested step by step, ideally copy and paste-able, so that other reviewers can test and check, and descendants can verify in the future.
If tests were not added, please describe why they were not added and/or why it was difficult to add.
-->
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] abstractdog commented on pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
abstractdog commented on pull request #1649:
URL: https://github.com/apache/hive/pull/1649#issuecomment-742396683
merged to master, thanks @pgaref for the review!
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] pgaref commented on a change in pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
pgaref commented on a change in pull request #1649:
URL: https://github.com/apache/hive/pull/1649#discussion_r528797991
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/exec/vector/ptf/VectorPTFEvaluatorCountDistinct.java
##########
@@ -0,0 +1,112 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.hadoop.hive.ql.exec.vector.ptf;
+
+import java.util.HashSet;
+import java.util.Set;
+
+import org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.ColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.VectorizedRowBatch;
+import org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression;
+import org.apache.hadoop.hive.ql.metadata.HiveException;
+import org.apache.hadoop.hive.ql.plan.ptf.WindowFrameDef;
+import org.apache.hive.common.util.Murmur3;
+
+import com.google.common.base.Preconditions;
+
+/**
+ * This class evaluates count(column) for a PTF group where a distinct keyword is applied to the
+ * partitioning column itself, e.g.:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt1) over(partition by txt1) as n,
+ * count(distinct txt2) over(partition by txt2) as m
+ * FROM example;
+ *
+ * In this case, the framework is still supposed to ensure sorting
+ * on the key (let's say txt1 for the first Reducer stage), but the original
+ * VectorPTFEvaluatorCount is not aware that a distinct keyword was applied
+ * to the key column. This case would be simple, because such function should
+ * return 1 every time. However, that's just a corner-case, a real scenario is
+ * when the partitioning column is not the same. In such cases, a real count
+ * distinct implementation is needed:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt2) over(partition by txt1) as n,
+ * count(distinct txt1) over(partition by txt2) as m
+ * FROM example;
+ */
+public abstract class VectorPTFEvaluatorCountDistinct extends VectorPTFEvaluatorCount {
+
+ protected Set<Object> uniqueObjects;
+
+ public VectorPTFEvaluatorCountDistinct(WindowFrameDef windowFrameDef,
+ VectorExpression inputVecExpr, int outputColumnNum) {
+ super(windowFrameDef, inputVecExpr, outputColumnNum);
+ resetEvaluator();
+ }
+
+ @Override
+ public void evaluateGroupBatch(VectorizedRowBatch batch) throws HiveException {
+
+ evaluateInputExpr(batch);
+
+ // We do not filter when PTF is in reducer.
+ Preconditions.checkState(!batch.selectedInUse);
+
+ final int size = batch.size;
+ if (size == 0) {
+ return;
+ }
+ ColumnVector colVector = batch.cols[inputColumnNum];
+ if (colVector.isRepeating) {
+ if (colVector.noNulls || !colVector.isNull[0]) {
+ countValue(colVector, 0);
+ }
+ } else {
+ boolean[] batchIsNull = colVector.isNull;
+ for (int i = 0; i < size; i++) {
+ if (!batchIsNull[i]) {
+ countValue(colVector, i);
+ }
+ }
+ }
+ }
+
+ protected void countValue(ColumnVector colVector, int i) {
+ Object value = getValue(colVector, i);
+ if (!uniqueObjects.contains(value)) {
+ uniqueObjects.add(value);
Review comment:
We do check if the element is present twice here -- add method already checks if the element is part of the Set so we dont need to call contains. I would just add a comment here instead.
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/exec/vector/ptf/VectorPTFEvaluatorCountDistinct.java
##########
@@ -0,0 +1,112 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.hadoop.hive.ql.exec.vector.ptf;
+
+import java.util.HashSet;
+import java.util.Set;
+
+import org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.ColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.VectorizedRowBatch;
+import org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression;
+import org.apache.hadoop.hive.ql.metadata.HiveException;
+import org.apache.hadoop.hive.ql.plan.ptf.WindowFrameDef;
+import org.apache.hive.common.util.Murmur3;
+
+import com.google.common.base.Preconditions;
+
+/**
+ * This class evaluates count(column) for a PTF group where a distinct keyword is applied to the
+ * partitioning column itself, e.g.:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt1) over(partition by txt1) as n,
+ * count(distinct txt2) over(partition by txt2) as m
+ * FROM example;
+ *
+ * In this case, the framework is still supposed to ensure sorting
+ * on the key (let's say txt1 for the first Reducer stage), but the original
+ * VectorPTFEvaluatorCount is not aware that a distinct keyword was applied
+ * to the key column. This case would be simple, because such function should
+ * return 1 every time. However, that's just a corner-case, a real scenario is
+ * when the partitioning column is not the same. In such cases, a real count
+ * distinct implementation is needed:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt2) over(partition by txt1) as n,
+ * count(distinct txt1) over(partition by txt2) as m
+ * FROM example;
+ */
+public abstract class VectorPTFEvaluatorCountDistinct extends VectorPTFEvaluatorCount {
+
+ protected Set<Object> uniqueObjects;
+
+ public VectorPTFEvaluatorCountDistinct(WindowFrameDef windowFrameDef,
+ VectorExpression inputVecExpr, int outputColumnNum) {
+ super(windowFrameDef, inputVecExpr, outputColumnNum);
+ resetEvaluator();
+ }
+
+ @Override
+ public void evaluateGroupBatch(VectorizedRowBatch batch) throws HiveException {
+
+ evaluateInputExpr(batch);
+
+ // We do not filter when PTF is in reducer.
+ Preconditions.checkState(!batch.selectedInUse);
+
+ final int size = batch.size;
+ if (size == 0) {
+ return;
+ }
+ ColumnVector colVector = batch.cols[inputColumnNum];
+ if (colVector.isRepeating) {
+ if (colVector.noNulls || !colVector.isNull[0]) {
+ countValue(colVector, 0);
+ }
+ } else {
+ boolean[] batchIsNull = colVector.isNull;
+ for (int i = 0; i < size; i++) {
+ if (!batchIsNull[i]) {
+ countValue(colVector, i);
Review comment:
Instead of updating the inherited count variable we could override getLongGroupResult() with uniqueObjects.size
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] abstractdog commented on a change in pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
abstractdog commented on a change in pull request #1649:
URL: https://github.com/apache/hive/pull/1649#discussion_r530182438
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/exec/vector/ptf/VectorPTFEvaluatorBytesCountDistinct.java
##########
@@ -0,0 +1,42 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.hadoop.hive.ql.exec.vector.ptf;
+
+import org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.ColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression;
+import org.apache.hadoop.hive.ql.plan.ptf.WindowFrameDef;
+import org.apache.hive.common.util.Murmur3;
+
+/**
+ * Bytes (String) implementation for VectorPTFEvaluatorCountDistinct.
+ */
+public class VectorPTFEvaluatorBytesCountDistinct extends VectorPTFEvaluatorCountDistinct {
+
+ public VectorPTFEvaluatorBytesCountDistinct(WindowFrameDef windowFrameDef,
+ VectorExpression inputVecExpr, int outputColumnNum) {
+ super(windowFrameDef, inputVecExpr, outputColumnNum);
+ resetEvaluator();
+ }
+
+ protected Object getValue(ColumnVector colVector, int i) {
+ BytesColumnVector inV = (BytesColumnVector) colVector;
+ return Murmur3.hash32(inV.vector[i], inV.start[i], inV.length[i], Murmur3.DEFAULT_SEED);
Review comment:
I used hashing for memory considerations, I was hoping that this way the unique set will consume less memory (storing hashes instead of strings of arbitrary length)...now I think that we could be fine with storing strings and prevent additional hashing, as we tend to optimize CPU cycles instead of memory in the very-first round...I'll change this to new String(byte[])
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] pgaref commented on pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
pgaref commented on pull request #1649:
URL: https://github.com/apache/hive/pull/1649#issuecomment-733617093
Thanks for the update @abstractdog !
+1 tests pending
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] abstractdog merged pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
abstractdog merged pull request #1649:
URL: https://github.com/apache/hive/pull/1649
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] pgaref commented on a change in pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
pgaref commented on a change in pull request #1649:
URL: https://github.com/apache/hive/pull/1649#discussion_r530261010
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDAFCount.java
##########
@@ -180,14 +181,13 @@ public void iterate(AggregationBuffer agg, Object[] parameters)
if (((CountAgg) agg).uniqueObjects == null) {
((CountAgg) agg).uniqueObjects = new HashSet<ObjectInspectorObject>();
}
- HashSet<ObjectInspectorObject> uniqueObjs = ((CountAgg) agg).uniqueObjects;
+ Set<ObjectInspectorObject> uniqueObjs = ((CountAgg) agg).uniqueObjects;
ObjectInspectorObject obj = new ObjectInspectorObject(
ObjectInspectorUtils.copyToStandardObject(parameters, inputOI, ObjectInspectorCopyOption.JAVA),
outputOI);
- if (!uniqueObjs.contains(obj)) {
- uniqueObjs.add(obj);
- } else {
+ boolean inserted = uniqueObjs.add(obj);
+ if (!inserted){
Review comment:
Thanks for taking care of this
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] pgaref commented on a change in pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
pgaref commented on a change in pull request #1649:
URL: https://github.com/apache/hive/pull/1649#discussion_r528820635
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/exec/vector/ptf/VectorPTFEvaluatorBytesCountDistinct.java
##########
@@ -0,0 +1,42 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.hadoop.hive.ql.exec.vector.ptf;
+
+import org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.ColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression;
+import org.apache.hadoop.hive.ql.plan.ptf.WindowFrameDef;
+import org.apache.hive.common.util.Murmur3;
+
+/**
+ * Bytes (String) implementation for VectorPTFEvaluatorCountDistinct.
+ */
+public class VectorPTFEvaluatorBytesCountDistinct extends VectorPTFEvaluatorCountDistinct {
+
+ public VectorPTFEvaluatorBytesCountDistinct(WindowFrameDef windowFrameDef,
+ VectorExpression inputVecExpr, int outputColumnNum) {
+ super(windowFrameDef, inputVecExpr, outputColumnNum);
+ resetEvaluator();
+ }
+
+ protected Object getValue(ColumnVector colVector, int i) {
+ BytesColumnVector inV = (BytesColumnVector) colVector;
+ return Murmur3.hash32(inV.vector[i], inV.start[i], inV.length[i], Murmur3.DEFAULT_SEED);
Review comment:
What is the benefit of returning Murmur hash directly here? Adding a String to the HashSet will also hash the input (so essentially we are hashing twice)
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] pgaref commented on a change in pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
pgaref commented on a change in pull request #1649:
URL: https://github.com/apache/hive/pull/1649#discussion_r530261010
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDAFCount.java
##########
@@ -180,14 +181,13 @@ public void iterate(AggregationBuffer agg, Object[] parameters)
if (((CountAgg) agg).uniqueObjects == null) {
((CountAgg) agg).uniqueObjects = new HashSet<ObjectInspectorObject>();
}
- HashSet<ObjectInspectorObject> uniqueObjs = ((CountAgg) agg).uniqueObjects;
+ Set<ObjectInspectorObject> uniqueObjs = ((CountAgg) agg).uniqueObjects;
ObjectInspectorObject obj = new ObjectInspectorObject(
ObjectInspectorUtils.copyToStandardObject(parameters, inputOI, ObjectInspectorCopyOption.JAVA),
outputOI);
- if (!uniqueObjs.contains(obj)) {
- uniqueObjs.add(obj);
- } else {
+ boolean inserted = uniqueObjs.add(obj);
+ if (!inserted){
Review comment:
Thanks for fixing this
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] abstractdog commented on a change in pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
abstractdog commented on a change in pull request #1649:
URL: https://github.com/apache/hive/pull/1649#discussion_r530185546
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/exec/vector/ptf/VectorPTFEvaluatorCountDistinct.java
##########
@@ -0,0 +1,112 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.hadoop.hive.ql.exec.vector.ptf;
+
+import java.util.HashSet;
+import java.util.Set;
+
+import org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.ColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.VectorizedRowBatch;
+import org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression;
+import org.apache.hadoop.hive.ql.metadata.HiveException;
+import org.apache.hadoop.hive.ql.plan.ptf.WindowFrameDef;
+import org.apache.hive.common.util.Murmur3;
+
+import com.google.common.base.Preconditions;
+
+/**
+ * This class evaluates count(column) for a PTF group where a distinct keyword is applied to the
+ * partitioning column itself, e.g.:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt1) over(partition by txt1) as n,
+ * count(distinct txt2) over(partition by txt2) as m
+ * FROM example;
+ *
+ * In this case, the framework is still supposed to ensure sorting
+ * on the key (let's say txt1 for the first Reducer stage), but the original
+ * VectorPTFEvaluatorCount is not aware that a distinct keyword was applied
+ * to the key column. This case would be simple, because such function should
+ * return 1 every time. However, that's just a corner-case, a real scenario is
+ * when the partitioning column is not the same. In such cases, a real count
+ * distinct implementation is needed:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt2) over(partition by txt1) as n,
+ * count(distinct txt1) over(partition by txt2) as m
+ * FROM example;
+ */
+public abstract class VectorPTFEvaluatorCountDistinct extends VectorPTFEvaluatorCount {
+
+ protected Set<Object> uniqueObjects;
+
+ public VectorPTFEvaluatorCountDistinct(WindowFrameDef windowFrameDef,
+ VectorExpression inputVecExpr, int outputColumnNum) {
+ super(windowFrameDef, inputVecExpr, outputColumnNum);
+ resetEvaluator();
+ }
+
+ @Override
+ public void evaluateGroupBatch(VectorizedRowBatch batch) throws HiveException {
+
+ evaluateInputExpr(batch);
+
+ // We do not filter when PTF is in reducer.
+ Preconditions.checkState(!batch.selectedInUse);
+
+ final int size = batch.size;
+ if (size == 0) {
+ return;
+ }
+ ColumnVector colVector = batch.cols[inputColumnNum];
+ if (colVector.isRepeating) {
+ if (colVector.noNulls || !colVector.isNull[0]) {
+ countValue(colVector, 0);
+ }
+ } else {
+ boolean[] batchIsNull = colVector.isNull;
+ for (int i = 0; i < size; i++) {
+ if (!batchIsNull[i]) {
+ countValue(colVector, i);
+ }
+ }
+ }
+ }
+
+ protected void countValue(ColumnVector colVector, int i) {
+ Object value = getValue(colVector, i);
+ if (!uniqueObjects.contains(value)) {
+ uniqueObjects.add(value);
Review comment:
btw, I copied this wrong approach from GenericUDAFCount, I'm fixing it there also
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] abstractdog commented on a change in pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
abstractdog commented on a change in pull request #1649:
URL: https://github.com/apache/hive/pull/1649#discussion_r530179906
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/exec/vector/ptf/VectorPTFEvaluatorCountDistinct.java
##########
@@ -0,0 +1,112 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.hadoop.hive.ql.exec.vector.ptf;
+
+import java.util.HashSet;
+import java.util.Set;
+
+import org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.ColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.VectorizedRowBatch;
+import org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression;
+import org.apache.hadoop.hive.ql.metadata.HiveException;
+import org.apache.hadoop.hive.ql.plan.ptf.WindowFrameDef;
+import org.apache.hive.common.util.Murmur3;
+
+import com.google.common.base.Preconditions;
+
+/**
+ * This class evaluates count(column) for a PTF group where a distinct keyword is applied to the
+ * partitioning column itself, e.g.:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt1) over(partition by txt1) as n,
+ * count(distinct txt2) over(partition by txt2) as m
+ * FROM example;
+ *
+ * In this case, the framework is still supposed to ensure sorting
+ * on the key (let's say txt1 for the first Reducer stage), but the original
+ * VectorPTFEvaluatorCount is not aware that a distinct keyword was applied
+ * to the key column. This case would be simple, because such function should
+ * return 1 every time. However, that's just a corner-case, a real scenario is
+ * when the partitioning column is not the same. In such cases, a real count
+ * distinct implementation is needed:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt2) over(partition by txt1) as n,
+ * count(distinct txt1) over(partition by txt2) as m
+ * FROM example;
+ */
+public abstract class VectorPTFEvaluatorCountDistinct extends VectorPTFEvaluatorCount {
+
+ protected Set<Object> uniqueObjects;
+
+ public VectorPTFEvaluatorCountDistinct(WindowFrameDef windowFrameDef,
+ VectorExpression inputVecExpr, int outputColumnNum) {
+ super(windowFrameDef, inputVecExpr, outputColumnNum);
+ resetEvaluator();
+ }
+
+ @Override
+ public void evaluateGroupBatch(VectorizedRowBatch batch) throws HiveException {
+
+ evaluateInputExpr(batch);
+
+ // We do not filter when PTF is in reducer.
+ Preconditions.checkState(!batch.selectedInUse);
+
+ final int size = batch.size;
+ if (size == 0) {
+ return;
+ }
+ ColumnVector colVector = batch.cols[inputColumnNum];
+ if (colVector.isRepeating) {
+ if (colVector.noNulls || !colVector.isNull[0]) {
+ countValue(colVector, 0);
+ }
+ } else {
+ boolean[] batchIsNull = colVector.isNull;
+ for (int i = 0; i < size; i++) {
+ if (!batchIsNull[i]) {
+ countValue(colVector, i);
Review comment:
right! updating it
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org
[GitHub] [hive] abstractdog commented on a change in pull request #1649: HIVE-24245: Vectorized PTF with count and distinct over partition producing incorrect results.
Posted by GitBox <gi...@apache.org>.
abstractdog commented on a change in pull request #1649:
URL: https://github.com/apache/hive/pull/1649#discussion_r530178604
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/exec/vector/ptf/VectorPTFEvaluatorCountDistinct.java
##########
@@ -0,0 +1,112 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.hadoop.hive.ql.exec.vector.ptf;
+
+import java.util.HashSet;
+import java.util.Set;
+
+import org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.ColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.VectorizedRowBatch;
+import org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression;
+import org.apache.hadoop.hive.ql.metadata.HiveException;
+import org.apache.hadoop.hive.ql.plan.ptf.WindowFrameDef;
+import org.apache.hive.common.util.Murmur3;
+
+import com.google.common.base.Preconditions;
+
+/**
+ * This class evaluates count(column) for a PTF group where a distinct keyword is applied to the
+ * partitioning column itself, e.g.:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt1) over(partition by txt1) as n,
+ * count(distinct txt2) over(partition by txt2) as m
+ * FROM example;
+ *
+ * In this case, the framework is still supposed to ensure sorting
+ * on the key (let's say txt1 for the first Reducer stage), but the original
+ * VectorPTFEvaluatorCount is not aware that a distinct keyword was applied
+ * to the key column. This case would be simple, because such function should
+ * return 1 every time. However, that's just a corner-case, a real scenario is
+ * when the partitioning column is not the same. In such cases, a real count
+ * distinct implementation is needed:
+ *
+ * SELECT
+ * txt1,
+ * txt2,
+ * count(distinct txt2) over(partition by txt1) as n,
+ * count(distinct txt1) over(partition by txt2) as m
+ * FROM example;
+ */
+public abstract class VectorPTFEvaluatorCountDistinct extends VectorPTFEvaluatorCount {
+
+ protected Set<Object> uniqueObjects;
+
+ public VectorPTFEvaluatorCountDistinct(WindowFrameDef windowFrameDef,
+ VectorExpression inputVecExpr, int outputColumnNum) {
+ super(windowFrameDef, inputVecExpr, outputColumnNum);
+ resetEvaluator();
+ }
+
+ @Override
+ public void evaluateGroupBatch(VectorizedRowBatch batch) throws HiveException {
+
+ evaluateInputExpr(batch);
+
+ // We do not filter when PTF is in reducer.
+ Preconditions.checkState(!batch.selectedInUse);
+
+ final int size = batch.size;
+ if (size == 0) {
+ return;
+ }
+ ColumnVector colVector = batch.cols[inputColumnNum];
+ if (colVector.isRepeating) {
+ if (colVector.noNulls || !colVector.isNull[0]) {
+ countValue(colVector, 0);
+ }
+ } else {
+ boolean[] batchIsNull = colVector.isNull;
+ for (int i = 0; i < size; i++) {
+ if (!batchIsNull[i]) {
+ countValue(colVector, i);
+ }
+ }
+ }
+ }
+
+ protected void countValue(ColumnVector colVector, int i) {
+ Object value = getValue(colVector, i);
+ if (!uniqueObjects.contains(value)) {
+ uniqueObjects.add(value);
Review comment:
yeah, thanks, forgot that Set takes care of uniqueness :)
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org
For additional commands, e-mail: gitbox-help@hive.apache.org