/**
* 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.udf.approx;
import java.util.ArrayList;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.hive.ql.exec.Description;
import org.apache.hadoop.hive.ql.exec.UDFArgumentTypeException;
import org.apache.hadoop.hive.ql.metadata.HiveException;
import org.apache.hadoop.hive.ql.parse.SemanticException;
import org.apache.hadoop.hive.ql.udf.generic.AbstractGenericUDAFResolver;
import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
import org.apache.hadoop.hive.serde2.io.DoubleWritable;
import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector;
import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspectorFactory;
import org.apache.hadoop.hive.serde2.objectinspector.PrimitiveObjectInspector;
import org.apache.hadoop.hive.serde2.objectinspector.StructField;
import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector;
import org.apache.hadoop.hive.serde2.objectinspector.primitive.DoubleObjectInspector;
import org.apache.hadoop.hive.serde2.objectinspector.primitive.LongObjectInspector;
import org.apache.hadoop.hive.serde2.objectinspector.primitive.PrimitiveObjectInspectorFactory;
import org.apache.hadoop.hive.serde2.objectinspector.primitive.PrimitiveObjectInspectorUtils;
import org.apache.hadoop.hive.serde2.typeinfo.PrimitiveTypeInfo;
import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
import org.apache.hadoop.io.BooleanWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.util.StringUtils;
/**
* GenericUDAFSum.
*
*/
@Description(name = "approx_sum", value = "_FUNC_(x) - Returns the approximate sum of a set of numbers with error bars")
public class ApproxUDAFSum extends AbstractGenericUDAFResolver {
static final Log LOG = LogFactory.getLog(ApproxUDAFSum.class.getName());
@Override
public GenericUDAFEvaluator getEvaluator(TypeInfo[] parameters)
throws SemanticException {
if (parameters.length != 3) {
throw new UDFArgumentTypeException(parameters.length - 1,
"Exactly one argument is expected.");
}
if (parameters[0].getCategory() != ObjectInspector.Category.PRIMITIVE) {
throw new UDFArgumentTypeException(0,
"Only primitive type arguments are accepted but "
+ parameters[0].getTypeName() + " is passed.");
}
switch (((PrimitiveTypeInfo) parameters[0]).getPrimitiveCategory()) {
case BYTE:
case SHORT:
case INT:
case LONG:
return new ApproxUDAFSumLong();
case FLOAT:
case DOUBLE:
return new ApproxUDAFSumDouble();
case DATE:
case TIMESTAMP:
case STRING:
case BOOLEAN:
default:
throw new UDFArgumentTypeException(0,
"Only numeric type arguments are accepted but "
+ parameters[0].getTypeName() + " is passed.");
}
}
/**
* ApproxUDAFSumDouble.
*
*/
public static class ApproxUDAFSumDouble extends GenericUDAFEvaluator {
// private PrimitiveObjectInspector inputOI;
// private DoubleWritable result;
// For PARTIAL1 and COMPLETE
private PrimitiveObjectInspector inputOI;
private PrimitiveObjectInspector totalRowsOI;
private PrimitiveObjectInspector sampleRowsOI;
// For PARTIAL2 and FINAL
private StructObjectInspector soi;
private StructField countField;
private StructField sumField;
private StructField varianceField;
private StructField totalRowsField;
private StructField sampleRowsField;
private LongObjectInspector countFieldOI;
private DoubleObjectInspector sumFieldOI;
private DoubleObjectInspector varianceFieldOI;
private LongObjectInspector totalRowsFieldOI;
private LongObjectInspector sampleRowsFieldOI;
// For PARTIAL1 and PARTIAL2
private Object[] partialResult;
// For FINAL and COMPLETE
Text result;
@Override
public ObjectInspector init(Mode m, ObjectInspector[] parameters) throws HiveException {
assert (parameters.length == 3);
super.init(m, parameters);
if (parameters.length == 3) {
totalRowsOI = (PrimitiveObjectInspector) parameters[2];
sampleRowsOI = (PrimitiveObjectInspector) parameters[1];
}
// result = new DoubleWritable(0);
// inputOI = (PrimitiveObjectInspector) parameters[0];
// return PrimitiveObjectInspectorFactory.writableDoubleObjectInspector;
// init input
if (mode == Mode.PARTIAL1 || mode == Mode.COMPLETE) {
inputOI = (PrimitiveObjectInspector) parameters[0];
} else {
soi = (StructObjectInspector) parameters[0];
countField = soi.getStructFieldRef("count");
sumField = soi.getStructFieldRef("sum");
varianceField = soi.getStructFieldRef("variance");
totalRowsField = soi.getStructFieldRef("totalRows");
sampleRowsField = soi.getStructFieldRef("sampleRows");
countFieldOI = (LongObjectInspector) countField
.getFieldObjectInspector();
sumFieldOI = (DoubleObjectInspector) sumField.getFieldObjectInspector();
varianceFieldOI = (DoubleObjectInspector) varianceField
.getFieldObjectInspector();
totalRowsFieldOI = (LongObjectInspector) totalRowsField
.getFieldObjectInspector();
sampleRowsFieldOI = (LongObjectInspector) sampleRowsField
.getFieldObjectInspector();
}
// init output
if (mode == Mode.PARTIAL1 || mode == Mode.PARTIAL2) {
// The output of a partial aggregation is a struct containing
// a long count and doubles sum and variance.
ArrayList<ObjectInspector> foi = new ArrayList<ObjectInspector>();
foi.add(PrimitiveObjectInspectorFactory.writableBooleanObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableLongObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableDoubleObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableDoubleObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableLongObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableLongObjectInspector);
ArrayList<String> fname = new ArrayList<String>();
fname.add("empty");
fname.add("count");
fname.add("sum");
fname.add("variance");
fname.add("totalRows");
fname.add("sampleRows");
partialResult = new Object[6];
partialResult[0] = new BooleanWritable();
partialResult[1] = new LongWritable(0);
partialResult[2] = new DoubleWritable(0);
partialResult[3] = new DoubleWritable(0);
partialResult[4] = new LongWritable(0);
partialResult[5] = new LongWritable(0);
return ObjectInspectorFactory.getStandardStructObjectInspector(fname,
foi);
} else {
result = new Text();
return PrimitiveObjectInspectorFactory.writableStringObjectInspector;
}
}
/** class for storing double sum value. */
static class SumDoubleAgg implements AggregationBuffer {
boolean empty;
long count; // number of elements
double sum;
double variance; // sum[x-avg^2] (this is actually n times the variance)
long totalRows;
long sampleRows;
}
@Override
public AggregationBuffer getNewAggregationBuffer() throws HiveException {
SumDoubleAgg result = new SumDoubleAgg();
reset(result);
return result;
}
@Override
public void reset(AggregationBuffer agg) throws HiveException {
SumDoubleAgg myagg = (SumDoubleAgg) agg;
myagg.empty = true;
myagg.count = 0;
myagg.sum = 0;
myagg.variance = 0;
myagg.totalRows = 0;
myagg.sampleRows = 0;
}
boolean warned = false;
@Override
public void iterate(AggregationBuffer agg, Object[] parameters) throws HiveException {
assert (parameters.length == 3);
if (parameters.length == 3) {
((SumDoubleAgg) agg).totalRows = PrimitiveObjectInspectorUtils.getLong(parameters[2],
totalRowsOI);
((SumDoubleAgg) agg).sampleRows = PrimitiveObjectInspectorUtils.getLong(parameters[1],
sampleRowsOI);
}
Object p = parameters[0];
if (p != null) {
SumDoubleAgg myagg = (SumDoubleAgg) agg;
try {
double v = PrimitiveObjectInspectorUtils.getDouble(p, inputOI);
myagg.count++;
myagg.sum += v;
if (myagg.count > 1) {
double t = myagg.count * v - myagg.sum;
myagg.variance += (t * t) / ((double) myagg.count * (myagg.count - 1));
}
} catch (NumberFormatException e) {
if (!warned) {
warned = true;
LOG.warn(getClass().getSimpleName() + " "
+ StringUtils.stringifyException(e));
LOG
.warn(getClass().getSimpleName()
+ " ignoring similar exceptions.");
}
}
}
}
@Override
public Object terminatePartial(AggregationBuffer agg) throws HiveException {
// return terminate(agg);
SumDoubleAgg myagg = (SumDoubleAgg) agg;
((BooleanWritable) partialResult[0]).set(myagg.empty);
((LongWritable) partialResult[1]).set(myagg.count);
((DoubleWritable) partialResult[2]).set(myagg.sum);
((DoubleWritable) partialResult[3]).set(myagg.variance);
((LongWritable) partialResult[4]).set(myagg.totalRows);
((LongWritable) partialResult[5]).set(myagg.sampleRows);
return partialResult;
}
@Override
public void merge(AggregationBuffer agg, Object partial) throws HiveException {
if (partial != null) {
SumDoubleAgg myagg = (SumDoubleAgg) agg;
myagg.empty = false;
Object partialCount = soi.getStructFieldData(partial, countField);
Object partialSum = soi.getStructFieldData(partial, sumField);
Object partialVariance = soi.getStructFieldData(partial, varianceField);
Object partialTotalRows = soi.getStructFieldData(partial, totalRowsField);
Object partialSampleRows = soi.getStructFieldData(partial, sampleRowsField);
long n = myagg.count;
long m = countFieldOI.get(partialCount);
long q = totalRowsFieldOI.get(partialTotalRows);
long r = sampleRowsFieldOI.get(partialSampleRows);
myagg.totalRows = q;
myagg.sampleRows = r;
if (n == 0) {
// Just copy the information since there is nothing so far
myagg.variance = varianceFieldOI.get(partialVariance);
myagg.count = countFieldOI.get(partialCount);
myagg.sum = sumFieldOI.get(partialSum);
}
if (m != 0 && n != 0) {
// Merge the two partials
double a = myagg.sum;
double b = sumFieldOI.get(partialSum);
myagg.empty = false;
myagg.count += m;
myagg.sum += b;
double t = (m / (double) n) * a - b;
myagg.variance += varianceFieldOI.get(partialVariance)
+ ((n / (double) m) / ((double) n + m)) * t * t;
}
}
}
@Override
public Object terminate(AggregationBuffer agg) throws HiveException {
SumDoubleAgg myagg = (SumDoubleAgg) agg;
if (myagg.empty) {
return null;
}
double approx_sum = ((double) myagg.sum * myagg.totalRows) / myagg.sampleRows;
double probability = ((double) myagg.count) / ((double) myagg.totalRows);
double mean = myagg.sum / myagg.count;
LOG.info("Sum: " + approx_sum);
LOG.info("TotalRows: " + myagg.totalRows);
LOG.info("Probability: " + probability);
LOG.info("Sampling Ratio: " + ((double) myagg.sampleRows) / myagg.totalRows);
StringBuilder sb = new StringBuilder();
sb.append(approx_sum);
sb.append(" +/- ");
sb.append(Math.ceil((2.575 * (1.0 * myagg.totalRows / myagg.sampleRows) * Math
.sqrt(probability
* ((myagg.variance) + ((1 - probability) * myagg.totalRows * mean * mean))))));
sb.append(" (99% Confidence) ");
result.set(sb.toString());
return result;
}
}
/**
* GenericUDAFSumLong.
*
*/
public static class ApproxUDAFSumLong extends GenericUDAFEvaluator {
// private PrimitiveObjectInspector inputOI;
// private LongWritable result;
// For PARTIAL1 and COMPLETE
private PrimitiveObjectInspector inputOI;
private PrimitiveObjectInspector totalRowsOI;
private PrimitiveObjectInspector sampleRowsOI;
// For PARTIAL2 and FINAL
private StructObjectInspector soi;
private StructField countField;
private StructField sumField;
private StructField varianceField;
private StructField totalRowsField;
private StructField sampleRowsField;
private LongObjectInspector countFieldOI;
private LongObjectInspector sumFieldOI;
private DoubleObjectInspector varianceFieldOI;
private LongObjectInspector totalRowsFieldOI;
private LongObjectInspector sampleRowsFieldOI;
// For PARTIAL1 and PARTIAL2
private Object[] partialResult;
// For FINAL and COMPLETE
Text result;
@Override
public ObjectInspector init(Mode m, ObjectInspector[] parameters) throws HiveException {
assert (parameters.length == 3);
super.init(m, parameters);
if (parameters.length == 3) {
totalRowsOI = (PrimitiveObjectInspector) parameters[2];
sampleRowsOI = (PrimitiveObjectInspector) parameters[1];
}
// init input
if (mode == Mode.PARTIAL1 || mode == Mode.COMPLETE) {
inputOI = (PrimitiveObjectInspector) parameters[0];
} else {
soi = (StructObjectInspector) parameters[0];
countField = soi.getStructFieldRef("count");
sumField = soi.getStructFieldRef("sum");
varianceField = soi.getStructFieldRef("variance");
totalRowsField = soi.getStructFieldRef("totalRows");
sampleRowsField = soi.getStructFieldRef("sampleRows");
countFieldOI = (LongObjectInspector) countField
.getFieldObjectInspector();
sumFieldOI = (LongObjectInspector) sumField.getFieldObjectInspector();
varianceFieldOI = (DoubleObjectInspector) varianceField
.getFieldObjectInspector();
totalRowsFieldOI = (LongObjectInspector) totalRowsField
.getFieldObjectInspector();
sampleRowsFieldOI = (LongObjectInspector) sampleRowsField
.getFieldObjectInspector();
}
// init output
if (mode == Mode.PARTIAL1 || mode == Mode.PARTIAL2) {
// The output of a partial aggregation is a struct containing
// a long count and doubles sum and variance.
ArrayList<ObjectInspector> foi = new ArrayList<ObjectInspector>();
foi.add(PrimitiveObjectInspectorFactory.writableBooleanObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableLongObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableLongObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableDoubleObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableLongObjectInspector);
foi.add(PrimitiveObjectInspectorFactory.writableLongObjectInspector);
ArrayList<String> fname = new ArrayList<String>();
fname.add("empty");
fname.add("count");
fname.add("sum");
fname.add("variance");
fname.add("totalRows");
fname.add("sampleRows");
partialResult = new Object[6];
partialResult[0] = new BooleanWritable();
partialResult[1] = new LongWritable(0);
partialResult[2] = new LongWritable(0);
partialResult[3] = new DoubleWritable(0);
partialResult[4] = new LongWritable(0);
partialResult[5] = new LongWritable(0);
return ObjectInspectorFactory.getStandardStructObjectInspector(fname,
foi);
} else {
result = new Text();
return PrimitiveObjectInspectorFactory.writableStringObjectInspector;
}
}
/** class for storing double sum value. */
static class SumLongAgg implements AggregationBuffer {
boolean empty;
long count; // number of elements
long sum;
double variance; // sum[x-avg^2] (this is actually n times the variance)
long totalRows;
long sampleRows;
}
@Override
public AggregationBuffer getNewAggregationBuffer() throws HiveException {
SumLongAgg result = new SumLongAgg();
reset(result);
return result;
}
@Override
public void reset(AggregationBuffer agg) throws HiveException {
SumLongAgg myagg = (SumLongAgg) agg;
myagg.empty = true;
myagg.count = 0;
myagg.sum = 0;
myagg.variance = 0;
myagg.totalRows = 0;
myagg.sampleRows = 0;
}
private boolean warned = false;
@Override
public void iterate(AggregationBuffer agg, Object[] parameters) throws HiveException {
assert (parameters.length == 3);
Object p = parameters[0];
if (p != null) {
SumLongAgg myagg = (SumLongAgg) agg;
if (parameters.length == 3) {
myagg.totalRows = PrimitiveObjectInspectorUtils.getLong(parameters[2],
totalRowsOI);
myagg.sampleRows = PrimitiveObjectInspectorUtils.getLong(parameters[1],
sampleRowsOI);
}
try {
long v = PrimitiveObjectInspectorUtils.getLong(p, inputOI);
myagg.count++;
myagg.sum += v;
if (myagg.count > 1) {
double t = myagg.count * v - myagg.sum;
myagg.variance += (t * t) / ((double) myagg.count * (myagg.count - 1));
}
} catch (NumberFormatException e) {
if (!warned) {
warned = true;
LOG.warn(getClass().getSimpleName() + " "
+ StringUtils.stringifyException(e));
}
}
}
}
@Override
public Object terminatePartial(AggregationBuffer agg) throws HiveException {
SumLongAgg myagg = (SumLongAgg) agg;
((BooleanWritable) partialResult[0]).set(myagg.empty);
((LongWritable) partialResult[1]).set(myagg.count);
((LongWritable) partialResult[2]).set(myagg.sum);
((DoubleWritable) partialResult[3]).set(myagg.variance);
((LongWritable) partialResult[4]).set(myagg.totalRows);
((LongWritable) partialResult[5]).set(myagg.sampleRows);
return partialResult;
}
@Override
public void merge(AggregationBuffer agg, Object partial) throws HiveException {
if (partial != null) {
SumLongAgg myagg = (SumLongAgg) agg;
myagg.empty = false;
Object partialCount = soi.getStructFieldData(partial, countField);
Object partialSum = soi.getStructFieldData(partial, sumField);
Object partialVariance = soi.getStructFieldData(partial, varianceField);
Object partialTotalRows = soi.getStructFieldData(partial, totalRowsField);
Object partialSampleRows = soi.getStructFieldData(partial, sampleRowsField);
long n = myagg.count;
long m = countFieldOI.get(partialCount);
long q = totalRowsFieldOI.get(partialTotalRows);
long r = sampleRowsFieldOI.get(partialSampleRows);
myagg.totalRows = q;
myagg.sampleRows = r;
if (n == 0) {
// Just copy the information since there is nothing so far
myagg.variance = varianceFieldOI.get(partialVariance);
myagg.count = countFieldOI.get(partialCount);
myagg.sum = sumFieldOI.get(partialSum);
}
if (m != 0 && n != 0) {
// Merge the two partials
long a = myagg.sum;
long b = sumFieldOI.get(partialSum);
myagg.empty = false;
myagg.count += m;
myagg.sum += b;
double t = (m / (double) n) * a - b;
myagg.variance += varianceFieldOI.get(partialVariance)
+ ((n / (double) m) / ((double) n + m)) * t * t;
}
}
}
@Override
public Object terminate(AggregationBuffer agg) throws HiveException {
SumLongAgg myagg = (SumLongAgg) agg;
if (myagg.empty) {
return null;
}
double approx_sum = ((double) myagg.sum * myagg.totalRows) / myagg.sampleRows;
double probability = ((double) myagg.count) / ((double) myagg.totalRows);
double mean = ((double) myagg.sum) / myagg.count;
LOG.info("Sum: " + approx_sum);
LOG.info("TotalRows: " + myagg.totalRows);
LOG.info("Probability: " + probability);
LOG.info("Sampling Ratio: " + ((double) myagg.sampleRows) / myagg.totalRows);
StringBuilder sb = new StringBuilder();
sb.append(approx_sum);
sb.append(" +/- ");
sb.append(Math.ceil((2.575 * (1.0 * myagg.totalRows / myagg.sampleRows) * Math
.sqrt(probability
* ((myagg.variance) + ((1 - probability) * myagg.totalRows * mean * mean))))));
sb.append(" (99% Confidence) ");
result.set(sb.toString());
return result;
}
}
}