AliasableEvalFunc (DataFu 1.2.0)

Overview

Package

Class

Use

Tree

Deprecated

Index

Help

PREV CLASS NEXT CLASS

FRAMES NO FRAMES

SUMMARY: NESTED | FIELD | CONSTR | METHOD

DETAIL: FIELD | CONSTR | METHOD

datafu.pig.util
Class AliasableEvalFunc<T>

java.lang.Object
  org.apache.pig.EvalFunc<T>
      datafu.pig.util.ContextualEvalFunc<T>
          datafu.pig.util.AliasableEvalFunc<T>


Type Parameters:: T -

Direct Known Subclasses:: BagGroup, BagLeftOuterJoin, Coalesce, TransposeTupleToBag

public abstract class AliasableEvalFunc<T>
extends ContextualEvalFunc<T>
extends ContextualEvalFunc<T>

Makes implementing and using UDFs easier by enabling named parameters.

This works by capturing the schema of the input tuple on the front-end and storing it into the UDFContext. It provides an easy means of referencing the parameters on the back-end to aid in writing schema-based UDFs.

A related class is SimpleEvalFunc. However they are actually fairly different. The primary purpose of SimpleEvalFunc is to skip the boilerplate under the assumption that the arguments in and out are well... simple. It also assumes that these arguments are in a well-defined positional ordering.

AliasableEvalFunc allows the UDF writer to avoid dealing with all positional assumptions and instead reference fields by their aliases. This practice allows for more readable code since the alias names should have more meaning to the reader than the position. This approach is also less error prone since it creates a more explicit contract for what input the UDF expects and prevents simple mistakes that positional-based UDFs could not easily catch, such as transposing two fields of the same type. If this contract is violated, say, by attempting to reference a field that is not present, a meaningful error message may be thrown.

Example: This example computes the monthly payments for mortgages depending on interest rate.

 public class MortgagePayment extends AliasableEvalFunc<DataBag> {
    ...
    public DataBag exec(Tuple input) throws IOException {
      DataBag output = BagFactory.getInstance().newDefaultBag();
      
      Double principal = getDouble(input, "principal"); // get a value from the input tuple by alias
      Integer numPayments = getInteger(input, "num_payments");
      DataBag interestRates = getBag(input, "interest_rates");
    
      for (Tuple interestTuple : interestRates) {
        Double interest = getDouble(interestTuple, getPrefixedAliasName("interest_rates", "interest_rate"));  // get a value from the inner bag tuple by alias
        double monthlyPayment = computeMonthlyPayment(principal, numPayments, interest);
        output.add(TupleFactory.getInstance().newTuple(monthlyPayment));
      }
      return output;
    }
  }

Author:: wvaughan

Field Summary

Fields inherited from class org.apache.pig.EvalFunc
`log, pigLogger, reporter, returnType`

Constructor Summary
`AliasableEvalFunc()`

Method Summary
`org.apache.pig.data.DataBag`	`getBag(org.apache.pig.data.Tuple tuple, java.lang.String alias)`
`java.lang.Boolean`	`getBoolean(org.apache.pig.data.Tuple tuple, java.lang.String alias)`
`java.lang.Double`	`getDouble(org.apache.pig.data.Tuple tuple, java.lang.String alias)`
`java.lang.Double`	`getDouble(org.apache.pig.data.Tuple tuple, java.lang.String alias, java.lang.Double defaultValue)`
`java.util.Map<java.lang.String,java.lang.Integer>`	`getFieldAliases()` Field aliases are generated from the input schema Each alias maps to a bag position Inner bags/tuples will have alias of outer.inner.foo
`java.lang.Float`	`getFloat(org.apache.pig.data.Tuple tuple, java.lang.String alias)`
`java.lang.Float`	`getFloat(org.apache.pig.data.Tuple tuple, java.lang.String alias, java.lang.Float defaultValue)`
`java.lang.Integer`	`getInteger(org.apache.pig.data.Tuple tuple, java.lang.String alias)`
`java.lang.Integer`	`getInteger(org.apache.pig.data.Tuple tuple, java.lang.String alias, java.lang.Integer defaultValue)`
`java.lang.Long`	`getLong(org.apache.pig.data.Tuple tuple, java.lang.String alias)`
`java.lang.Long`	`getLong(org.apache.pig.data.Tuple tuple, java.lang.String alias, java.lang.Long defaultValue)`
`java.lang.Object`	`getObject(org.apache.pig.data.Tuple tuple, java.lang.String alias)`
`abstract org.apache.pig.impl.logicalLayer.schema.Schema`	`getOutputSchema(org.apache.pig.impl.logicalLayer.schema.Schema input)` Specify the output schema as in {link EvalFunc#outputSchema(Schema)}.
`java.lang.Integer`	`getPosition(java.lang.String alias)`
`java.lang.Integer`	`getPosition(java.lang.String prefix, java.lang.String alias)`
`java.lang.String`	`getPrefixedAliasName(java.lang.String prefix, java.lang.String alias)`
`java.lang.String`	`getString(org.apache.pig.data.Tuple tuple, java.lang.String alias)`
`java.lang.String`	`getString(org.apache.pig.data.Tuple tuple, java.lang.String alias, java.lang.String defaultValue)`
`org.apache.pig.impl.logicalLayer.schema.Schema`	`outputSchema(org.apache.pig.impl.logicalLayer.schema.Schema input)` A wrapper method which captures the schema and then calls getOutputSchema

Methods inherited from class datafu.pig.util.ContextualEvalFunc
`getContextProperties, getInstanceName, getInstanceProperties, setUDFContextSignature`

Methods inherited from class org.apache.pig.EvalFunc
`exec, finish, getArgToFuncMapping, getCacheFiles, getInputSchema, getLogger, getPigLogger, getReporter, getReturnType, getSchemaName, isAsynchronous, progress, setInputSchema, setPigLogger, setReporter, warn`

Methods inherited from class java.lang.Object
`clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait`

Constructor Detail