org.apache.pig

Class EvalFunc<T>

java.lang.Object

org.apache.pig.EvalFunc<T>

Direct Known Subclasses:

ABS, ABS, AccumulatorEvalFunc, AddDuration, AlgebraicBigDecimalMathBase, AlgebraicBigIntegerMathBase, AlgebraicByteArrayMathBase, AlgebraicByteArrayMathBase.Initial, AlgebraicDoubleMathBase, AlgebraicFloatMathBase, AlgebraicIntMathBase, AlgebraicLongMathBase, AlgebraicMathBase.Final, AlgebraicMathBase.Intermediate, ARITY, Assert, AVG, AVG.Final, AVG.Initial, AVG.Intermediate, BagSize, BagToString, BagToTuple, Base, Base, BigDecimalAbs, BigDecimalAvg, BigDecimalAvg.Final, BigDecimalAvg.Initial, BigDecimalAvg.Intermediate, BigIntegerAbs, BigIntegerAvg, BigIntegerAvg.Final, BigIntegerAvg.Initial, BigIntegerAvg.Intermediate, Bin, BinCond, BuildBloomBase, CONCAT, ConstantSize, copySign, COR, COR, COR.Final, COR.Final, COR.Initial, COR.Initial, COR.Intermed, COR.Intermed, COUNT, COUNT_STAR, COUNT_STAR.Final, COUNT_STAR.Initial, COUNT_STAR.Intermediate, COUNT.Final, COUNT.Initial, COUNT.Intermediate, COV, COV, COV.Final, COV.Final, COV.Initial, COV.Initial, COV.Intermed, COV.Intermed, CubeDimensions, CurrentTime, CustomFormatToISO, DateExtractor, DateTimeMax, DateTimeMax.Final, DateTimeMax.Initial, DateTimeMax.Intermediate, DateTimeMin, DateTimeMin.Final, DateTimeMin.Initial, DateTimeMin.Intermediate, DaysBetween, Decode, DIFF, DiffDate, Distinct, Distinct.Final, Distinct.Initial, Distinct.Intermediate, DoubleAbs, DoubleAvg, DoubleAvg.Final, DoubleAvg.Initial, DoubleAvg.Intermediate, DoubleCopySign, DoubleGetExponent, DoubleMax, DoubleMin, DoubleNextAfter, DoubleNextup, DoubleRound, DoubleRound, DoubleRoundTo, DoubleSignum, DoubleUlp, ENDSWITH, ExtremalTupleByNthField, ExtremalTupleByNthField.HelperClass, FilterFunc, FindQuantiles, FloatAbs, FloatAbs, FloatAvg, FloatAvg.Final, FloatAvg.Initial, FloatAvg.Intermediate, FloatCopySign, FloatGetExponent, FloatMax, FloatMin, FloatNextAfter, FloatNextup, FloatRound, FloatRound, FloatRoundTo, FloatSignum, FloatUlp, GenericInvoker, GetDay, getExponent, GetHour, GetMemNumRows, GetMilliSecond, GetMinute, GetMonth, GetSecond, GetWeek, GetWeekYear, GetYear, GFAny, GFCross, GFReplicate, GroovyEvalFunc, HashFNV, HiveUDAF, HiveUDAF.Final, HiveUDAF.Initial, HiveUDAF.Intermediate, HiveUDF, HiveUDTF, HostExtractor, HoursBetween, IdentityColumn, INDEXOF, IntAbs, IntAbs, IntAvg, IntAvg.Final, IntAvg.Initial, IntAvg.Intermediate, IntMax, IntMin, INVERSEMAP, InvokerGenerator, IsDouble, IsFirstReduceOfKey, IsFloat, IsInt, IsLong, IsNumeric, ISODaysBetween, ISOHoursBetween, ISOMinutesBetween, ISOMonthsBetween, ISOSecondsBetween, ISOToDay, ISOToHour, ISOToMinute, ISOToMonth, ISOToSecond, ISOToUnix, ISOToWeek, ISOToYear, ISOYearsBetween, JrubyAlgebraicEvalFunc.AlgebraicFunctionWrapper, JrubyEvalFunc, JsFunction, JythonFunction, KEYSET, LAST_INDEX_OF, LCFIRST, LENGTH, LongAbs, LongAbs, LongAvg, LongAvg.Final, LongAvg.Initial, LongAvg.Intermediate, LongMax, LongMin, LookupInFiles, LOWER, LTRIM, MapSize, MAX, MaxTupleBy1stField, MaxTupleBy1stField.Final, MaxTupleBy1stField.Initial, MaxTupleBy1stField.Intermediate, MilliSecondsBetween, MIN, MinutesBetween, MonthsBetween, nextAfter, NEXTUP, Over, PartitionSkewedKeys, PluckTuple, RANDOM, ReadScalars, ReadScalarsTez, REGEX_EXTRACT, REGEX_EXTRACT_ALL, REGEX_SEARCH, RegexExtract, RegexExtractAll, RegexMatch, REPLACE, REPLACE_MULTI, Reverse, RollupDimensions, ROUND, ROUND, ROUND_TO, RTRIM, SCALB, SearchEngineExtractor, SearchQuery, SearchTermExtractor, SecondsBetween, SIGNUM, SIZE, SPRINTF, STARTSWITH, Stitch, StreamingUDF, StringConcat, StringMax, StringMax.Final, StringMax.Initial, StringMax.Intermediate, StringMin, StringMin.Final, StringMin.Initial, StringMin.Intermediate, StringSize, STRSPLIT, STRSPLITTOBAG, Stuff, SUBSTRING, SUBTRACT, SubtractDuration, ToBag, TOBAG, ToDate, ToDate2ARGS, ToDate3ARGS, ToDateISO, TOKENIZE, TOMAP, ToMilliSeconds, Top, Top.Final, TOP.Final, Top.Initial, TOP.Initial, Top.Intermed, TOP.Intermed, ToString, ToTuple, TOTUPLE, ToUnixTime, TRIM, TupleSize, TypedOutputEvalFunc, UCFIRST, ULP, UniqueID, UnixToISO, UPPER, UPPER, VALUELIST, VALUESET, WeeksBetween, XPath, XPathAll, YearsBetween

@InterfaceAudience.Public
 @InterfaceStability.Stable
public abstract class EvalFunc<T>
extends java.lang.Object

The class is used to implement functions to be applied to fields in a dataset. The function is applied to each Tuple in the set. The programmer should not make assumptions about state maintained between invocations of the exec() method since the Pig runtime will schedule and localize invocations based on information provided at runtime. The programmer also should not make assumptions about when or how many times the class will be instantiated, since it may be instantiated multiple times in both the front and back end.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	EvalFunc.SchemaType EvalFunc's schema type.

Field Summary

Fields

Modifier and Type	Field and Description
protected org.apache.commons.logging.Log	log Logging object.
protected PigLogger	pigLogger Logger for aggregating warnings.
protected PigProgressable	reporter Reporter to send heartbeats to Hadoop.
protected java.lang.reflect.Type	returnType Return type of this instance of EvalFunc.

Constructor Summary

Constructors

Constructor and Description
EvalFunc()

Method Summary

Modifier and Type	Method and Description
boolean	allowCompileTimeCalculation() Whether the UDF should be evaluated at compile time if all inputs are constant.
abstract T	exec(Tuple input) This callback method must be implemented by all subclasses.
void	finish() Placeholder for cleanup to be performed at the end.
java.util.List<FuncSpec>	getArgToFuncMapping() Allow a UDF to specify type specific implementations of itself.
java.util.List<java.lang.String>	getCacheFiles() Allow a UDF to specify a list of hdfs files it would like placed in the distributed cache.
Schema	getInputSchema() This method is intended to be called by the user in EvalFunc to get the input schema of the EvalFunc
LoadCaster	getLoadCaster() This will be called on both the front end and the back end during execution.
org.apache.commons.logging.Log	getLogger()
PigLogger	getPigLogger()
PigProgressable	getReporter()
java.lang.reflect.Type	getReturnType() Get the Type that this EvalFunc returns.
protected java.lang.String	getSchemaName(java.lang.String name, Schema input)
EvalFunc.SchemaType	getSchemaType() Returns the EvalFunc.SchemaType of the EvalFunc.
java.util.List<java.lang.String>	getShipFiles() Allow a UDF to specify a list of local files it would like placed in the distributed cache.
boolean	isAsynchronous() Deprecated.
boolean	needEndOfAllInputProcessing()
Schema	outputSchema(Schema input) Report the schema of the output of this UDF.
void	progress() Utility method to allow UDF to report progress.
void	setEndOfAllInput(boolean endOfAllInput)
void	setInputSchema(Schema input) This method is for internal use.
void	setPigLogger(PigLogger pigLogger) Set the PigLogger object.
void	setReporter(PigProgressable reporter) Set the reporter.
void	setUDFContextSignature(java.lang.String signature) This method will be called by Pig both in the front end and back end to pass a unique signature to the EvalFunc.
void	warn(java.lang.String msg, java.lang.Enum warningEnum) Issue a warning.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

reporter

protected PigProgressable reporter

Reporter to send heartbeats to Hadoop. If exec will take more than a a few seconds PigProgressable.progress() should be called occasionally to avoid timeouts. Default Hadoop timeout is 600 seconds.

log

protected org.apache.commons.logging.Log log

Logging object. Log calls made on the front end will be sent to pig's log on the client. Log calls made on the backend will be sent to stdout and can be seen in the Hadoop logs.

pigLogger

protected PigLogger pigLogger

Logger for aggregating warnings. Any warnings to be sent to the user should be logged to this via PigLogger.warn(java.lang.Object, java.lang.String, java.lang.Enum<?>).

returnType

protected java.lang.reflect.Type returnType

Return type of this instance of EvalFunc.

Constructor Detail

EvalFunc

public EvalFunc()

Method Detail

getSchemaName

protected java.lang.String getSchemaName(java.lang.String name,
                                         Schema input)

getReturnType

public java.lang.reflect.Type getReturnType()

Get the Type that this EvalFunc returns.

Returns:

Type

progress

public final void progress()

Utility method to allow UDF to report progress. If exec will take more than a a few seconds PigProgressable.progress() should be called occasionally to avoid timeouts. Default Hadoop timeout is 600 seconds.

warn

public final void warn(java.lang.String msg,
                       java.lang.Enum warningEnum)

Issue a warning. Warning messages are aggregated and reported to the user.

Parameters:

msg - String message of the warning

warningEnum - type of warning

finish

public void finish()

Placeholder for cleanup to be performed at the end. User defined functions can override. Default implementation is a no-op.

exec

public abstract T exec(Tuple input)
                throws java.io.IOException

This callback method must be implemented by all subclasses. This is the method that will be invoked on every Tuple of a given dataset. Since the dataset may be divided up in a variety of ways the programmer should not make assumptions about state that is maintained between invocations of this method.

Parameters:

input - the Tuple to be processed.

Returns:

result, of type T.

Throws:

java.io.IOException

outputSchema

public Schema outputSchema(Schema input)

Report the schema of the output of this UDF. Pig will make use of this in error checking, optimization, and planning. The schema of input data to this UDF is provided.

The default implementation interprets the OutputSchema annotation, if one is present. Otherwise, it returns null (no known output schema).

Parameters:

input - Schema of the input

Returns:

Schema of the output

isAsynchronous

@Deprecated
public boolean isAsynchronous()

Deprecated.

This function should be overriden to return true for functions that return their values asynchronously. Currently pig never attempts to execute a function asynchronously.

Returns:

true if the function can be executed asynchronously.

getReporter

public PigProgressable getReporter()

setReporter

public final void setReporter(PigProgressable reporter)

Set the reporter. Called by Pig to provide a reference of the reporter to the UDF.

Parameters:

reporter - Hadoop reporter

getArgToFuncMapping

public java.util.List<FuncSpec> getArgToFuncMapping()
                                             throws FrontendException

Allow a UDF to specify type specific implementations of itself. For example, an implementation of arithmetic sum might have int and float implementations, since integer arithmetic performs much better than floating point arithmetic. Pig's typechecker will call this method and using the returned list plus the schema of the function's input data, decide which implementation of the UDF to use.

Returns:

A List containing FuncSpec objects representing the EvalFunc class which can handle the inputs corresponding to the schema in the objects. Each FuncSpec should be constructed with a schema that describes the input for that implementation. For example, the sum function above would return two elements in its list:

1. FuncSpec(this.getClass().getName(), new Schema(new Schema.FieldSchema(null, DataType.DOUBLE)))
2. FuncSpec(IntSum.getClass().getName(), new Schema(new Schema.FieldSchema(null, DataType.INTEGER)))

This would indicate that the main implementation is used for doubles, and the special implementation IntSum is used for ints.

Throws:

FrontendException

getCacheFiles

public java.util.List<java.lang.String> getCacheFiles()

Allow a UDF to specify a list of hdfs files it would like placed in the distributed cache. These files will be put in the cache for every job the UDF is used in. The default implementation returns null.

Returns:

A list of files

getShipFiles

public java.util.List<java.lang.String> getShipFiles()

Allow a UDF to specify a list of local files it would like placed in the distributed cache. These files will be put in the cache for every job the UDF is used in. Check for FuncUtils for utility function to facilitate it The default implementation returns null.

Returns:

A list of files

getPigLogger

public PigLogger getPigLogger()

setPigLogger

public final void setPigLogger(PigLogger pigLogger)

Set the PigLogger object. Called by Pig to provide a reference to the UDF.

Parameters:

pigLogger - PigLogger object.

getLogger

public org.apache.commons.logging.Log getLogger()

setUDFContextSignature

public void setUDFContextSignature(java.lang.String signature)

This method will be called by Pig both in the front end and back end to pass a unique signature to the EvalFunc. The signature can be used to store into the UDFContext any information which the EvalFunc needs to store between various method invocations in the front end and back end.

Parameters:

signature - a unique signature to identify this EvalFunc

setInputSchema

public void setInputSchema(Schema input)

This method is for internal use. It is called by Pig core in both front-end and back-end to setup the right input schema for EvalFunc

getInputSchema

public Schema getInputSchema()

This method is intended to be called by the user in EvalFunc to get the input schema of the EvalFunc

getSchemaType

public EvalFunc.SchemaType getSchemaType()

Returns the EvalFunc.SchemaType of the EvalFunc. User defined functions can override this method to return EvalFunc.SchemaType.VARARG. In this case the last FieldSchema added to the Schema in getArgToFuncMapping() will be considered as a vararg field.

Returns:

the schema type of the UDF

allowCompileTimeCalculation

public boolean allowCompileTimeCalculation()

Whether the UDF should be evaluated at compile time if all inputs are constant. This is applicable for most UDF, however, if a UDF will access hdfs file which is not available at compile time, it has to be false

Returns:

Whether or not compile time calculation is allowed, default to false to ensure legacy UDF will get the right behavior

needEndOfAllInputProcessing

public boolean needEndOfAllInputProcessing()

setEndOfAllInput

public void setEndOfAllInput(boolean endOfAllInput)

getLoadCaster

public LoadCaster getLoadCaster()
                         throws java.io.IOException

This will be called on both the front end and the back end during execution.

Returns:

the LoadCaster associated with this eval. Returning null indicates that casts from bytearray will pick the one associated with the parameters when they all come from the same loadcaster type.

Throws:

java.io.IOException - if there is an exception during LoadCaster