Package com.apple.foundationdb.record.provider.foundationdb.indexes

Interface AtomicMutation

All Known Implementing Classes:

AtomicMutation.Standard

@API(MAINTAINED)
public interface AtomicMutation

The particular operation to be performed by an AtomicMutationIndexMaintainer index.

Nested Class Summary

Nested Classes

Modifier and Type	Interface	Description
static class	AtomicMutation.Standard	The atomic mutations implemented straightforwardly by the FDB API.

Method Summary

Modifier and Type	Method	Description
boolean	allowsNegative()	Determine whether this type allows negative values.
BiFunction<Tuple,Tuple,Tuple>	getAggregator()	Get a function to aggregate multiple index entries.
byte[]	getCompareAndClearParam()	Get a value that causes the index entry to be removed if the result of the mutation matches.
Tuple	getIdentity()	Get the initial value for aggregating multiple index entries.
byte[]	getMutationParam(IndexEntry value, boolean remove)	Get the underlying argument to the FDB API.
MutationType	getMutationType()	Get the underlying mutation performed by the FDB API.
boolean	hasLongValue()	Determine whether this type aggregates long (integer) values.
boolean	hasSingleValue()	Determine whether this type aggregates exactly one value.
boolean	hasValues()	Determine whether this type aggregates values (as opposed to something like counting records).
boolean	isIdempotent()	Determine whether this type is idempotent.

Method Detail

getMutationType

@Nonnull
MutationType getMutationType()

Get the underlying mutation performed by the FDB API.

Returns:

the underlying mutation type

getMutationParam

@Nullable
byte[] getMutationParam(IndexEntry value,
                        boolean remove)

Get the underlying argument to the FDB API.

Parameters:

value - the Tuple form of the value being stored into the index

remove - true if the entry is being removed from the index

Returns:

a byte array to pass to the FDB API or null to do nothing for this mutation

getAggregator

@Nonnull
BiFunction<Tuple,Tuple,Tuple> getAggregator()

Get a function to aggregate multiple index entries. For example, summing subtotals or maxing individual maxima.

Returns:

a function that combines a running aggregate with a single entry to produce a new aggregate

See Also:

RecordCursor.reduce(U, java.util.function.BiFunction<U, ? super T, U>)

getIdentity

@Nullable
Tuple getIdentity()

Get the initial value for aggregating multiple index entries.

Returns:

the initial value of a running aggregate for this type

See Also:

RecordCursor.reduce(U, java.util.function.BiFunction<U, ? super T, U>)

hasValues

boolean hasValues()

Determine whether this type aggregates values (as opposed to something like counting records). The values are specified in the grouped part of the GroupingKeyExpression.

Returns:

true if values are allowed

hasSingleValue

boolean hasSingleValue()

Determine whether this type aggregates exactly one value.

Returns:

true if only a single value is allowed

hasLongValue

boolean hasLongValue()

Determine whether this type aggregates long (integer) values.

Returns:

true if only a long value is allowed

allowsNegative

boolean allowsNegative()

Determine whether this type allows negative values. If negative values are not allowed, an exception is thrown when a record is saved that contains a negative value in the indexed field.

Returns:

true if negative values are allowed

isIdempotent

boolean isIdempotent()

Determine whether this type is idempotent. Max and min type operations are idempotent; sum and count type operations are not.

Returns:

true if updating the index multiple times with the same value yields the same result

getCompareAndClearParam

@Nullable
byte[] getCompareAndClearParam()

Get a value that causes the index entry to be removed if the result of the mutation matches.

Returns:

a byte array to pass to COMPARE_AND_CLEAR or null to do nothing for this mutation