Package org.apache.lucene.document.column

Class DictionaryColumn

java.lang.Object

org.apache.lucene.document.column.Column

org.apache.lucene.document.column.DictionaryColumn

public abstract class DictionaryColumn
extends Column

A Column that provides string or binary values via a pre-defined term dictionary plus per-doc ordinals into that dictionary. Used for SORTED and SORTED_SET doc values, for stored binary or string fields, and for term inversion (tokenized or untokenized).

Iteration is performed via cursors. tuples() is always available and yields (docID, ordinal) pairs. values() is a bulk cursor over consecutive doc-ids; it must be overridden when Column.density() is DENSE and is only consulted in that case.

The caller supplies a fixed List<BytesRef> dictionary at construction. Per-doc ordinals returned by cursors index into this dictionary.

Duplicate dictionary entries are permitted; two slots with the same bytes will both resolve to the same Lucene-level ordinal. The dictionary may be in any order.

The dictionary list and the backing byte arrays of its entries must not be mutated after the column is constructed.

WARNING: This API is experimental and might change in incompatible ways in the next release.

Nested Class Summary

Nested classes/interfaces inherited from class org.apache.lucene.document.column.Column

Column.Density

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	DictionaryColumn(String name, IndexableFieldType fieldType, Column.Density density, List<BytesRef> dictionary)	Creates a DictionaryColumn.

Method Summary

Modifier and Type	Method	Description
final List<BytesRef>	dictionary()	Returns the term dictionary.
StoredValue.Type	storedType()	The stored-field type emitted for this column.
abstract OrdinalsTupleCursor	tuples()	Returns a fresh tuple cursor starting at the beginning of the batch.
OrdinalsCursor	values()	Returns a fresh dense cursor for doc-ids [0, numDocs), producing exactly one ordinal per doc.

Methods inherited from class org.apache.lucene.document.column.Column

density, fieldType, name

Methods inherited from class java.lang.Object

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

Constructor Details

DictionaryColumn

protected DictionaryColumn(String name,
 IndexableFieldType fieldType,
 Column.Density density,
 List<BytesRef> dictionary)

Creates a DictionaryColumn.

Parameters:

name - the field name

fieldType - describes how this field should be indexed

density - whether every batch-local doc-id has a value

dictionary - the term universe; entries must be non-null and no longer than ByteBlockPool.BYTE_BLOCK_SIZE - 2. Must contain at least one entry. Duplicate entries are allowed but incur a minor per-batch cost.

Method Details

dictionary

public final List<BytesRef> dictionary()

Returns the term dictionary. The list is indexed by ordinal; cursors must produce values in [0, dictionary().size()).

tuples

public abstract OrdinalsTupleCursor tuples()

Returns a fresh tuple cursor starting at the beginning of the batch. Always available, regardless of Column.density().

values

public OrdinalsCursor values()

Returns a fresh dense cursor for doc-ids [0, numDocs), producing exactly one ordinal per doc. Must be overridden when Column.density() is DENSE; the default implementation throws UnsupportedOperationException and is never called for SPARSE columns.

storedType

public StoredValue.Type storedType()

The stored-field type emitted for this column. The default is StoredValue.Type.BINARY. Only StoredValue.Type.BINARY and StoredValue.Type.STRING are supported; subclasses may override to emit string stored values.