Class CharTokenizer
An abstract base class for simple, character-oriented tokenizers.
You must specify the required LuceneVersion compatibility when creating CharTokenizer:
- As of 3.1, CharTokenizer uses an int based API to normalize and detect token codepoints. See IsTokenChar(Int32) and Normalize(Int32) for details.
A new CharTokenizer API has been introduced with Lucene 3.1. This API moved from UTF-16 code units to UTF-32 codepoints to eventually add support for supplementary characters. The old char based API has been deprecated and should be replaced with the int based methods IsTokenChar(Int32) and Normalize(Int32).
As of Lucene 3.1 each CharTokenizer - constructor expects a LuceneVersion argument. Based on the given LuceneVersion either the new API or a backwards compatibility layer is used at runtime. For LuceneVersion < 3.1 the backwards compatibility layer ensures correct behavior even for indexes build with previous versions of Lucene. If a LuceneVersion >= 3.1 is used CharTokenizer requires the new API to be implemented by the instantiated class. Yet, the old char based API is not required anymore even if backwards compatibility must be preserved. CharTokenizer subclasses implementing the new API are fully backwards compatible if instantiated with LuceneVersion < 3.1.
Note: If you use a subclass of CharTokenizer with LuceneVersion >= 3.1 on an index build with a version < 3.1, created tokens might not be compatible with the terms in your index.
Inherited Members
Assembly: Lucene.Net.Analysis.Common.dll
Syntax
[Serializable]
public abstract class CharTokenizer : Tokenizer, IDisposableConstructors
| Name | Description |
|---|---|
| CharTokenizer(LuceneVersion, AttributeSource.AttributeFactory, TextReader) | Creates a new CharTokenizer instance |
| CharTokenizer(LuceneVersion, TextReader) | Creates a new CharTokenizer instance |
Methods
| Name | Description |
|---|---|
| End() | |
| IncrementToken() | |
| IsTokenChar(Int32) | Returns true iff a codepoint should be included in a token. This tokenizer generates as tokens adjacent sequences of codepoints which satisfy this predicate. Codepoints for which this is false are used to define token boundaries and are not included in tokens. |
| Normalize(Int32) | Called on each token character to normalize it before it is added to the token. The default implementation does nothing. Subclasses may use this to, e.g., lowercase tokens. |
| Reset() |
