The org.apache.lucene.analysis.standard package contains three
fast grammar-based tokenizers constructed with JFlex:
StandardTokenizer:
as of Lucene 3.1, implements the Word Break rules from the Unicode Text
Segmentation algorithm, as specified in
Unicode Standard Annex #29.
Unlike UAX29URLEmailTokenizer, URLs and email addresses are
not tokenized as single tokens, but are instead split up into
tokens according to the UAX#29 word break rules.
StandardAnalyzer includes
StandardTokenizer,
StandardFilter,
LowerCaseFilter
and StopFilter.
When the Version specified in the constructor is lower than
3.1, the ClassicTokenizer
implementation is invoked.ClassicTokenizer:
this class was formerly (prior to Lucene 3.1) named
StandardTokenizer. (Its tokenization rules are not
based on the Unicode Text Segmentation algorithm.)
ClassicAnalyzer includes
ClassicTokenizer,
StandardFilter,
LowerCaseFilter
and StopFilter.
UAX29URLEmailTokenizer:
implements the Word Break rules from the Unicode Text Segmentation
algorithm, as specified in
Unicode Standard Annex #29.
URLs and email addresses are also tokenized according to the relevant RFCs.