com.aliasi.sentences
Class SentenceAnnotateFilter

java.lang.Object
  org.xml.sax.helpers.DefaultHandler
      com.aliasi.xml.SimpleElementHandler
          com.aliasi.xml.SAXFilterHandler
              com.aliasi.xml.ElementStackFilter
                  com.aliasi.xml.TextContentFilter
                      com.aliasi.sentences.SentenceAnnotateFilter

All Implemented Interfaces:

ContentHandler, DTDHandler, EntityResolver, ErrorHandler

public class SentenceAnnotateFilter
extends TextContentFilter

A SentenceAnnotateFilter applies sentence-boundary annotation to the text content of the specified elements. An instance is constructed with a sentence model and a tokenizer factory. Optionally, an array of elements to annotate may be provided; if no array is specified, all text content is annotated.

The element sent is used to wrap sentences. If the filtered element contains only whitespace, it is not annotated. There will be no whitespace characters at the start or end of a sentence element's text content. All inter-sentence whitespace is retained, but included between sentence elements in the filtered element's content. For instance, the input <p> A b. C d. </p> will yield <p> <sent>A b.</sent> <sent>C d.</sent> </p>. Note that the text of a sentence element starts with the first character of the first token and ends with the last character of the last token. Inter-sentential whitespace winds up as text content outside of the sentence. In this case, there is a single whitespace before the first sentence, two spaces between the sentences, and a single space after the second sentence.

Since:

LingPipe1.0

Version:

1.0.3

Author:

Bob Carpenter

Field Summary

static String

SENTENCE_ELEMENT Element used to group sentences in sentence annotation, namely "sent".

Fields inherited from class com.aliasi.xml.SAXFilterHandler

mHandler

Fields inherited from class com.aliasi.xml.SimpleElementHandler

CDATA_ATTS_TYPE, EMPTY_ATTS, NO_OP_DEFAULT_HANDLER

Constructor Summary

SentenceAnnotateFilter(SentenceModel sentenceModel, TokenizerFactory tokenizerFactory)
Constructs a sentence annotation filter with the specified sentence model and tokenizer factory.

SentenceAnnotateFilter(SentenceModel sentenceModel, TokenizerFactory tokenizerFactory, String[] elements)
Constructs a sentence annotation filter with the specified sentence model and tokenizer factory, and elements whose text content should be annotated.

Method Summary

void	characters(char[] cs, int start, int length) Annotates characters if all characters are being annotated, otherwise annotates if in an annotated element, otherwise passing characters directly to contained handler.
void	filteredCharacters(char[] cs, int start, int length) Performs sentence-boundary annotation of the specified characters.

Methods inherited from class com.aliasi.xml.TextContentFilter

filterElement

Methods inherited from class com.aliasi.xml.ElementStackFilter

currentAttributes, currentElement, endElement, getAttributesStack, getElementStack, noElement, startDocument, startElement

Methods inherited from class com.aliasi.xml.SAXFilterHandler

endDocument, endPrefixMapping, error, fatalError, ignorableWhitespace, notationDecl, processingInstruction, resolveEntity, setDocumentLocator, setHandler, skippedEntity, startPrefixMapping, unparsedEntityDecl, warning

Methods inherited from class com.aliasi.xml.SimpleElementHandler

addSimpleAttribute, characters, characters, characters, characters, createAttributes, createAttributes, createAttributes, createAttributes, createAttributes, createAttributes, endSimpleElement, endSimpleElement, startEndSimpleElement, startEndSimpleElement, startSimpleElement, startSimpleElement, startSimpleElement, startSimpleElement, startSimpleElement, startSimpleElement, startSimpleElement

Methods inherited from class java.lang.Object

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

Field Detail

SENTENCE_ELEMENT

public static final String SENTENCE_ELEMENT

Element used to group sentences in sentence annotation, namely "sent".

See Also:

Constant Field Values

Constructor Detail

SentenceAnnotateFilter

public SentenceAnnotateFilter(SentenceModel sentenceModel,
                              TokenizerFactory tokenizerFactory)

Constructs a sentence annotation filter with the specified sentence model and tokenizer factory.

Parameters:

sentenceModel - Sentence model to use for boundary detection.

tokenizerFactory - Factory to produce tokenizers for text.

SentenceAnnotateFilter

public SentenceAnnotateFilter(SentenceModel sentenceModel,
                              TokenizerFactory tokenizerFactory,
                              String[] elements)

Constructs a sentence annotation filter with the specified sentence model and tokenizer factory, and elements whose text content should be annotated.

Parameters:

sentenceModel - Sentence model to use for boundary detection.

tokenizerFactory - Factory to produce tokenizers for text.

elements - List of elements to be annotated.

Method Detail

characters

public void characters(char[] cs,
                       int start,
                       int length)
                throws SAXException

Annotates characters if all characters are being annotated, otherwise annotates if in an annotated element, otherwise passing characters directly to contained handler. All boundary events will be passed to the contained handler.

Specified by:

characters in interface ContentHandler

Overrides:

characters in class TextContentFilter

Parameters:

cs - Character array to filter.

start - First character to filter.

length - Number of characters to filter.

Throws:

SAXException - If there is an exception thrown by the contained handler.

filteredCharacters

public void filteredCharacters(char[] cs,
                               int start,
                               int length)
                        throws SAXException

Performs sentence-boundary annotation of the specified characters. Markup and text SAX events are delegated to the contained handler.

Specified by:

filteredCharacters in class TextContentFilter

Parameters:

cs - Character array to annotate.

start - First character to annotate.

length - Number of characters to annotate.

Throws:

SAXException - If there is an exception thrown by the contained handler.