de.digitalcollections.solrocr.formats

Class OcrPassageFormatter

java.lang.Object

org.apache.lucene.search.uhighlight.PassageFormatter

de.digitalcollections.solrocr.formats.OcrPassageFormatter

Direct Known Subclasses:

AltoPassageFormatter, HocrPassageFormatter, MiniOcrPassageFormatter

public abstract class OcrPassageFormatter
extends org.apache.lucene.search.uhighlight.PassageFormatter

Takes care of formatting fragments of the OCR format into OcrSnippet instances.

Field Summary

Fields

Modifier and Type	Field and Description
protected boolean	absoluteHighlights
protected String	endHlTag
protected String	startHlTag

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	OcrPassageFormatter(String startHlTag, String endHlTag, boolean absoluteHighlights)

Method Summary

Modifier and Type	Method and Description
protected void	addHighlightsToSnippet(List<List<OcrBox>> hlBoxes, OcrSnippet snippet)
abstract String	determineStartPage(String ocrFragment, int startOffset, IterableCharSequence content) Determine the id of the page an OCR fragment resides on.
OcrSnippet[]	format(org.apache.lucene.search.uhighlight.Passage[] passages, IterableCharSequence content) Format the passages that point to subsequences of the document text into OcrSnippet instances
Object	format(org.apache.lucene.search.uhighlight.Passage[] passages, String content) Convenience implementation to format document text that is available as a String.
protected String	getTextFromXml(String xmlFragment) Helper method to get plaintext from XML/HTML-like fragments
protected List<OcrBox>	mergeBoxes(List<OcrBox> boxes) Merge adjacent OCR boxes into a single one, taking line breaks into account
protected OcrSnippet	parseFragment(String ocrFragment, String pageId) Parse an OcrSnippet from an OCR fragment.
protected abstract List<OcrBox>	parseWords(String ocrFragment, String startPage) Parse word boxes from an OCR fragment.

Methods inherited from class java.lang.Object

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

Field Detail

startHlTag

protected final String startHlTag

endHlTag

protected final String endHlTag

absoluteHighlights

protected final boolean absoluteHighlights

Constructor Detail

OcrPassageFormatter

protected OcrPassageFormatter(String startHlTag,
                              String endHlTag,
                              boolean absoluteHighlights)

Method Detail

format

public OcrSnippet[] format(org.apache.lucene.search.uhighlight.Passage[] passages,
                           IterableCharSequence content)

Format the passages that point to subsequences of the document text into OcrSnippet instances

Parameters:

passages - in the the document text that contain highlighted text

content - of the OCR field, implemented as an IterableCharSequence

Returns:

the parsed snippet representation of the passages

getTextFromXml

protected String getTextFromXml(String xmlFragment)

Helper method to get plaintext from XML/HTML-like fragments

determineStartPage

public abstract String determineStartPage(String ocrFragment,
                                          int startOffset,
                                          IterableCharSequence content)

Determine the id of the page an OCR fragment resides on.

parseFragment

protected OcrSnippet parseFragment(String ocrFragment,
                                   String pageId)

Parse an OcrSnippet from an OCR fragment.

parseWords

protected abstract List<OcrBox> parseWords(String ocrFragment,
                                           String startPage)

Parse word boxes from an OCR fragment.

addHighlightsToSnippet

protected void addHighlightsToSnippet(List<List<OcrBox>> hlBoxes,
                                      OcrSnippet snippet)

mergeBoxes

protected List<OcrBox> mergeBoxes(List<OcrBox> boxes)

Merge adjacent OCR boxes into a single one, taking line breaks into account

format

public Object format(org.apache.lucene.search.uhighlight.Passage[] passages,
                     String content)

Convenience implementation to format document text that is available as a String. Wraps the String in a IterableCharSequence implementation and calls format(Passage[], IterableCharSequence)

Specified by:

format in class org.apache.lucene.search.uhighlight.PassageFormatter

Parameters:

passages - in the the document text that contain highlighted text

content - of the OCR field, implemented as an IterableCharSequence

Returns:

the parsed snippet representation of the passages