org.knowceans.corpus.parsers.nips
Class NipsXmlReader

java.lang.Object
  org.knowceans.corpus.parsers.nips.NipsXmlReader

public class NipsXmlReader
extends java.lang.Object

NipsXmlReader parsers one XML file converted from NIPS PDF documents using XPDF-basd pdftohtml

Author:

heinrich

Field Summary

(package private) java.lang.String	_bold
(package private) java.lang.String	_fontspec
(package private) java.lang.String	_italics
(package private) java.lang.String	_page
(package private) java.lang.String	_text
(package private) java.util.regex.Pattern	boldpattern
(package private) java.util.HashMap<java.lang.String,java.lang.Double>	fontsizes
(package private) java.util.regex.Pattern	fontspecpattern
private boolean	includerefs whether to include references in the text
private boolean	inrefs
(package private) java.util.regex.Pattern	italicspattern
(package private) java.util.regex.Pattern	pagepattern
(package private) java.util.regex.Pattern	textpattern

Constructor Summary

NipsXmlReader()
initialise reader (e.g., compile regex patterns)

Method Summary

private java.lang.String	clean(java.lang.String in) cleans the string.
private java.lang.String	cleanUp(java.lang.String in)
void	extract(java.lang.String filename, NipsDocument doc)
private java.util.Vector<java.lang.String>	extractPage(java.lang.StringBuffer content) extracts the content of a page.
private java.lang.String[]	getHead(java.lang.StringBuffer content) extract title, authors and abstract from (the first page of a) document.
private java.util.Vector<java.lang.String>	getPages(java.lang.StringBuffer content)
static void	main(java.lang.String[] args)
private void	processText(java.lang.StringBuffer content, NipsDocument doc)
private java.lang.String	replaceUmlauts(java.lang.String in) inserts umlauts TODO: direct unicode multicharacter replacements
private int	setFonts(java.lang.StringBuffer content) sets the font sizes for the current page

Methods inherited from class java.lang.Object

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

Field Detail

_page

java.lang.String _page

_fontspec

java.lang.String _fontspec

_text

java.lang.String _text

_bold

java.lang.String _bold

_italics

java.lang.String _italics

pagepattern

java.util.regex.Pattern pagepattern

fontspecpattern

java.util.regex.Pattern fontspecpattern

textpattern

java.util.regex.Pattern textpattern

boldpattern

java.util.regex.Pattern boldpattern

italicspattern

java.util.regex.Pattern italicspattern

fontsizes

java.util.HashMap<java.lang.String,java.lang.Double> fontsizes

inrefs

private boolean inrefs

includerefs

private boolean includerefs

whether to include references in the text

Constructor Detail

NipsXmlReader

public NipsXmlReader()

initialise reader (e.g., compile regex patterns)

Method Detail

main

public static void main(java.lang.String[] args)

extract

public void extract(java.lang.String filename,
                    NipsDocument doc)

Parameters:

filename -

doc - document record (existing or will be created)

processText

private void processText(java.lang.StringBuffer content,
                         NipsDocument doc)

Parameters:

content -

getHead

private java.lang.String[] getHead(java.lang.StringBuffer content)

extract title, authors and abstract from (the first page of a) document.

Parameters:

content - the content that the heading data is extracted from. The extracted parts are stripped from

Returns:

title, author names (semicolon-separated), abstract.

cleanUp

private java.lang.String cleanUp(java.lang.String in)

Parameters:

in -

Returns:

setFonts

private int setFonts(java.lang.StringBuffer content)

sets the font sizes for the current page

Parameters:

content -

Returns:

end of last group

replaceUmlauts

private java.lang.String replaceUmlauts(java.lang.String in)

inserts umlauts TODO: direct unicode multicharacter replacements

Parameters:

in -

Returns:

clean

private java.lang.String clean(java.lang.String in)

cleans the string.

Parameters:

in -

Returns:

getPages

private java.util.Vector<java.lang.String> getPages(java.lang.StringBuffer content)

Parameters:

content -

Returns:

extractPage

private java.util.Vector<java.lang.String> extractPage(java.lang.StringBuffer content)

extracts the content of a page. Sections are recognised by boldface > 13pt

Parameters:

content -

Returns: