"Fossies" - the Fresh Open Source Software Archive  

Source code changes of the file "pdfbox/src/main/java/org/apache/pdfbox/text/PDFMarkedContentExtractor.java" between
pdfbox-2.0.23-src.zip and pdfbox-2.0.24-src.zip

About: Apache PDFBox is a Java PDF library tool that allows creation of new PDF documents, manipulation of existing documents and the ability to extract content from documents.

PDFMarkedContentExtractor.java  (pdfbox-2.0.23-src)	:	PDFMarkedContentExtractor.java  (pdfbox-2.0.24-src)
skipping to change at line 43		skipping to change at line 43
import org.apache.pdfbox.contentstream.operator.markedcontent.DrawObject;		import org.apache.pdfbox.contentstream.operator.markedcontent.DrawObject;
import org.apache.pdfbox.contentstream.operator.markedcontent.EndMarkedContentSe quence;		import org.apache.pdfbox.contentstream.operator.markedcontent.EndMarkedContentSe quence;
/**		/**
* This is an stream engine to extract the marked content of a pdf.		* This is an stream engine to extract the marked content of a pdf.
*		*
* @author Johannes Koch		* @author Johannes Koch
*/		*/
public class PDFMarkedContentExtractor extends LegacyPDFStreamEngine		public class PDFMarkedContentExtractor extends LegacyPDFStreamEngine
{		{
private final boolean suppressDuplicateOverlappingText = true;		private boolean suppressDuplicateOverlappingText = true;
private final List<PDMarkedContent> markedContents = new ArrayList<PDMarkedC ontent>();		private final List<PDMarkedContent> markedContents = new ArrayList<PDMarkedC ontent>();
private final Deque<PDMarkedContent> currentMarkedContents = new ArrayDeque< PDMarkedContent>();		private final Deque<PDMarkedContent> currentMarkedContents = new ArrayDeque< PDMarkedContent>();
private final Map<String, List<TextPosition>> characterListMapping = new Has hMap<String, List<TextPosition>>();		private final Map<String, List<TextPosition>> characterListMapping = new Has hMap<String, List<TextPosition>>();
/**		/**
* Instantiate a new PDFTextStripper object.		* Instantiate a new PDFTextStripper object.
*/		*/
public PDFMarkedContentExtractor() throws IOException		public PDFMarkedContentExtractor() throws IOException
{		{
this(null);		this(null);
skipping to change at line 72		skipping to change at line 72
{		{
addOperator(new BeginMarkedContentSequenceWithProperties());		addOperator(new BeginMarkedContentSequenceWithProperties());
addOperator(new BeginMarkedContentSequence());		addOperator(new BeginMarkedContentSequence());
addOperator(new EndMarkedContentSequence());		addOperator(new EndMarkedContentSequence());
addOperator(new DrawObject());		addOperator(new DrawObject());
// todo: DP - Marked Content Point		// todo: DP - Marked Content Point
// todo: MP - Marked Content Point with Properties		// todo: MP - Marked Content Point with Properties
}		}
/**		/**
		* @return the suppressDuplicateOverlappingText setting.
		*/
		public boolean isSuppressDuplicateOverlappingText()
		{
		return suppressDuplicateOverlappingText;
		}
		/**
		* By default the class will attempt to remove text that overlaps each other
		. Word paints the
		* same character several times in order to make it look bold. By setting th
		is to false all text
		* will be extracted, which means that certain sections will be duplicated,
		but better
		* performance will be noticed.
		*
		* @param suppressDuplicateOverlappingText The suppressDuplicateOverlappingT
		ext setting to set.
		*/
		public void setSuppressDuplicateOverlappingText(boolean suppressDuplicateOve
		rlappingText)
		{
		this.suppressDuplicateOverlappingText = suppressDuplicateOverlappingText
		;
		}
		/**
* This will determine of two floating point numbers are within a specified variance.		* This will determine of two floating point numbers are within a specified variance.
*		*
* @param first The first number to compare to.		* @param first The first number to compare to.
* @param second The second number to compare to.		* @param second The second number to compare to.
* @param variance The allowed variance.		* @param variance The allowed variance.
*/		*/
private boolean within( float first, float second, float variance )		private boolean within( float first, float second, float variance )
{		{
return second > first - variance && second < first + variance;		return second > first - variance && second < first + variance;
}		}
skipping to change at line 159		skipping to change at line 180
// of padding are applied, then backed off (not sure why this is don e, but there		// of padding are applied, then backed off (not sure why this is don e, but there
// are cases where the padding is on the order of 10x the character width, and		// are cases where the padding is on the order of 10x the character width, and
// the TJ just backs up to compensate after each character). Also, we subtract		// the TJ just backs up to compensate after each character). Also, we subtract
// an amount to allow for kerning (a percentage of the width of the last		// an amount to allow for kerning (a percentage of the width of the last
// character).		// character).
//		//
boolean suppressCharacter = false;		boolean suppressCharacter = false;
float tolerance = (text.getWidth()/textCharacter.length())/3.0f;		float tolerance = (text.getWidth()/textCharacter.length())/3.0f;
for (TextPosition sameTextCharacter : sameTextCharacters)		for (TextPosition sameTextCharacter : sameTextCharacters)
{		{
TextPosition character = sameTextCharacter;		String charCharacter = sameTextCharacter.getUnicode();
String charCharacter = character.getUnicode();		float charX = sameTextCharacter.getX();
float charX = character.getX();		float charY = sameTextCharacter.getY();
float charY = character.getY();
//only want to suppress		//only want to suppress
if( charCharacter != null &&		if( charCharacter != null &&
//charCharacter.equals( textCharacter ) &&		//charCharacter.equals( textCharacter ) &&
within( charX, textX, tolerance ) &&		within( charX, textX, tolerance ) &&
within( charY,		within( charY,
textY,		textY,
tolerance ) )		tolerance ) )
{		{
suppressCharacter = true;		suppressCharacter = true;
break;		break;
End of changes. 3 change blocks.
5 lines changed or deleted		31 lines changed or added

---