Index: pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFParser.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFParser.java (revision 1226367) +++ pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFParser.java (working copy) @@ -66,7 +66,7 @@ /** Collects all Xref/trailer objects and resolves them into single * object using startxref reference. */ - private XrefTrailerResolver xrefTrailerResolver = new XrefTrailerResolver(); + protected XrefTrailerResolver xrefTrailerResolver = new XrefTrailerResolver(); /** * Temp file directory. @@ -659,7 +659,7 @@ * @return false on parsing error * @throws IOException If an IO error occurs. */ - private boolean parseStartXref() throws IOException + protected boolean parseStartXref() throws IOException { if(pdfSource.peek() != 's') { @@ -686,7 +686,7 @@ * @return false on parsing error * @throws IOException If an IO error occurs. */ - private boolean parseXrefTable( long startByteOffset ) throws IOException + protected boolean parseXrefTable( long startByteOffset ) throws IOException { if(pdfSource.peek() != 'x') { @@ -767,7 +767,7 @@ * @return false on parsing error * @throws IOException If an IO error occurs. */ - private boolean parseTrailer() throws IOException + protected boolean parseTrailer() throws IOException { if(pdfSource.peek() != 't') { Index: pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFXrefStreamParser.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFXrefStreamParser.java (revision 1226367) +++ pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFXrefStreamParser.java (working copy) @@ -150,9 +150,26 @@ break; case 2: /* + * object stored in object stream; 2nd argument is object number of object stream; + * 3rd argument index of object within object stream + * + * For sequential PDFParser we do not need this information + * because * These objects are handled by the dereferenceObjects() method * since they're only pointing to object numbers + * + * However for XRef aware parsers we have to know which objects contain + * object streams. We will store this information in normal xref mapping + * table but add object stream number with minus sign in order to + * distinguish from file offsets */ + int objstmObjNr = 0; + for(int i = 0; i < w1; i++) + { + objstmObjNr += (currLine[i + w0] & 0x00ff) << ((w1 - i - 1) * 8); + } + objKey = new COSObjectKey( objID.intValue(), 0 ); + xrefTrailerResolver.setXRef( objKey, -objstmObjNr ); break; default: break; Index: pdfbox/src/main/java/org/apache/pdfbox/pdfparser/XrefTrailerResolver.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/pdfparser/XrefTrailerResolver.java (revision 1226367) +++ pdfbox/src/main/java/org/apache/pdfbox/pdfparser/XrefTrailerResolver.java (working copy) @@ -19,8 +19,11 @@ import java.util.ArrayList; import java.util.Collections; import java.util.HashMap; +import java.util.HashSet; import java.util.List; import java.util.Map; +import java.util.Set; +import java.util.Map.Entry; import org.apache.commons.logging.Log; import org.apache.commons.logging.LogFactory; @@ -111,6 +114,13 @@ } /** + * Returns the trailer last set by {@link #setTrailer(COSDictionary)}. + */ + public COSDictionary getCurrentTrailer() { + return curXrefTrailerObj.trailer; + } + + /** * Sets the byte position of the first XRef * (has to be called after very last startxref was read). * This is used to resolve chain of active XRef/trailer. @@ -203,8 +213,37 @@ * {@link #setStartxref(int)} was not called before. * */ public Map getXrefTable() { return ( resolvedXrefTrailer == null ) ? null : resolvedXrefTrailer.xrefTable; } + + /** Returns object numbers which are referenced as contained + * in object stream with specified object number. + * + * This will scan resolved xref table for all entries having negated + * stream object number as value. + * + * @param objstmObjNr object number of object stream for which contained object numbers + * should be returned + * + * @return set of object numbers referenced for given object stream + * or null if {@link #setStartxref(long)} was not + * called before so that no resolved xref table exists + */ + public Set getContainedObjectNumbers( final int objstmObjNr ) + { + if ( resolvedXrefTrailer == null ) + return null; + + final Set refObjNrs = new HashSet(); + final int cmpVal = - objstmObjNr; + + for ( Entry xrefEntry : resolvedXrefTrailer.xrefTable.entrySet() ) { + if ( xrefEntry.getValue() == cmpVal ) + refObjNrs.add( xrefEntry.getKey().getNumber() ); + } + + return refObjNrs; + } } Index: pdfbox/src/main/java/org/apache/pdfbox/pdmodel/PDDocument.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/pdmodel/PDDocument.java (revision 1226367) +++ pdfbox/src/main/java/org/apache/pdfbox/pdmodel/PDDocument.java (working copy) @@ -47,6 +47,7 @@ import org.apache.pdfbox.exceptions.InvalidPasswordException; import org.apache.pdfbox.exceptions.SignatureException; import org.apache.pdfbox.io.RandomAccess; +import org.apache.pdfbox.pdfparser.NonSequentialPDFParser; import org.apache.pdfbox.pdfparser.PDFParser; import org.apache.pdfbox.pdfwriter.COSWriter; import org.apache.pdfbox.pdmodel.common.COSArrayList; @@ -1090,6 +1091,24 @@ return parser.getPDDocument(); } + + /** + * Parses PDF with non sequential parser. + * + * @param file file to be loaded + * @param scratchFile location to store temp PDFBox data for this document + * + * @return loaded document + * + * @throws IOException in case of a file reading or parsing error + */ + public static PDDocument loadNonSeq( File file, RandomAccess scratchFile ) throws IOException + { + NonSequentialPDFParser parser = new NonSequentialPDFParser( file, scratchFile ); + parser.parse(); + return parser.getPDDocument(); + } + /** * This will save this document to the filesystem. * Index: pdfbox/src/main/java/org/apache/pdfbox/io/PushBackInputStream.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/io/PushBackInputStream.java (revision 1226367) +++ pdfbox/src/main/java/org/apache/pdfbox/io/PushBackInputStream.java (working copy) @@ -198,4 +198,21 @@ return data; } + /** + * Signals this class that underlying {@link InputStream} seeked to a new + * position. We set this position and clear push back buffer. + * + * @param newOffset new offset of underlying stream + */ + public void setNewInputStreamOffset( long newOffset ) throws IOException { + + // clear unread buffer by skipping over all bytes of buffer + int unreadLength = buf.length - pos; + if ( unreadLength > 0 ) { + skip( unreadLength ); + } + + offset = newOffset; + } + }