You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@pdfbox.apache.org by le...@apache.org on 2012/05/15 21:30:09 UTC
svn commit: r1338849 [3/3] -
/pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/
Modified: pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFObjectStreamParser.java
URL: http://svn.apache.org/viewvc/pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFObjectStreamParser.java?rev=1338849&r1=1338848&r2=1338849&view=diff
==============================================================================
--- pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFObjectStreamParser.java (original)
+++ pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFObjectStreamParser.java Tue May 15 19:30:09 2012
@@ -40,7 +40,7 @@ public class PDFObjectStreamParser exten
/**
* Log instance.
*/
- private static final Log log =
+ private static final Log LOG =
LogFactory.getLog(PDFObjectStreamParser.class);
private List<COSObject> streamObjects = null;
@@ -53,13 +53,14 @@ public class PDFObjectStreamParser exten
* @since Apache PDFBox 1.3.0
* @param strm The stream to parse.
* @param doc The document for the current parsing.
- * @param forceParcing flag to skip malformed or otherwise unparseable
+ * @param forceParsing flag to skip malformed or otherwise unparseable
* input where possible
* @throws IOException If there is an error initializing the stream.
*/
public PDFObjectStreamParser(
COSStream strm, COSDocument doc, boolean forceParsing)
- throws IOException {
+ throws IOException
+ {
super(strm.getUnfilteredStream(), forceParsing);
setDocument( doc );
stream = strm;
@@ -74,7 +75,8 @@ public class PDFObjectStreamParser exten
* @throws IOException If there is an error initializing the stream.
*/
public PDFObjectStreamParser(COSStream strm, COSDocument doc)
- throws IOException {
+ throws IOException
+ {
this(strm, doc, FORCE_PARSING);
}
@@ -109,9 +111,9 @@ public class PDFObjectStreamParser exten
COSInteger.get( objectNumbers.get( objectCounter).intValue() );
object.setObjectNumber( objNum );
streamObjects.add( object );
- if(log.isDebugEnabled())
+ if(LOG.isDebugEnabled())
{
- log.debug( "parsed=" + object );
+ LOG.debug( "parsed=" + object );
}
objectCounter++;
}
Modified: pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFStreamParser.java
URL: http://svn.apache.org/viewvc/pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFStreamParser.java?rev=1338849&r1=1338848&r2=1338849&view=diff
==============================================================================
--- pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFStreamParser.java (original)
+++ pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFStreamParser.java Tue May 15 19:30:09 2012
@@ -55,13 +55,14 @@ public class PDFStreamParser extends Bas
* @since Apache PDFBox 1.3.0
* @param stream The stream to read data from.
* @param raf The random access file.
- * @param forceParcing flag to skip malformed or otherwise unparseable
+ * @param forceParsing flag to skip malformed or otherwise unparseable
* input where possible
* @throws IOException If there is an error reading from the stream.
*/
public PDFStreamParser(
InputStream stream, RandomAccess raf, boolean forceParsing)
- throws IOException {
+ throws IOException
+ {
super(stream, forceParsing);
file = raf;
}
@@ -75,7 +76,8 @@ public class PDFStreamParser extends Bas
* @throws IOException If there is an error reading from the stream.
*/
public PDFStreamParser(InputStream stream, RandomAccess raf)
- throws IOException {
+ throws IOException
+ {
this(stream, raf, FORCE_PARSING);
}
@@ -96,12 +98,13 @@ public class PDFStreamParser extends Bas
*
* @since Apache PDFBox 1.3.0
* @param stream The stream to parse.
- * @param forceParcing flag to skip malformed or otherwise unparseable
+ * @param forceParsing flag to skip malformed or otherwise unparseable
* input where possible
* @throws IOException If there is an error initializing the stream.
*/
public PDFStreamParser(COSStream stream, boolean forceParsing)
- throws IOException {
+ throws IOException
+ {
this(stream.getUnfilteredStream(), stream.getScratchFile(), forceParsing);
}
@@ -150,6 +153,9 @@ public class PDFStreamParser extends Bas
return streamObjects;
}
+ /**
+ * {@inheritDoc}
+ */
public void close() throws IOException
{
pdfSource.close();
@@ -190,7 +196,8 @@ public class PDFStreamParser extends Bas
}
/** {@inheritDoc} */
- public Object next() {
+ public Object next()
+ {
tryNext();
Object tmp = token;
if (tmp == null)
@@ -474,7 +481,8 @@ public class PDFStreamParser extends Bas
nextChar = pdfSource.peek();
buffer.append( currentChar );
// Type3 Glyph description has operators with a number in the name
- if (currentChar == 'd' && (nextChar == '0' || nextChar == '1') ) {
+ if (currentChar == 'd' && (nextChar == '0' || nextChar == '1') )
+ {
buffer.append( (char)pdfSource.read() );
nextChar = pdfSource.peek();
}
Modified: pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFXrefStreamParser.java
URL: http://svn.apache.org/viewvc/pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFXrefStreamParser.java?rev=1338849&r1=1338848&r2=1338849&view=diff
==============================================================================
--- pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFXrefStreamParser.java (original)
+++ pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/PDFXrefStreamParser.java Tue May 15 19:30:09 2012
@@ -49,19 +49,19 @@ public class PDFXrefStreamParser extends
* @param doc The document for the current parsing.
* @param forceParsing flag to skip malformed or otherwise unparseable
* input where possible
- * @param xrefTrailerResolver resolver to read the xref/trailer information
+ * @param resolver resolver to read the xref/trailer information
*
* @throws IOException If there is an error initializing the stream.
*/
public PDFXrefStreamParser(
COSStream strm, COSDocument doc, boolean forceParsing,
- XrefTrailerResolver xrefTrailerResolver )
+ XrefTrailerResolver resolver )
throws IOException
{
super(strm.getUnfilteredStream(), forceParsing);
setDocument(doc);
stream = strm;
- this.xrefTrailerResolver = xrefTrailerResolver;
+ this.xrefTrailerResolver = resolver;
}
/**
@@ -163,11 +163,11 @@ public class PDFXrefStreamParser extends
* 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);
- }
+ 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;
Modified: pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/VisualSignatureParser.java
URL: http://svn.apache.org/viewvc/pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/VisualSignatureParser.java?rev=1338849&r1=1338848&r2=1338849&view=diff
==============================================================================
--- pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/VisualSignatureParser.java (original)
+++ pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/VisualSignatureParser.java Tue May 15 19:30:09 2012
@@ -29,51 +29,74 @@ import org.apache.pdfbox.cos.COSObject;
import org.apache.pdfbox.pdfwriter.COSWriter;
import org.apache.pdfbox.persistence.util.COSObjectKey;
-public class VisualSignatureParser extends BaseParser {
+public class VisualSignatureParser extends BaseParser
+{
/**
* Log instance.
*/
- private static final Log log = LogFactory.getLog(PDFParser.class);
+ private static final Log LOG = LogFactory.getLog(PDFParser.class);
- public VisualSignatureParser(InputStream input) throws IOException {
+ /**
+ * Constructor.
+ *
+ * @param input the inputstream to be read.
+ *
+ * @throws IOException If something went wrong
+ */
+ public VisualSignatureParser(InputStream input) throws IOException
+ {
super(input);
}
- public void parse() throws IOException {
+ /**
+ * {@inheritDoc}
+ */
+ public void parse() throws IOException
+ {
document = new COSDocument();
skipToNextObj();
boolean wasLastParsedObjectEOF = false;
- try {
- while(!wasLastParsedObjectEOF) {
- if(pdfSource.isEOF()) {
+ try
+ {
+ while(!wasLastParsedObjectEOF)
+ {
+ if(pdfSource.isEOF())
+ {
break;
}
- try {
+ try
+ {
wasLastParsedObjectEOF = parseObject();
- } catch(IOException e) {
+ }
+ catch(IOException e)
+ {
/*
* Warning is sent to the PDFBox.log and to the Console that
* we skipped over an object
*/
- log.warn("Parsing Error, Skipping Object", e);
+ LOG.warn("Parsing Error, Skipping Object", e);
skipToNextObj();
}
skipSpaces();
}
- } catch(IOException e) {
+ }
+ catch(IOException e)
+ {
/*
* PDF files may have random data after the EOF marker. Ignore errors if
* last object processed is EOF.
*/
- if(!wasLastParsedObjectEOF) {
+ if(!wasLastParsedObjectEOF)
+ {
throw e;
}
}
}
- private void skipToNextObj() throws IOException {
+ private void skipToNextObj() throws IOException
+ {
byte[] b = new byte[16];
Pattern p = Pattern.compile("\\d+\\s+\\d+\\s+obj.*", Pattern.DOTALL);
/* Read a buffer of data each time to see if it starts with a
@@ -81,9 +104,11 @@ public class VisualSignatureParser exten
* rarely be needing this function. We could update this to use the
* circular buffer, like in readUntilEndStream().
*/
- while(!pdfSource.isEOF()) {
+ while(!pdfSource.isEOF())
+ {
int l = pdfSource.read(b);
- if(l < 1) {
+ if(l < 1)
+ {
break;
}
String s = new String(b, "US-ASCII");
@@ -91,80 +116,105 @@ public class VisualSignatureParser exten
|| s.startsWith("xref")
|| s.startsWith("startxref")
|| s.startsWith("stream")
- || p.matcher(s).matches()) {
+ || p.matcher(s).matches())
+ {
pdfSource.unread(b);
break;
- } else {
+ }
+ else
+ {
pdfSource.unread(b, 1, l - 1);
}
}
}
- private boolean parseObject() throws IOException {
+ private boolean parseObject() throws IOException
+ {
boolean isEndOfFile = false;
skipSpaces();
//peek at the next character to determine the type of object we are parsing
char peekedChar = (char) pdfSource.peek();
//ignore endobj and endstream sections.
- while(peekedChar == 'e') {
+ while(peekedChar == 'e')
+ {
//there are times when there are multiple endobj, so lets
//just read them and move on.
readString();
skipSpaces();
peekedChar = (char) pdfSource.peek();
}
- if(pdfSource.isEOF()) {
+ if(pdfSource.isEOF())
+ {
// end of file we will return a false and call it a day.
- } else if(peekedChar == 'x') {
+ }
+ else if(peekedChar == 'x')
+ {
//xref table. Note: The contents of the Xref table are currently ignored
return true;
- } else if(peekedChar == 't' || peekedChar == 's') {
+ }
+ else if(peekedChar == 't' || peekedChar == 's')
+ {
// Note: startxref can occur in either a trailer section or by itself
- if(peekedChar == 't') {
+ if(peekedChar == 't')
+ {
return true;
}
- if(peekedChar == 's') {
+ if(peekedChar == 's')
+ {
skipToNextObj();
//verify that EOF exists
String eof = readExpectedString("%%EOF");
- if(eof.indexOf("%%EOF") == -1 && !pdfSource.isEOF()) {
+ if(eof.indexOf("%%EOF") == -1 && !pdfSource.isEOF())
+ {
throw new IOException("expected='%%EOF' actual='" + eof + "' next=" + readString()
+ " next=" + readString());
}
isEndOfFile = true;
}
- } else {
+ }
+ else
+ {
//we are going to parse an normal object
int number = -1;
int genNum = -1;
String objectKey = null;
boolean missingObjectNumber = false;
- try {
+ try
+ {
char peeked = (char) pdfSource.peek();
- if(peeked == '<') {
+ if(peeked == '<')
+ {
missingObjectNumber = true;
- } else {
+ }
+ else
+ {
number = readInt();
}
- } catch(IOException e) {
+ }
+ catch(IOException e)
+ {
//ok for some reason "GNU Ghostscript 5.10" puts two endobj
//statements after an object, of course this is nonsense
//but because we want to support as many PDFs as possible
//we will simply try again
number = readInt();
}
- if(!missingObjectNumber) {
+ if(!missingObjectNumber)
+ {
skipSpaces();
genNum = readInt();
objectKey = readString(3);
//System.out.println( "parseObject() num=" + number +
//" genNumber=" + genNum + " key='" + objectKey + "'" );
- if(!objectKey.equals("obj")) {
+ if(!objectKey.equals("obj"))
+ {
throw new IOException("expected='obj' actual='" + objectKey + "' " + pdfSource);
}
- } else {
+ }
+ else
+ {
number = -1;
genNum = -1;
}
@@ -173,13 +223,17 @@ public class VisualSignatureParser exten
COSBase pb = parseDirObject();
String endObjectKey = readString();
- if(endObjectKey.equals("stream")) {
+ if(endObjectKey.equals("stream"))
+ {
pdfSource.unread(endObjectKey.getBytes());
pdfSource.unread(' ');
- if(pb instanceof COSDictionary) {
+ if(pb instanceof COSDictionary)
+ {
pb = parseCOSStream((COSDictionary) pb, getDocument().getScratchFile());
- } else {
+ }
+ else
+ {
// this is not legal
// the combination of a dict and the stream/endstream forms a complete stream object
throw new IOException("stream not preceded by dictionary");
@@ -192,8 +246,10 @@ public class VisualSignatureParser exten
pb.setNeedToBeUpdate(true);
pdfObject.setObject(pb);
- if(!endObjectKey.equals("endobj")) {
- if(endObjectKey.startsWith("endobj")) {
+ if(!endObjectKey.equals("endobj"))
+ {
+ if(endObjectKey.startsWith("endobj"))
+ {
/*
* Some PDF files don't contain a new line after endobj so we
* need to make sure that the next object number is getting read separately
@@ -201,19 +257,26 @@ public class VisualSignatureParser exten
* instead of "endobj"
*/
pdfSource.unread(endObjectKey.substring(6).getBytes());
- } else if(!pdfSource.isEOF()) {
- try {
+ }
+ else if(!pdfSource.isEOF())
+ {
+ try
+ {
//It is possible that the endobj is missing, there
//are several PDFs out there that do that so skip it and move on.
Float.parseFloat(endObjectKey);
pdfSource.unread(COSWriter.SPACE);
pdfSource.unread(endObjectKey.getBytes());
- } catch(NumberFormatException e) {
+ }
+ catch(NumberFormatException e)
+ {
//we will try again incase there was some garbage which
//some writers will leave behind.
String secondEndObjectKey = readString();
- if(!secondEndObjectKey.equals("endobj")) {
- if(isClosing()) {
+ if(!secondEndObjectKey.equals("endobj"))
+ {
+ if(isClosing())
+ {
//found a case with 17506.pdf object 41 that was like this
//41 0 obj [/Pattern /DeviceGray] ] endobj
//notice the second array close, here we are reading it
@@ -222,7 +285,8 @@ public class VisualSignatureParser exten
}
skipSpaces();
String thirdPossibleEndObj = readString();
- if(!thirdPossibleEndObj.equals("endobj")) {
+ if(!thirdPossibleEndObj.equals("endobj"))
+ {
throw new IOException("expected='endobj' firstReadAttempt='" + endObjectKey + "' "
+ "secondReadAttempt='" + secondEndObjectKey + "' " + pdfSource);
}
@@ -235,8 +299,17 @@ public class VisualSignatureParser exten
return isEndOfFile;
}
- public COSDocument getDocument() throws IOException {
- if(document == null) {
+ /**
+ * Returns the underlying COSDocument.
+ *
+ * @return the COSDocument
+ *
+ * @throws IOException If something went wrong
+ */
+ public COSDocument getDocument() throws IOException
+ {
+ if(document == null)
+ {
throw new IOException("You must call parse() before calling getDocument()");
}
return document;
Modified: pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/XrefTrailerResolver.java
URL: http://svn.apache.org/viewvc/pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/XrefTrailerResolver.java?rev=1338849&r1=1338848&r2=1338849&view=diff
==============================================================================
--- pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/XrefTrailerResolver.java (original)
+++ pdfbox/trunk/pdfbox/src/main/java/org/apache/pdfbox/pdfparser/XrefTrailerResolver.java Tue May 15 19:30:09 2012
@@ -122,9 +122,13 @@ public class XrefTrailerResolver
/**
* Returns the trailer last set by {@link #setTrailer(COSDictionary)}.
+ *
+ * @return the current trailer.
+ *
*/
- public COSDictionary getCurrentTrailer() {
- return curXrefTrailerObj.trailer;
+ public COSDictionary getCurrentTrailer()
+ {
+ return curXrefTrailerObj.trailer;
}
/**
@@ -137,6 +141,9 @@ public class XrefTrailerResolver
* in byte position order.
* Thus for incomplete PDF documents with missing
* startxref one could call this method with parameter value -1.
+ *
+ * @param startxrefBytePosValue starting position of the first XRef
+ *
*/
public void setStartxref( long startxrefBytePosValue )
{
@@ -242,17 +249,20 @@ public class XrefTrailerResolver
*/
public Set<Long> getContainedObjectNumbers( final int objstmObjNr )
{
- if ( resolvedXrefTrailer == null )
- return null;
-
- final Set<Long> refObjNrs = new HashSet<Long>();
- final int cmpVal = - objstmObjNr;
-
- for ( Entry<COSObjectKey,Long> xrefEntry : resolvedXrefTrailer.xrefTable.entrySet() ) {
- if ( xrefEntry.getValue() == cmpVal )
- refObjNrs.add( xrefEntry.getKey().getNumber() );
- }
-
- return refObjNrs;
+ if ( resolvedXrefTrailer == null )
+ {
+ return null;
+ }
+ final Set<Long> refObjNrs = new HashSet<Long>();
+ final int cmpVal = - objstmObjNr;
+
+ for ( Entry<COSObjectKey,Long> xrefEntry : resolvedXrefTrailer.xrefTable.entrySet() )
+ {
+ if ( xrefEntry.getValue() == cmpVal )
+ {
+ refObjNrs.add( xrefEntry.getKey().getNumber() );
+ }
+ }
+ return refObjNrs;
}
}