Author: tilman
Date: Sat Dec 20 17:13:48 2025
New Revision: 1930760
Log:
PDFBOX-6130: parse XMP files without processing instructions; improve javadoc;
fix problem with attributes that had no prefix; put prefix into QName (for
debugging); add test
Modified:
pdfbox/branches/2.0/xmpbox/src/main/java/org/apache/xmpbox/xml/DomXmpParser.java
pdfbox/branches/2.0/xmpbox/src/test/java/org/apache/xmpbox/xml/DomXmpParserTest.java
Modified:
pdfbox/branches/2.0/xmpbox/src/main/java/org/apache/xmpbox/xml/DomXmpParser.java
==============================================================================
---
pdfbox/branches/2.0/xmpbox/src/main/java/org/apache/xmpbox/xml/DomXmpParser.java
Sat Dec 20 16:00:25 2025 (r1930759)
+++
pdfbox/branches/2.0/xmpbox/src/main/java/org/apache/xmpbox/xml/DomXmpParser.java
Sat Dec 20 17:13:48 2025 (r1930760)
@@ -109,9 +109,10 @@ public class DomXmpParser
/**
* Enable or disable strict parsing mode.
*
- * @param strictParsing Whether to be strict when parsing XMP: true (the
default) means that
- * malformed XMP will result in an exception, false means that if
malformed content is
- * encountered, the parser will continue its work if possible.
+ * @param strictParsing Whether to be strict or lenient when parsing XMP.
True (the default)
+ * means that malformed XMP will result in an exception, false (lenient)
means that if malformed
+ * content is encountered, the parser will continue its work if possible.
Use strict mode if you
+ * want to work with PDF/A files. Use lenient mode if you care more about
getting metadata.
*/
public void setStrictParsing(boolean strictParsing)
{
@@ -151,7 +152,17 @@ public class DomXmpParser
// expect xpacket processing instruction
if (!(node instanceof ProcessingInstruction))
{
- throw new XmpParsingException(ErrorType.XpacketBadStart, "xmp
should start with a processing instruction");
+ if (strictParsing)
+ {
+ throw new XmpParsingException(ErrorType.XpacketBadStart, "xmp
should start with a processing instruction");
+ }
+ else
+ {
+ xmp =
XMPMetadata.createXMPMetadata(XmpConstants.DEFAULT_XPACKET_BEGIN,
+ XmpConstants.DEFAULT_XPACKET_ID,
+ XmpConstants.DEFAULT_XPACKET_BYTES,
+ XmpConstants.DEFAULT_XPACKET_ENCODING);
+ }
}
else
{
@@ -178,7 +189,14 @@ public class DomXmpParser
// expect xpacket end
if (!(node instanceof ProcessingInstruction))
{
- throw new XmpParsingException(ErrorType.XpacketBadEnd, "xmp should
end with a processing instruction");
+ if (strictParsing)
+ {
+ throw new XmpParsingException(ErrorType.XpacketBadEnd, "xmp
should end with a processing instruction");
+ }
+ else
+ {
+ xmp.setEndXPacket(XmpConstants.DEFAULT_XPACKET_END);
+ }
}
else
{
@@ -191,7 +209,7 @@ public class DomXmpParser
throw new XmpParsingException(ErrorType.XpacketBadEnd,
"xmp should end after xpacket end processing instruction");
}
- // xpacket is OK and the is no more nodes
+ // xpacket is OK and there are no more nodes
// Now, parse the content of root
Element rdfRdf = findDescriptionsParent(root);
nsFinder.push(rdfRdf); // PDFBOX-6099: push namespaces in rdf:RDF
@@ -320,7 +338,7 @@ public class DomXmpParser
{
ComplexPropertyContainer container = schema.getContainer();
PropertyType type = checkPropertyDefinition(xmp,
- new QName(attr.getNamespaceURI(), attr.getLocalName()));
+ new QName(attr.getNamespaceURI(), attr.getLocalName(),
attr.getPrefix()));
if (type == null)
{
@@ -715,8 +733,11 @@ public class DomXmpParser
((XMPSchema) sp).setAboutAsSimple(attr.getValue());
}
}
- else
+ else if (XMLConstants.XML_NS_URI.equals(attr.getNamespaceURI()))
{
+ // This part was the fallback before PDFBOX-6130, now
restricted:
+ // Do not load "ordinary" attributes here because these will
be handled by
+ // tryParseAttributesAsProperties() and
parseDescriptionRootAttr()
Attribute attribute = new Attribute(XMLConstants.XML_NS_URI,
attr.getLocalName(), attr.getValue());
sp.setAttribute(attribute);
}
@@ -921,12 +942,20 @@ public class DomXmpParser
private Element findDescriptionsParent(Element root) throws
XmpParsingException
{
- Element rdfRdf;
+ Element rdfRdf = null;
// check if already rdf element, as xmpmeta wrapper can be optional
if (!XmpConstants.RDF_NAMESPACE.equals(root.getNamespaceURI()))
{
// always <x:xmpmeta xmlns:x="adobe:ns:meta/">
- expectNaming(root, "adobe:ns:meta/", "x", "xmpmeta");
+ if (!strictParsing && "xapmeta".equals(root.getLocalName()))
+ {
+ // older XMP content
+ expectNaming(root, "adobe:ns:meta/", "x", "xapmeta");
+ }
+ else
+ {
+ expectNaming(root, "adobe:ns:meta/", "x", "xmpmeta");
+ }
// should only have one child
NodeList nl = root.getChildNodes();
if (nl.getLength() == 0)
@@ -937,14 +966,25 @@ public class DomXmpParser
else if (nl.getLength() > 1)
{
// only expect one element
- throw new XmpParsingException(ErrorType.Format, "More than one
element found in x:xmpmeta");
+ if (strictParsing)
+ {
+ throw new XmpParsingException(ErrorType.Format, "More than
one element found in x:xmpmeta");
+ }
+ }
+ // find element (there may be a text before the element)
+ for (int i = 0; i < nl.getLength(); ++i)
+ {
+ if (nl.item(i) instanceof Element)
+ {
+ rdfRdf = (Element) nl.item(i);
+ break;
+ }
}
- else if (!(root.getFirstChild() instanceof Element))
+ if (rdfRdf == null)
{
// should be an element
throw new XmpParsingException(ErrorType.Format, "x:xmpmeta
does not contains rdf:RDF element");
} // else let's parse
- rdfRdf = (Element) root.getFirstChild();
}
else
{
Modified:
pdfbox/branches/2.0/xmpbox/src/test/java/org/apache/xmpbox/xml/DomXmpParserTest.java
==============================================================================
---
pdfbox/branches/2.0/xmpbox/src/test/java/org/apache/xmpbox/xml/DomXmpParserTest.java
Sat Dec 20 16:00:25 2025 (r1930759)
+++
pdfbox/branches/2.0/xmpbox/src/test/java/org/apache/xmpbox/xml/DomXmpParserTest.java
Sat Dec 20 17:13:48 2025 (r1930760)
@@ -24,6 +24,7 @@ import java.io.ByteArrayOutputStream;
import java.io.IOException;
import java.io.InputStream;
import java.io.UnsupportedEncodingException;
+import java.nio.charset.StandardCharsets;
import java.util.Calendar;
import java.util.List;
import javax.xml.transform.TransformerException;
@@ -34,6 +35,8 @@ import org.apache.xmpbox.schema.DublinCo
import org.apache.xmpbox.schema.ExifSchema;
import org.apache.xmpbox.schema.PDFAIdentificationSchema;
import org.apache.xmpbox.schema.PhotoshopSchema;
+import org.apache.xmpbox.schema.TiffSchema;
+import org.apache.xmpbox.schema.XMPBasicSchema;
import org.apache.xmpbox.schema.XMPMediaManagementSchema;
import org.apache.xmpbox.schema.XMPSchema;
import org.apache.xmpbox.schema.XMPageTextSchema;
@@ -677,6 +680,10 @@ public class DomXmpParserTest
{
assertEquals("Expecting local name 'xmpmeta' and found 'xapmeta'",
ex.getMessage());
}
+ DomXmpParser xmpParser2 = new DomXmpParser();
+ xmpParser2.setStrictParsing(false);
+ XMPMetadata xmp2 = xmpParser2.parse(s.getBytes("utf-8"));
+ assertEquals(0, xmp2.getAllSchemas().size());
}
@Test
@@ -1327,4 +1334,96 @@ public class DomXmpParserTest
assertEquals("uuid:0b306144-6a43-dcbd-6b3e-c6b6b1df873d",
xmpMediaManagementSchema.getInstanceID());
assertEquals("uuid:0b306144-6a43-dcbd-6b3e-c6b6b1df873d",
xmpMediaManagementSchema.getDocumentID());
}
-}
\ No newline at end of file
+
+ @Test
+ public void testNoProcessingInstruction() throws XmpParsingException,
UnsupportedEncodingException, TransformerException
+ {
+ // From file 000163.pdf
+ // Coastal Services Magazine Volume 11_6 November/December
+ String s = "<x:xmpmeta xmlns:x=\"adobe:ns:meta/\" x:xmptk=\"Adobe XMP
Core 4.1-c037 46.282696, Mon Apr 02 2007 18:36:42 \">\n" +
+ " <rdf:RDF
xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\";>\n" +
+ " <rdf:Description rdf:about=\"\"\n" +
+ " xmlns:xapMM=\"http://ns.adobe.com/xap/1.0/mm/\"\n"; +
+ "
xmlns:stRef=\"http://ns.adobe.com/xap/1.0/sType/ResourceRef#\"\n"; +
+ " xmlns:tiff=\"http://ns.adobe.com/tiff/1.0/\"\n"; +
+ " xmlns:xap=\"http://ns.adobe.com/xap/1.0/\"\n"; +
+ " xmlns:exif=\"http://ns.adobe.com/exif/1.0/\"\n"; +
+ " xmlns:dc=\"http://purl.org/dc/elements/1.1/\"\n"; +
+ " xmlns:photoshop=\"http://ns.adobe.com/photoshop/1.0/\"\n";
+
+ "
xapMM:DocumentID=\"uuid:F1FEDA1D7D03DA11B0F6E4B4E63B0143\"\n" +
+ "
xapMM:InstanceID=\"uuid:7A28FBF56920DA11B4BBB356C0A5C72B\"\n" +
+ " tiff:Orientation=\"1\"\n" +
+ " tiff:XResolution=\"3050000/10000\"\n" +
+ " tiff:YResolution=\"3050000/10000\"\n" +
+ " tiff:ResolutionUnit=\"2\"\n" +
+ " tiff:NativeDigest=\"123456\"\n" +
+ " xap:ModifyDate=\"2005-09-08T09:13:10-04:00\"\n" +
+ " xap:CreatorTool=\"Adobe Photoshop CS2 Windows\"\n" +
+ " xap:CreateDate=\"2005-08-02T13:47:24-04:00\"\n" +
+ " xap:MetadataDate=\"2005-09-08T09:13:10-04:00\"\n" +
+ " exif:ColorSpace=\"-1\"\n" +
+ " exif:PixelXDimension=\"1525\"\n" +
+ " exif:PixelYDimension=\"387\"\n" +
+ " exif:NativeDigest=\"12345678\"\n" +
+ " dc:format=\"image/tiff\"\n" +
+ " photoshop:ColorMode=\"4\"\n" +
+ " photoshop:ICCProfile=\"U.S. Web Coated (SWOP) v2\"\n" +
+ " photoshop:History=\"\">\n" +
+ " <xapMM:DerivedFrom\n" +
+ "
stRef:instanceID=\"adobe:docid:photoshop:28ff3dc5-4801-11d8-85d1-bb49d244e2ef\"\n"
+
+ "
stRef:documentID=\"adobe:docid:photoshop:28ff3dc5-4801-11d8-85d1-bb49d244e2ef\"/>\n"
+
+ " </rdf:Description>\n" +
+ " </rdf:RDF>\n" +
+ "</x:xmpmeta>";
+ try
+ {
+ new DomXmpParser().parse(s.getBytes("utf-8"));
+ fail("XmpParsingException expected");
+ }
+ catch (XmpParsingException ex)
+ {
+ assertEquals("xmp should start with a processing instruction",
ex.getMessage());
+ }
+ DomXmpParser xmpParser2 = new DomXmpParser();
+ xmpParser2.setStrictParsing(false);
+ XMPMetadata xmp2 =
xmpParser2.parse(s.getBytes(StandardCharsets.UTF_8));
+ DublinCoreSchema dublinCoreSchema = xmp2.getDublinCoreSchema();
+ assertEquals("image/tiff", dublinCoreSchema.getFormat());
+ XMPMediaManagementSchema xmpMediaManagementSchema =
xmp2.getXMPMediaManagementSchema();
+ assertEquals("uuid:F1FEDA1D7D03DA11B0F6E4B4E63B0143",
xmpMediaManagementSchema.getDocumentID());
+ TiffSchema tiffSchema = (TiffSchema) xmp2.getSchema(TiffSchema.class);
+ assertEquals("[Orientation=IntegerType:1]",
tiffSchema.getProperty(TiffSchema.ORIENTATION).toString());
+ PhotoshopSchema photoshopSchema = xmp2.getPhotoshopSchema();
+ assertEquals((Integer) 4, photoshopSchema.getColorMode());
+ ExifSchema exifSchema = (ExifSchema) xmp2.getSchema(ExifSchema.class);
+ assertEquals("[PixelXDimension=IntegerType:1525]",
exifSchema.getProperty(ExifSchema.PIXEL_X_DIMENSION).toString());
+ XMPBasicSchema xmpBasicSchema = xmp2.getXMPBasicSchema();
+ assertEquals("Adobe Photoshop CS2 Windows",
xmpBasicSchema.getCreatorTool());
+ XmpSerializer serializer = new XmpSerializer();
+ ByteArrayOutputStream baos = new ByteArrayOutputStream();
+ serializer.serialize(xmp2, baos, true);
+ // check that there are no isolated properties
+ // (Happened before the change at the bottom of loadAttributes())
+ String s2 = baos.toString("utf-8");
+ assertFalse(s2.contains(" ColorMode="));
+ assertFalse(s2.contains(" CreateDate="));
+ assertFalse(s2.contains(" CreatorTool="));
+ assertFalse(s2.contains(" DocumentID="));
+ // now make sure that parsing again still brings the same data
+ DomXmpParser xmpParser3 = new DomXmpParser();
+ xmpParser3.setStrictParsing(false);
+ XMPMetadata xmp3 = xmpParser3.parse(baos.toByteArray());
+ DublinCoreSchema dublinCoreSchema3 = xmp3.getDublinCoreSchema();
+ assertEquals("image/tiff", dublinCoreSchema3.getFormat());
+ XMPMediaManagementSchema xmpMediaManagementSchema3 =
xmp3.getXMPMediaManagementSchema();
+ assertEquals("uuid:F1FEDA1D7D03DA11B0F6E4B4E63B0143",
xmpMediaManagementSchema3.getDocumentID());
+ TiffSchema tiffSchema3 = (TiffSchema) xmp3.getSchema(TiffSchema.class);
+ assertEquals("[Orientation=IntegerType:1]",
tiffSchema3.getProperty(TiffSchema.ORIENTATION).toString());
+ PhotoshopSchema photoshopSchema3 = xmp3.getPhotoshopSchema();
+ assertEquals((Integer) 4, photoshopSchema3.getColorMode());
+ ExifSchema exifSchema3 = (ExifSchema) xmp3.getSchema(ExifSchema.class);
+ assertEquals("[PixelXDimension=IntegerType:1525]",
exifSchema3.getProperty(ExifSchema.PIXEL_X_DIMENSION).toString());
+ XMPBasicSchema xmpBasicSchema3 = xmp3.getXMPBasicSchema();
+ assertEquals("Adobe Photoshop CS2 Windows",
xmpBasicSchema3.getCreatorTool());
+ }
+}
