public XmpDirectory Extract([NotNull] byte[] xmpBytes, int offset, int length)
{
if (offset < 0)
{
throw new ArgumentOutOfRangeException(nameof(offset), "Must be zero or greater.");
}
if (length < 0)
{
throw new ArgumentOutOfRangeException(nameof(length), "Must be zero or greater.");
}
if (xmpBytes.Length < offset + length)
{
throw new ArgumentException("Extends beyond length of byte array.", nameof(length));
}
// Trim any trailing null bytes
// https://github.com/drewnoakes/metadata-extractor-dotnet/issues/154
while (xmpBytes[offset + length - 1] == 0)
{
length--;
}
var directory = new XmpDirectory();
try
{
var xmpMeta = XmpMetaFactory.ParseFromBuffer(xmpBytes, offset, length);
directory.SetXmpMeta(xmpMeta);
}
catch (XmpException e)
{
directory.AddError("Error processing XMP data: " + e.Message);
}
return(directory);
}
private static IReadOnlyList <XmpDirectory> ReadJpegSegments(IEnumerable <JpegSegment> segments)
{
// This is a modified version of XmpReader.ReadJpegSegments
// The original version specifically tests the buffer against the classic XMP preamble and discards everything else.
// Here we also test against the extended XMP preamble and regroup all the extended segments into a single buffer.
// Note that we do not import any of the usual tags as in ProcessXmpTags(). Users of the class have to go
// through XmpMeta methods.
var directories = new List <XmpDirectory>();
byte[] extendedData = null;
int maxPreambleLength = Math.Max(XmpJpegPreamble.Length, XmpJpegExtendedPreamble.Length);
try
{
foreach (var segment in segments)
{
string extractedPreamble = Encoding.UTF8.GetString(segment.Bytes, 0, maxPreambleLength);
if (extractedPreamble.StartsWith(XmpJpegPreamble))
{
ParseNormalXmpSegment(directories, segment.Bytes);
}
else if (extractedPreamble.StartsWith(XmpJpegExtendedPreamble))
{
ParseExtendedXmpSegment(ref extendedData, segment.Bytes);
}
}
}
catch (JpegProcessingException)
{
return(directories);
}
if (extendedData == null)
{
return(directories);
}
// We have collected extended XMP data, let's parse the reconstructed buffer.
XmpDirectory directory = new XmpDirectory();
try
{
var xmpMeta = XmpMetaFactory.ParseFromBuffer(extendedData);
directory.SetXmpMeta(xmpMeta);
}
catch (XmpException e)
{
directory.AddError("Error processing XMP data: " + e.Message);
}
directories.Add(directory);
return(directories);
}
public XmpDirectory Extract([NotNull] byte[] xmpBytes, int offset, int length)
{
var directory = new XmpDirectory();
try
{
var xmpMeta = XmpMetaFactory.ParseFromBuffer(xmpBytes, offset, length);
directory.SetXmpMeta(xmpMeta);
}
catch (XmpException e)
{
directory.AddError("Error processing XMP data: " + e.Message);
}
return(directory);
}
/// <summary>
/// Performs the XMP data extraction.
/// <para />
/// The extraction is done with Adobe's XMPCore library.
/// </summary>
public XmpDirectory Extract([NotNull] byte[] xmpBytes)
{
var directory = new XmpDirectory();
try
{
var xmpMeta = XmpMetaFactory.ParseFromBuffer(xmpBytes);
ProcessXmpTags(directory, xmpMeta);
}
catch (XmpException e)
{
directory.AddError("Error processing XMP data: " + e.Message);
}
return(directory);
}
public XmpDirectory Extract([NotNull] byte[] xmpBytes, int offset, int length)
{
if (offset < 0)
{
throw new ArgumentOutOfRangeException(nameof(offset), "Must be zero or greater.");
}
if (length < 0)
{
throw new ArgumentOutOfRangeException(nameof(length), "Must be zero or greater.");
}
if (xmpBytes.Length < offset + length)
{
throw new ArgumentException("Extends beyond length of byte array.", nameof(length));
}
// Trim any trailing null bytes
// https://github.com/drewnoakes/metadata-extractor-dotnet/issues/154
while (xmpBytes[offset + length - 1] == 0)
{
length--;
}
var directory = new XmpDirectory();
try
{
// Limit photoshop:DocumentAncestors node as it can reach over 100000 items and make parsing extremely slow.
// This is not a typical value but it may happen https://forums.adobe.com/thread/2081839
var parseOptions = new ParseOptions();
parseOptions.SetXMPNodesToLimit(new Dictionary <string, int>()
{
{ "photoshop:DocumentAncestors", 1000 }
});
var xmpMeta = XmpMetaFactory.ParseFromBuffer(xmpBytes, offset, length, parseOptions);
directory.SetXmpMeta(xmpMeta);
}
catch (XmpException e)
{
directory.AddError("Error processing XMP data: " + e.Message);
}
return(directory);
}
private static void ParseNormalXmpSegment(List <XmpDirectory> directories, byte[] segmentBytes)
{
byte[] xmpBytes = new byte[segmentBytes.Length - XmpJpegPreamble.Length];
Array.Copy(segmentBytes, XmpJpegPreamble.Length, xmpBytes, 0, xmpBytes.Length);
XmpDirectory directory = new XmpDirectory();
try
{
var xmpMeta = XmpMetaFactory.ParseFromBuffer(xmpBytes);
directory.SetXmpMeta(xmpMeta);
}
catch (XmpException e)
{
directory.AddError("Error processing XMP data: " + e.Message);
}
directories.Add(directory);
}
private void CompareResults(String orig, String curr)
{
PdfReader cmpReader = new PdfReader(CMP_FOLDER + orig);
PdfReader outReader = new PdfReader(OUT_FOLDER + curr);
byte[] cmpBytes = cmpReader.Metadata, outBytes = outReader.Metadata;
IXmpMeta xmpMeta = XmpMetaFactory.ParseFromBuffer(cmpBytes);
XmpUtils.RemoveProperties(xmpMeta, XmpConst.NS_XMP, XmpBasicProperties.CREATEDATE, true, true);
XmpUtils.RemoveProperties(xmpMeta, XmpConst.NS_XMP, XmpBasicProperties.MODIFYDATE, true, true);
XmpUtils.RemoveProperties(xmpMeta, XmpConst.NS_XMP, XmpBasicProperties.METADATADATE, true, true);
XmpUtils.RemoveProperties(xmpMeta, XmpConst.NS_PDF, PdfProperties.PRODUCER, true, true);
cmpBytes = XmpMetaFactory.SerializeToBuffer(xmpMeta, new SerializeOptions(SerializeOptions.SORT));
xmpMeta = XmpMetaFactory.ParseFromBuffer(outBytes);
XmpUtils.RemoveProperties(xmpMeta, XmpConst.NS_XMP, XmpBasicProperties.CREATEDATE, true, true);
XmpUtils.RemoveProperties(xmpMeta, XmpConst.NS_XMP, XmpBasicProperties.MODIFYDATE, true, true);
XmpUtils.RemoveProperties(xmpMeta, XmpConst.NS_XMP, XmpBasicProperties.METADATADATE, true, true);
XmpUtils.RemoveProperties(xmpMeta, XmpConst.NS_PDF, PdfProperties.PRODUCER, true, true);
outBytes = XmpMetaFactory.SerializeToBuffer(xmpMeta, new SerializeOptions(SerializeOptions.SORT));
XmlDiff xmldiff = new XmlDiff(XmlDiffOptions.None);
if (!xmldiff.Compare(new XmlTextReader(new MemoryStream(cmpBytes)),
new XmlTextReader(new MemoryStream(outBytes))))
{
String currXmlName = curr.Replace(".pdf", ".xml");
FileStream outStream = new FileStream(OUT_FOLDER + "cmp_" + currXmlName, FileMode.Create);
outStream.Write(cmpBytes, 0, cmpBytes.Length);
outStream = new FileStream(OUT_FOLDER + currXmlName, FileMode.Create);
outStream.Write(outBytes, 0, outBytes.Length);
Assert.Fail("The XMP packages are different!");
}
}
