/// <summary>
/// This method will look for markup annotations on EVERY page
/// </summary>
/// <returns>Count of apparent markup annotations</returns>
private int ScanLinearizedPdf()
{
int CountAnnots = 0;
for (int P = 1; P < MyReader.NumberOfPages; P++)
{
PdfDictionary Next = MyReader.GetPageN(P);
PdfArray annotArray = Next.GetAsArray(PdfName.ANNOTS);
if (annotArray != null)
{
for (int i = 0; i < annotArray.Size - 1; i++)
{
PdfDictionary NextAnnot = annotArray.GetAsDict(i);
if (NextAnnot.Contains(PdfName.POPUP) ||
(NextAnnot.Contains(PdfName.T) && NextAnnot.Contains(PdfName.POPUP)) ||
NextAnnot.Contains(PdfName.INTENT)
)
{
CountAnnots++;
}
}
}
}
return(CountAnnots);
}
public void CopySignedDocuments()
{
string file = RESOURCES + "hello_signed1.pdf";
Directory.CreateDirectory("PdfCopyTest/");
Document pdfDocument = new Document();
PdfCopy copier = new PdfCopy(pdfDocument, new FileStream("PdfCopyTest/CopySignedDocuments.pdf", FileMode.Create));
pdfDocument.Open();
PdfReader reader1 = new PdfReader(file);
copier.AddPage(copier.GetImportedPage(reader1, 1));
copier.FreeReader(reader1);
reader1 = new PdfReader(file);
copier.AddPage(copier.GetImportedPage(reader1, 1));
copier.FreeReader(reader1);
pdfDocument.Close();
PdfReader reader = new PdfReader("PdfCopyTest/CopySignedDocuments.pdf");
PdfDictionary sig = (PdfDictionary)reader.GetPdfObject(9);
PdfDictionary sigRef = sig.GetAsArray(PdfName.REFERENCE).GetAsDict(0);
Assert.True(PdfName.SIGREF.Equals(sigRef.GetAsName(PdfName.TYPE)));
Assert.False(sigRef.Contains(PdfName.DATA));
sig = (PdfDictionary)reader.GetPdfObject(21);
sigRef = sig.GetAsArray(PdfName.REFERENCE).GetAsDict(0);
Assert.True(PdfName.SIGREF.Equals(sigRef.GetAsName(PdfName.TYPE)));
Assert.False(sigRef.Contains(PdfName.DATA));
}
protected override void CheckPdfObject(PdfWriter writer, int key, Object obj1)
{
if (obj1 is PdfNumber)
{
PdfNumber number = (PdfNumber)obj1;
if (Math.Abs(number.DoubleValue) > maxRealValue && number.ToString().Contains("."))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("real.number.is.out.of.range"));
}
}
else if (obj1 is PdfString)
{
PdfString str = (PdfString)obj1;
if (str.GetBytes().Length > maxStringLength)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("pdf.string.is.too.long"));
}
}
else if (obj1 is PdfArray)
{
PdfArray array = (PdfArray)obj1;
if (array.Size > maxArrayLength)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("pdf.array.is.out.of.bounds"));
}
}
else if (obj1 is PdfDictionary)
{
PdfDictionary dictionary = (PdfDictionary)obj1;
if (dictionary.Size > maxDictionaryLength)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("pdf.dictionary.is.out.of.bounds"));
}
if (PdfName.CATALOG.Equals(dictionary.GetAsName(PdfName.TYPE)))
{
if (dictionary.Contains(PdfName.AA))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("the.document.catalog.dictionary.shall.not.include.an.aa.entry"));
}
if (CheckStructure(conformanceLevel))
{
PdfDictionary markInfo = dictionary.GetAsDict(PdfName.MARKINFO);
if (markInfo == null || markInfo.GetAsBoolean(PdfName.MARKED) == null || markInfo.GetAsBoolean(PdfName.MARKED).BooleanValue == false)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("document.catalog.dictionary.shall.include.a.markinfo.dictionary.whose.entry.marked.shall.have.a.value.of.true"));
}
if (!dictionary.Contains(PdfName.LANG))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("document.catalog.dictionary.should.contain.lang.entry"));
}
}
}
}
}
public void PdfDictionaryContainsReturnsFalseIfKeyIsNull()
{
PdfDictionary dictionary = new PdfDictionary();
bool contained = dictionary.Contains(null);
Assert.False(contained);
}
private void CheckAnnotationSize(Stream inputStream, int expectedAnnotationsSize)
{
PdfReader reader = new PdfReader(inputStream);
PdfDictionary pageDictionary = reader.GetPageN(1);
if (pageDictionary.Contains(PdfName.ANNOTS))
{
PdfArray annotations = pageDictionary.GetAsArray(PdfName.ANNOTS);
Assert.True(annotations.Size == expectedAnnotationsSize);
}
}
protected override void CheckGState(PdfWriter writer, int key, Object obj1)
{
if (obj1 is PdfDictionary)
{
PdfDictionary gs = (PdfDictionary)obj1;
PdfObject obj = gs.Get(PdfName.BM);
if (obj != null && !PdfGState.BM_NORMAL.Equals(obj) && !PdfGState.BM_COMPATIBLE.Equals(obj))
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage("blend.mode.1.not.allowed", obj.ToString()));
}
obj = gs.Get(PdfName.CA);
double v = 0.0;
if (obj != null && (v = ((PdfNumber)obj).DoubleValue) != 1.0)
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage("transparency.is.not.allowed.ca.eq.1", v.ToString()));
}
obj = gs.Get(PdfName.ca);
v = 0.0;
if (obj != null && (v = ((PdfNumber)obj).DoubleValue) != 1.0)
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage("transparency.is.not.allowed.ca.eq.1", v.ToString()));
}
if (gs.Contains(PdfName.TR))
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage("an.extgstate.dictionary.shall.not.contain.the.tr.key"));
}
PdfName tr2 = gs.GetAsName(PdfName.TR2);
if (tr2 != null && !tr2.Equals(PdfName.DEFAULT))
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage(
"an.extgstate.dictionary.shall.not.contain.the.TR2.key.with.a.value.other.than.default"));
}
PdfName ri = gs.GetAsName(PdfName.RI);
if (ri != null &&
!(PdfName.RELATIVECOLORIMETRIC.Equals(ri) || PdfName.ABSOLUTECOLORIMETRIC.Equals(ri) ||
PdfName.PERCEPTUAL.Equals(ri) || PdfName.SATURATION.Equals(ri)))
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage("1.value.of.ri.key.is.not.allowed", ri.ToString()));
}
if (gs.Get(PdfName.SMASK) != null && !PdfName.NONE.Equals(gs.GetAsName(PdfName.SMASK)))
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage("the.smask.key.is.not.allowed.in.extgstate"));
}
}
}
/**
* Parses the samples of the image from the underlying content parser, ignoring all filters.
* The parser must be positioned immediately after the ID operator that ends the inline image's dictionary.
* The parser will be left positioned immediately following the EI operator.
* This is primarily useful if no filters have been applied.
* @param imageDictionary the dictionary of the inline image
* @param ps the content parser
* @return the samples of the image
* @throws IOException if anything bad happens during parsing
*/
private static byte[] ParseUnfilteredSamples(PdfDictionary imageDictionary, PdfDictionary colorSpaceDic, PdfContentParser ps)
{
// special case: when no filter is specified, we just read the number of bits
// per component, multiplied by the width and height.
if (imageDictionary.Contains(PdfName.FILTER))
{
throw new ArgumentException("Dictionary contains filters");
}
PdfNumber h = imageDictionary.GetAsNumber(PdfName.HEIGHT);
int bytesToRead = ComputeBytesPerRow(imageDictionary, colorSpaceDic) * h.IntValue;
byte[] bytes = new byte[bytesToRead];
PRTokeniser tokeniser = ps.GetTokeniser();
int shouldBeWhiteSpace = tokeniser.Read(); // skip next character (which better be a whitespace character - I suppose we could check for this)
// from the PDF spec: Unless the image uses ASCIIHexDecode or ASCII85Decode as one of its filters, the ID operator shall be followed by a single white-space character, and the next character shall be interpreted as the first byte of image data.
// unfortunately, we've seen some PDFs where there is no space following the ID, so we have to capture this case and handle it
int startIndex = 0;
if (!PRTokeniser.IsWhitespace(shouldBeWhiteSpace) || shouldBeWhiteSpace == 0) // tokeniser treats 0 as whitespace, but for our purposes, we shouldn't)
{
bytes[0] = (byte)shouldBeWhiteSpace;
startIndex++;
}
for (int i = startIndex; i < bytesToRead; i++)
{
int ch = tokeniser.Read();
if (ch == -1)
{
throw new InlineImageParseException("End of content stream reached before end of image data");
}
bytes[i] = (byte)ch;
}
PdfObject ei = ps.ReadPRObject();
if (!ei.ToString().Equals("EI"))
{
// Some PDF producers seem to add another non-whitespace character after the image data.
// Let's try to handle that case here.
PdfObject ei2 = ps.ReadPRObject();
if (!ei2.ToString().Equals("EI"))
{
throw new InlineImageParseException("EI not found after end of image data");
}
}
return(bytes);
}
protected override void CheckFileSpec(PdfWriter writer, int key, Object obj1)
{
if (obj1 is PdfFileSpecification)
{
PdfDictionary fileSpec = (PdfFileSpecification)obj1;
if (!fileSpec.Contains(PdfName.UF) || !fileSpec.Contains(PdfName.F) ||
!fileSpec.Contains(PdfName.DESC))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("file.specification.dictionary.shall.contain.f.uf.and.desc.entries"));
}
PdfObject obj = fileSpec.Get(PdfName.AFRELATIONSHIP);
if (obj == null || !obj.IsName() || !allowedAFRelationships.Contains(obj as PdfName))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("file.specification.dictionary.shall.contain.correct.afrelationship.key"));
}
if (fileSpec.Contains(PdfName.EF))
{
PdfDictionary dict = GetDirectDictionary(fileSpec.Get(PdfName.EF));
if (dict == null || !dict.Contains(PdfName.F))
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage("ef.key.of.file.specification.dictionary.shall.contain.dictionary.with.valid.f.key"));
}
PdfDictionary embeddedFile = GetDirectDictionary(dict.Get(PdfName.F));
if (embeddedFile == null)
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage("ef.key.of.file.specification.dictionary.shall.contain.dictionary.with.valid.f.key"));
}
CheckEmbeddedFile(embeddedFile);
}
}
}
public static void Test_TraceObjects(string file)
{
Trace.WriteLine($"pdf \"{file}\"");
using (PdfReader pdfReader = new PdfReader(file))
{
int objectCount = pdfReader.XrefSize;
for (int i = 0; i < objectCount; i++)
{
PdfObject obj = pdfReader.GetPdfObject(i);
//if (obj != null)
if (obj is PdfDictionary)
{
PdfDictionary objDic = (PdfDictionary)obj;
string type = null;
if (objDic.Contains(PdfName.TYPE))
{
type = objDic.Get(PdfName.TYPE).ToString();
}
string subtype = null;
if (objDic.Contains(PdfName.SUBTYPE))
{
subtype = objDic.Get(PdfName.SUBTYPE).ToString();
}
Trace.WriteLine($"object {i + 1,3} object type {obj.GetType()} dictionary type {type} subtype {subtype}");
}
else if (obj != null)
{
Trace.WriteLine($"object {i + 1,3} object type {obj.GetType()}");
}
else
{
Trace.WriteLine($"object {i + 1,3} null");
}
}
}
}
public PageInfo(PdfReader reader, int pageNumber)
{
PageNumber = pageNumber;
ImagesInfo = new ImagesInfo();
TextsInfo = new TextsInfo();
_reader = reader;
_page = reader.GetPageN(pageNumber);
PageUnits = _page.Contains(PdfName.USERUNIT) ? _page.GetAsNumber(PdfName.USERUNIT).FloatValue : 72;
Rectangle mediabox = reader.GetPageSize(_page);
PageWidth = Utilities.PointsToMillimeters(mediabox.Width);
PageHeight = Utilities.PointsToMillimeters(mediabox.Height);
_RenderPage(PageNumber, PageUnits);
}
public static int GetImagesCount(PdfReader pdfReader)
{
int objectCount = pdfReader.XrefSize;
int count = 0;
for (int i = 0; i < objectCount; i++)
{
PdfObject obj = pdfReader.GetPdfObject(i);
if (obj is PdfDictionary)
{
PdfDictionary objDic = (PdfDictionary)obj;
if (objDic.Contains(PdfName.TYPE) && objDic.Get(PdfName.TYPE).ToString() == "/XObject" && objDic.Contains(PdfName.SUBTYPE) && objDic.Get(PdfName.SUBTYPE).ToString() == "/Image")
{
count++;
}
}
}
return(count);
}
public static void Test_ExtractImage(string file, int index, string imageFile)
{
Trace.WriteLine($"extract image index {index} from pdf \"{file}\" to \"{imageFile}\"");
if (!zPath.IsPathRooted(imageFile))
{
imageFile = zPath.Combine(zPath.GetDirectoryName(file), imageFile);
}
using (PdfReader pdfReader = new PdfReader(file))
{
PdfObject obj = pdfReader.GetPdfObject(index);
if (!(obj is PdfDictionary))
{
Trace.WriteLine("object is not dictionary");
return;
}
PdfDictionary objDic = (PdfDictionary)obj;
if (!objDic.Contains(PdfName.TYPE) || objDic.Get(PdfName.TYPE).ToString() != "/XObject" || !objDic.Contains(PdfName.SUBTYPE) || objDic.Get(PdfName.SUBTYPE).ToString() != "/Image")
{
Trace.WriteLine("object is not an image");
return;
}
//iTextSharp.text.pdf
//Image
//iTextSharp.text.pdf.PdfImage
//PdfImage pdfImage = new PdfImage();
byte[] bytes = PdfReader.FlateDecode(PdfReader.GetStreamBytesRaw((PRStream)obj), true);
//byte[] bytes = PdfReader.GetStreamBytesRaw((PRStream)obj);
// error : The byte array is not a recognized imageformat
iTextSharp.text.Image image = iTextSharp.text.Image.GetInstance(bytes);
Trace.WriteLine($"width {image.Width} height {image.Height} dpiX {image.DpiX} dpiY {image.DpiY}");
//image.IsContent();
//image.IsImgRaw();
//image.IsJpeg();
//image.OriginalType;
//image.RawData;
//image.Type;
}
}
/**
* Parses the samples of the image from the underlying content parser, ignoring all filters.
* The parser must be positioned immediately after the ID operator that ends the inline image's dictionary.
* The parser will be left positioned immediately following the EI operator.
* This is primarily useful if no filters have been applied.
* @param imageDictionary the dictionary of the inline image
* @param ps the content parser
* @return the samples of the image
* @throws IOException if anything bad happens during parsing
*/
private static byte[] ParseUnfilteredSamples(PdfDictionary imageDictionary, PdfContentParser ps)
{
// special case: when no filter is specified, we just read the number of bits
// per component, multiplied by the width and height.
if (imageDictionary.Contains(PdfName.FILTER))
{
throw new ArgumentException("Dictionary contains filters");
}
PdfNumber h = imageDictionary.GetAsNumber(PdfName.HEIGHT);
int bytesToRead = ComputeBytesPerRow(imageDictionary) * h.IntValue;
byte[] bytes = new byte[bytesToRead];
PRTokeniser tokeniser = ps.GetTokeniser();
tokeniser.Read(); // skip next character (which better be a whitespace character - I suppose we could check for this)
for (int i = 0; i < bytesToRead; i++)
{
int ch = tokeniser.Read();
if (ch == -1)
{
throw new InlineImageParseException("End of content stream reached before end of image data");
}
bytes[i] = (byte)ch;
}
PdfObject ei = ps.ReadPRObject();
if (!ei.ToString().Equals("EI"))
{
throw new InlineImageParseException("EI not found after end of image data");
}
return(bytes);
}
/**
* Extracts locations from the redact annotations contained in the document and applied to the given page.
*/
private IList <PdfCleanUpLocation> ExtractLocationsFromRedactAnnots(int page, PdfDictionary pageDict)
{
List <PdfCleanUpLocation> locations = new List <PdfCleanUpLocation>();
if (pageDict.Contains(PdfName.ANNOTS))
{
PdfArray annotsArray = pageDict.GetAsArray(PdfName.ANNOTS);
for (int i = 0; i < annotsArray.Size; ++i)
{
PdfIndirectReference annotIndirRef = annotsArray.GetAsIndirectObject(i);
PdfDictionary annotDict = annotsArray.GetAsDict(i);
PdfName annotSubtype = annotDict.GetAsName(PdfName.SUBTYPE);
if (annotSubtype.Equals(PdfName.REDACT))
{
SaveRedactAnnotIndirRef(page, annotIndirRef.ToString());
locations.AddRange(ExtractLocationsFromRedactAnnot(page, i, annotDict));
}
}
}
return(locations);
}
public static void Test_TraceImages(string file)
{
Trace.WriteLine($"pdf \"{file}\"");
using (PdfReader pdfReader = new PdfReader(file))
{
int objectCount = pdfReader.XrefSize;
for (int i = 0; i < objectCount; i++)
{
PdfObject obj = pdfReader.GetPdfObject(i);
if (obj is PdfDictionary)
{
PdfDictionary objDic = (PdfDictionary)obj;
if (objDic.Contains(PdfName.TYPE) && objDic.Get(PdfName.TYPE).ToString() == "/XObject" && objDic.Contains(PdfName.SUBTYPE) && objDic.Get(PdfName.SUBTYPE).ToString() == "/Image")
{
string filter = objDic.Get(PdfName.FILTER).ToString();
int width = int.Parse(objDic.Get(PdfName.WIDTH).ToString());
int height = int.Parse(objDic.Get(PdfName.HEIGHT).ToString());
string bpp = objDic.Get(PdfName.BITSPERCOMPONENT).ToString();
Trace.WriteLine($"object {i + 1,3} image width {width} height {height} filter {filter} bits per component {bpp}");
}
}
}
}
}
/**
* Parses the samples of the image from the underlying content parser, accounting for filters
* The parser must be positioned immediately after the ID operator that ends the inline image's dictionary.
* The parser will be left positioned immediately following the EI operator.
* <b>Note:</b>This implementation does not actually apply the filters at this time
* @param imageDictionary the dictionary of the inline image
* @param ps the content parser
* @return the samples of the image
* @throws IOException if anything bad happens during parsing
*/
private static byte[] ParseInlineImageSamples(PdfDictionary imageDictionary, PdfDictionary colorSpaceDic, PdfContentParser ps)
{
// by the time we get to here, we have already parsed the ID operator
if (!imageDictionary.Contains(PdfName.FILTER))
{
return(ParseUnfilteredSamples(imageDictionary, colorSpaceDic, ps));
}
// read all content until we reach an EI operator surrounded by whitespace.
// The following algorithm has two potential issues: what if the image stream
// contains <ws>EI<ws> ?
// Plus, there are some streams that don't have the <ws> before the EI operator
// it sounds like we would have to actually decode the content stream, which
// I'd rather avoid right now.
MemoryStream baos = new MemoryStream();
MemoryStream accumulated = new MemoryStream();
int ch;
int found = 0;
PRTokeniser tokeniser = ps.GetTokeniser();
byte[] ff = null;
while ((ch = tokeniser.Read()) != -1)
{
if (found == 0 && PRTokeniser.IsWhitespace(ch))
{
found++;
accumulated.WriteByte((byte)ch);
}
else if (found == 1 && ch == 'E')
{
found++;
accumulated.WriteByte((byte)ch);
}
else if (found == 1 && PRTokeniser.IsWhitespace(ch))
{
// this clause is needed if we have a white space character that is part of the image data
// followed by a whitespace character that precedes the EI operator. In this case, we need
// to flush the first whitespace, then treat the current whitespace as the first potential
// character for the end of stream check. Note that we don't increment 'found' here.
baos.Write(ff = accumulated.ToArray(), 0, ff.Length);
accumulated.SetLength(0);
accumulated.WriteByte((byte)ch);
}
else if (found == 2 && ch == 'I')
{
found++;
accumulated.WriteByte((byte)ch);
}
else if (found == 3 && PRTokeniser.IsWhitespace(ch))
{
byte[] tmp = baos.ToArray();
if (InlineImageStreamBytesAreComplete(tmp, imageDictionary))
{
return(tmp);
}
byte[] accumulatedArr = accumulated.ToArray();
baos.Write(accumulatedArr, 0, accumulatedArr.Length);
accumulated.SetLength(0);
baos.WriteByte((byte)ch);
found = 0;
}
else
{
baos.Write(ff = accumulated.ToArray(), 0, ff.Length);
accumulated.SetLength(0);
baos.WriteByte((byte)ch);
found = 0;
}
}
throw new InlineImageParseException("Could not find image data or EI");
}
protected override void CheckAnnotation(PdfWriter writer, int key, Object obj1)
{
if (obj1 is PdfFormField)
{
PdfFormField field = (PdfFormField)obj1;
if (!field.Contains(PdfName.SUBTYPE))
{
return;
}
if (field.Contains(PdfName.AA) || field.Contains(PdfName.A))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("widget.annotation.dictionary.or.field.dictionary.shall.not.include.a.or.aa.entry"));
}
}
if (obj1 is PdfAnnotation)
{
PdfAnnotation annot = (PdfAnnotation)obj1;
PdfName subtype = annot.Get(PdfName.SUBTYPE) as PdfName;
if (subtype != null && !allowedAnnotTypes.Contains(subtype))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("annotation.type.1.not.allowed", subtype.ToString()));
}
PdfNumber ca = annot.GetAsNumber(PdfName.CA);
if (ca != null && ca.FloatValue != 1.0)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("an.annotation.dictionary.shall.not.contain.the.ca.key.with.a.value.other.than.1"));
}
PdfNumber f = annot.GetAsNumber(PdfName.F);
if (f == null)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("an.annotation.dictionary.shall.contain.the.f.key"));
}
int flags = f.IntValue;
if (CheckFlag(flags, PdfAnnotation.FLAGS_PRINT) == false || CheckFlag(flags, PdfAnnotation.FLAGS_HIDDEN) ||
CheckFlag(flags, PdfAnnotation.FLAGS_INVISIBLE) || CheckFlag(flags, PdfAnnotation.FLAGS_NOVIEW))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("the.f.keys.print.flag.bit.shall.be.set.to.1.and.its.hidden.invisible.and.noview.flag.bits.shall.be.set.to.0"));
}
if (PdfName.TEXT.Equals(annot.GetAsName(PdfName.SUBTYPE)))
{
if (CheckFlag(flags, PdfAnnotation.FLAGS_NOZOOM) == false || CheckFlag(flags, PdfAnnotation.FLAGS_NOROTATE) == false)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("text.annotations.should.set.the.nozoom.and.norotate.flag.bits.of.the.f.key.to.1"));
}
}
if (annot.Contains(PdfName.C) || annot.Contains(PdfName.IC))
{
ICC_Profile colorProfile = ((PdfAWriter)writer).ColorProfile;
String cs = "";
cs = System.Text.Encoding.ASCII.GetString(colorProfile.Data, 16, 4);
if (!"RGB".Equals(cs.ToUpper()))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("destoutputprofile.in.the.pdfa1.outputintent.dictionary.shall.be.rgb"));
}
}
PdfDictionary ap = GetDirectDictionary(annot.Get(PdfName.AP));
if (ap != null)
{
if (ap.Contains(PdfName.R) || ap.Contains(PdfName.D))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("appearance.dictionary.shall.contain.only.the.n.key.with.stream.value"));
}
PdfObject n = ap.Get(PdfName.N);
if (!(n is PdfIndirectReference))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("appearance.dictionary.shall.contain.only.the.n.key.with.stream.value"));
}
}
if (PdfName.WIDGET.Equals(annot.GetAsName(PdfName.SUBTYPE)) && (annot.Contains(PdfName.AA) || annot.Contains(PdfName.A)))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("widget.annotation.dictionary.or.field.dictionary.shall.not.include.a.or.aa.entry"));
}
if (CheckStructure(conformanceLevel))
{
if (contentAnnotations.Contains(subtype) && !annot.Contains(PdfName.CONTENTS))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("annotation.of.type.1.should.have.contents.key", subtype.ToString()));
}
}
}
}
protected override void CheckPdfObject(PdfWriter writer, int key, Object obj1)
{
if (obj1 is PdfNumber)
{
PdfNumber number = (PdfNumber)obj1;
if (Math.Abs(number.DoubleValue) > maxRealValue && number.ToString().Contains("."))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("real.number.is.out.of.range"));
}
}
else if (obj1 is PdfString)
{
PdfString str = (PdfString)obj1;
if (str.GetBytes().Length > maxStringLength)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("pdf.string.is.too.long"));
}
}
else if (obj1 is PdfArray)
{
PdfArray array = (PdfArray)obj1;
if (array.Size > maxArrayLength)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("pdf.array.is.out.of.bounds"));
}
}
else if (obj1 is PdfDictionary)
{
PdfDictionary dictionary = (PdfDictionary)obj1;
if (dictionary.Size > maxDictionaryLength)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("pdf.dictionary.is.out.of.bounds"));
}
PdfName type = dictionary.GetAsName(PdfName.TYPE);
if (PdfName.CATALOG.Equals(type))
{
if (!dictionary.Contains(PdfName.METADATA))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("the.document.catalog.dictionary.shall.contain.metadata"));
}
if (dictionary.Contains(PdfName.AA))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("the.document.catalog.dictionary.shall.not.include.an.aa.entry"));
}
if (dictionary.Contains(PdfName.NAMES))
{
PdfDictionary names = GetDirectDictionary(dictionary.Get(PdfName.NAMES));
if (names != null && names.Contains(PdfName.EMBEDDEDFILES))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("the.document.catalog.dictionary.shall.not.include.embeddedfiles.names.entry"));
}
}
if (CheckStructure(conformanceLevel))
{
PdfDictionary markInfo = GetDirectDictionary(dictionary.Get(PdfName.MARKINFO));
if (markInfo == null || markInfo.GetAsBoolean(PdfName.MARKED) == null || markInfo.GetAsBoolean(PdfName.MARKED).BooleanValue == false)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("document.catalog.dictionary.shall.include.a.markinfo.dictionary.whose.entry.marked.shall.have.a.value.of.true"));
}
if (!dictionary.Contains(PdfName.LANG))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("document.catalog.dictionary.should.contain.lang.entry"));
}
}
}
else if (PdfName.PAGE.Equals(type))
{
if (dictionary.Contains(PdfName.AA))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("page.dictionary.shall.not.include.aa.entry"));
}
}
else if (PdfName.OUTPUTINTENT.Equals(type))
{
PdfObject destOutputIntent = dictionary.Get(PdfName.DESTOUTPUTPROFILE);
if (destOutputIntent != null && pdfaDestOutputIntent != null)
{
if (pdfaDestOutputIntent.IndRef != destOutputIntent.IndRef)
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage(
"if.outputintents.array.more.than.one.entry.the.same.indirect.object"));
}
}
else
{
pdfaDestOutputIntent = destOutputIntent;
}
PdfName gts = dictionary.GetAsName(PdfName.S);
if (pdfaDestOutputIntent != null)
{
if (PdfName.GTS_PDFA1.Equals(gts))
{
if (pdfaOutputIntentColorSpace != null)
{
throw new PdfAConformanceException(obj1,
MessageLocalization.GetComposedMessage("a.pdfa.file.may.have.only.one.pdfa.outputintent"));
}
pdfaOutputIntentColorSpace = "";
ICC_Profile icc_profile = writer.ColorProfile;
pdfaOutputIntentColorSpace = Encoding.GetEncoding("US-ASCII").GetString(icc_profile.Data, 16, 4);
}
}
else
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("outputintent.shall.have.gtspdfa1.and.destoutputintent"));
}
}
}
}
protected override void CheckPdfObject(PdfWriter writer, int key, Object obj1)
{
if (obj1 is PdfNumber)
{
PdfNumber number = (PdfNumber)obj1;
if (Math.Abs(number.DoubleValue) > maxRealValue && number.ToString().Contains("."))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("real.number.is.out.of.range"));
}
}
else if (obj1 is PdfString)
{
PdfString str = (PdfString)obj1;
if (str.GetBytes().Length > maxStringLength)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("pdf.string.is.too.long"));
}
}
else if (obj1 is PdfArray)
{
PdfArray array = (PdfArray)obj1;
if (array.Size > maxArrayLength)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("pdf.array.is.out.of.bounds"));
}
}
else if (obj1 is PdfDictionary)
{
PdfDictionary dictionary = (PdfDictionary)obj1;
if (dictionary.Size > maxDictionaryLength)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("pdf.dictionary.is.out.of.bounds"));
}
PdfName type = dictionary.GetAsName(PdfName.TYPE);
if (PdfName.CATALOG.Equals(type))
{
if (!dictionary.Contains(PdfName.METADATA))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("the.document.catalog.dictionary.shall.contain.metadata"));
}
if (dictionary.Contains(PdfName.AA))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("the.document.catalog.dictionary.shall.not.include.an.aa.entry"));
}
if (dictionary.Contains(PdfName.NAMES))
{
PdfDictionary names = GetDirectDictionary(dictionary.Get(PdfName.NAMES));
if (names != null && names.Contains(PdfName.EMBEDDEDFILES))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("the.document.catalog.dictionary.shall.not.include.embeddedfiles.names.entry"));
}
}
if (CheckStructure(conformanceLevel))
{
PdfDictionary markInfo = GetDirectDictionary(dictionary.Get(PdfName.MARKINFO));
if (markInfo == null || markInfo.GetAsBoolean(PdfName.MARKED) == null || markInfo.GetAsBoolean(PdfName.MARKED).BooleanValue == false)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("document.catalog.dictionary.shall.include.a.markinfo.dictionary.whose.entry.marked.shall.have.a.value.of.true"));
}
if (!dictionary.Contains(PdfName.LANG))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("document.catalog.dictionary.should.contain.lang.entry"));
}
}
PdfArray outputIntents = GetDirectArray(dictionary.Get(PdfName.OUTPUTINTENTS));
bool pdfa1OutputIntentFound = false;
if (outputIntents != null && outputIntents.Size > 0)
{
for (int i = 0; i < outputIntents.Size; i++)
{
PdfDictionary outputIntentDictionary = GetDirectDictionary(outputIntents[i]);
PdfName gts = outputIntentDictionary.GetAsName(PdfName.S);
if (PdfName.GTS_PDFA1.Equals(gts))
{
if (pdfa1OutputIntentFound)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("a.pdfa.file.may.have.only.one.pdfa.outputintent"));
}
pdfa1OutputIntentFound = true;
}
if (outputIntentDictionary != null)
{
PdfObject destOutputIntent = outputIntentDictionary.Get(PdfName.DESTOUTPUTPROFILE);
if (destOutputIntent == null && PdfName.GTS_PDFA1.Equals(gts))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("outputintent.shall.have.gtspdfa1.and.destoutputintent"));
}
}
}
}
if ((rgbUsed || cmykUsed || grayUsed) && !pdfa1OutputIntentFound)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("if.device.rgb.cmyk.gray.used.in.file.that.file.shall.contain.pdfa.outputintent"));
}
}
else if (PdfName.PAGE.Equals(type))
{
if (dictionary.Contains(PdfName.AA))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("page.dictionary.shall.not.include.aa.entry"));
}
}
else if (PdfName.OUTPUTINTENT.Equals(type))
{
PdfObject iccProfileStream = dictionary.Get(PdfName.DESTOUTPUTPROFILE);
String inputColorSpace = "";
if (iccProfileStream != null)
{
ICC_Profile icc_profile = writer.ColorProfile;
inputColorSpace = Encoding.GetEncoding("US-ASCII").GetString(icc_profile.Data, 16, 4);
}
PdfName gts = dictionary.GetAsName(PdfName.S);
if (!PdfName.GTS_PDFA1.Equals(gts))
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("outputintent.shall.have.gtspdfa1.and.destoutputintent"));
}
if ("RGB ".Equals(inputColorSpace))
{
if (cmykUsed)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("devicecmyk.may.be.used.only.if.the.file.has.a.cmyk.pdfa.outputIntent"));
}
}
else if ("CMYK".Equals(inputColorSpace))
{
if (rgbUsed)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("devicergb.may.be.used.only.if.the.file.has.a.rgb.pdfa.outputIntent"));
}
}
else
{
if (cmykUsed)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("devicecmyk.may.be.used.only.if.the.file.has.a.cmyk.pdfa.outputIntent"));
}
if (rgbUsed)
{
throw new PdfAConformanceException(obj1, MessageLocalization.GetComposedMessage("devicergb.may.be.used.only.if.the.file.has.a.rgb.pdfa.outputIntent"));
}
}
}
}
}
/**
* Parses the samples of the image from the underlying content parser, ignoring all filters.
* The parser must be positioned immediately after the ID operator that ends the inline image's dictionary.
* The parser will be left positioned immediately following the EI operator.
* This is primarily useful if no filters have been applied.
* @param imageDictionary the dictionary of the inline image
* @param ps the content parser
* @return the samples of the image
* @throws IOException if anything bad happens during parsing
*/
private static byte[] ParseUnfilteredSamples(PdfDictionary imageDictionary, PdfDictionary colorSpaceDic, PdfContentParser ps) {
// special case: when no filter is specified, we just read the number of bits
// per component, multiplied by the width and height.
if (imageDictionary.Contains(PdfName.FILTER))
throw new ArgumentException("Dictionary contains filters");
PdfNumber h = imageDictionary.GetAsNumber(PdfName.HEIGHT);
int bytesToRead = ComputeBytesPerRow(imageDictionary, colorSpaceDic) * h.IntValue;
byte[] bytes = new byte[bytesToRead];
PRTokeniser tokeniser = ps.GetTokeniser();
int shouldBeWhiteSpace = tokeniser.Read(); // skip next character (which better be a whitespace character - I suppose we could check for this)
// from the PDF spec: Unless the image uses ASCIIHexDecode or ASCII85Decode as one of its filters, the ID operator shall be followed by a single white-space character, and the next character shall be interpreted as the first byte of image data.
// unfortunately, we've seen some PDFs where there is no space following the ID, so we have to capture this case and handle it
int startIndex = 0;
if (!PRTokeniser.IsWhitespace(shouldBeWhiteSpace) || shouldBeWhiteSpace == 0){ // tokeniser treats 0 as whitespace, but for our purposes, we shouldn't)
bytes[0] = (byte)shouldBeWhiteSpace;
startIndex++;
}
for (int i = startIndex; i < bytesToRead; i++){
int ch = tokeniser.Read();
if (ch == -1)
throw new InlineImageParseException("End of content stream reached before end of image data");
bytes[i] = (byte)ch;
}
PdfObject ei = ps.ReadPRObject();
if (!ei.ToString().Equals("EI"))
throw new InlineImageParseException("EI not found after end of image data");
return bytes;
}
/**
* Parses the samples of the image from the underlying content parser, accounting for filters
* The parser must be positioned immediately after the ID operator that ends the inline image's dictionary.
* The parser will be left positioned immediately following the EI operator.
* <b>Note:</b>This implementation does not actually apply the filters at this time
* @param imageDictionary the dictionary of the inline image
* @param ps the content parser
* @return the samples of the image
* @throws IOException if anything bad happens during parsing
*/
private static byte[] ParseInlineImageSamples(PdfDictionary imageDictionary, PdfDictionary colorSpaceDic, PdfContentParser ps) {
// by the time we get to here, we have already parsed the ID operator
if (!imageDictionary.Contains(PdfName.FILTER)){
return ParseUnfilteredSamples(imageDictionary, colorSpaceDic, ps);
}
// read all content until we reach an EI operator surrounded by whitespace.
// The following algorithm has two potential issues: what if the image stream
// contains <ws>EI<ws> ?
// Plus, there are some streams that don't have the <ws> before the EI operator
// it sounds like we would have to actually decode the content stream, which
// I'd rather avoid right now.
MemoryStream baos = new MemoryStream();
MemoryStream accumulated = new MemoryStream();
int ch;
int found = 0;
PRTokeniser tokeniser = ps.GetTokeniser();
byte[] ff = null;
while ((ch = tokeniser.Read()) != -1){
if (found == 0 && PRTokeniser.IsWhitespace(ch)){
found++;
accumulated.WriteByte((byte)ch);
} else if (found == 1 && ch == 'E'){
found++;
accumulated.WriteByte((byte)ch);
} else if (found == 1 && PRTokeniser.IsWhitespace(ch)){
// this clause is needed if we have a white space character that is part of the image data
// followed by a whitespace character that precedes the EI operator. In this case, we need
// to flush the first whitespace, then treat the current whitespace as the first potential
// character for the end of stream check. Note that we don't increment 'found' here.
baos.Write(ff = accumulated.ToArray(), 0, ff.Length);
accumulated.SetLength(0);
accumulated.WriteByte((byte)ch);
} else if (found == 2 && ch == 'I'){
found++;
accumulated.WriteByte((byte)ch);
} else if (found == 3 && PRTokeniser.IsWhitespace(ch)){
try
{
byte[] tmp = baos.ToArray();
new PdfImageObject(imageDictionary, tmp, colorSpaceDic);
return tmp;
}
catch (Exception)
{
byte[] tmp = accumulated.ToArray();
baos.Write(tmp, 0, tmp.Length);
accumulated.SetLength(0);
baos.WriteByte((byte)ch);
found = 0;
}
} else {
baos.Write(ff = accumulated.ToArray(), 0, ff.Length);
accumulated.SetLength(0);
baos.WriteByte((byte)ch);
found = 0;
}
}
throw new InlineImageParseException("Could not find image data or EI");
}
/**
* Determine if an MCID is available
* @return true if the MCID is available, false otherwise
*/
virtual public bool HasMcid()
{
return(dictionary.Contains(PdfName.MCID));
}
public static void Test_ExtractImages(string file, string imageDirectory)
{
// from http://stackoverflow.com/questions/802269/extract-images-using-itextsharp/804392#804392
Trace.WriteLine($"extract images from pdf \"{file}\" to \"{imageDirectory}\"");
if (!zPath.IsPathRooted(imageDirectory))
{
imageDirectory = zPath.Combine(zPath.GetDirectoryName(file), imageDirectory);
}
using (PdfReader pdfReader = new PdfReader(file))
{
int index = 1;
int objectCount = pdfReader.XrefSize;
for (int i = 0; i < objectCount; i++)
{
PdfObject obj = pdfReader.GetPdfObject(i);
if (obj is PdfDictionary)
{
PdfDictionary objDic = (PdfDictionary)obj;
if (objDic.Contains(PdfName.TYPE) && objDic.Get(PdfName.TYPE).ToString() == "/XObject" && objDic.Contains(PdfName.SUBTYPE) && objDic.Get(PdfName.SUBTYPE).ToString() == "/Image")
{
string filter = objDic.Get(PdfName.FILTER).ToString();
int width = int.Parse(objDic.Get(PdfName.WIDTH).ToString());
int height = int.Parse(objDic.Get(PdfName.HEIGHT).ToString());
string bpp = objDic.Get(PdfName.BITSPERCOMPONENT).ToString();
Trace.WriteLine($"object {i + 1} image width {width} height {height} filter {filter} bits per component {bpp}");
if (filter == "/FlateDecode")
{
byte[] arr = PdfReader.FlateDecode(PdfReader.GetStreamBytesRaw((PRStream)obj), true);
Trace.WriteLine($" bytes count {arr.Length}");
// PixelFormat.Format24bppRgb
// System.Drawing.Imaging.PixelFormat 8 bits
Bitmap bmp = new Bitmap(width, height, PixelFormat.Format8bppIndexed);
// PixelFormat.Format24bppRgb
BitmapData bmpData = bmp.LockBits(new Rectangle(0, 0, width, height), ImageLockMode.WriteOnly, PixelFormat.Format8bppIndexed);
Marshal.Copy(arr, 0, bmpData.Scan0, arr.Length);
bmp.UnlockBits(bmpData);
//bmp.Save(zPath.Combine(imageDirectory, $"image-{index++:000}.jpeg"), ImageFormat.Jpeg);
bmp.Save(zPath.Combine(imageDirectory, $"image-{index++:000}.png"), ImageFormat.Png);
}
}
}
//if (obj != null && obj.IsStream())
//{
// PdfDictionary objDic = (PdfDictionary)obj;
// if (objDic.Contains(PdfName.SUBTYPE) && objDic.Get(PdfName.SUBTYPE).ToString() == "/Image")
// {
// string filter = objDic.Get(PdfName.FILTER).ToString();
// int width = int.Parse(objDic.Get(PdfName.WIDTH).ToString());
// int height = int.Parse(objDic.Get(PdfName.HEIGHT).ToString());
// string bpp = objDic.Get(PdfName.BITSPERCOMPONENT).ToString();
// Trace.WriteLine($"object {i + 1} image width {width} height {height} filter {filter} bits per component {bpp}");
// //string extent = ".";
// //byte[] img = null;
// switch (filter)
// {
// case "/FlateDecode":
// byte[] arr = PdfReader.FlateDecode(PdfReader.GetStreamBytesRaw((PRStream)obj), true);
// Bitmap bmp = new Bitmap(width, height, PixelFormat.Format24bppRgb);
// BitmapData bmpData = bmp.LockBits(new Rectangle(0, 0, width, height), ImageLockMode.WriteOnly, PixelFormat.Format24bppRgb);
// Marshal.Copy(arr, 0, bmpData.Scan0, arr.Length);
// bmp.UnlockBits(bmpData);
// //bmp.Save("c:\\temp\\bmp1.png", ImageFormat.Png);
// bmp.Save(zPath.Combine(imageDirectory, $"image-{index++:000}.jpeg"), ImageFormat.Jpeg);
// break;
// default:
// break;
// }
// }
//}
}
}
}
