/**
* If the child of a structured element is a dictionary, we inspect the
* child; we may also draw a tag.
*
* @param k
* the child dictionary to inspect
*/
public void InspectChildDictionary(PdfDictionary k)
{
if (k == null)
{
return;
}
PdfName s = k.GetAsName(PdfName.S);
if (s != null)
{
String tagN = PdfName.DecodeName(s.ToString());
String tag = FixTagName(tagN);
outp.Write("<");
outp.Write(tag);
outp.Write(">");
PdfDictionary dict = k.GetAsDict(PdfName.PG);
if (dict != null)
{
ParseTag(tagN, k.GetDirectObject(PdfName.K), dict);
}
InspectChild(k.GetDirectObject(PdfName.K));
outp.Write("</");
outp.Write(tag);
outp.WriteLine(">");
}
else
{
InspectChild(k.GetDirectObject(PdfName.K));
}
}
/**
* If the child of a structured element is a dictionary, we inspect the
* child; we may also draw a tag.
*
* @param k
* the child dictionary to inspect
*/
virtual public void InspectChildDictionary(PdfDictionary k, bool inspectAttributes)
{
if (k == null)
{
return;
}
PdfName s = k.GetAsName(PdfName.S);
if (s != null)
{
String tagN = PdfName.DecodeName(s.ToString());
String tag = FixTagName(tagN);
outp.Write("<");
outp.Write(tag);
if (inspectAttributes)
{
PdfDictionary a = k.GetAsDict(PdfName.A);
if (a != null)
{
Dictionary <PdfName, PdfObject> .KeyCollection keys = a.Keys;
foreach (PdfName key in keys)
{
outp.Write(' ');
PdfObject value = a.Get(key);
value = PdfReader.GetPdfObject(value);
outp.Write(XmlName(key));
outp.Write("=\"");
outp.Write(value.ToString());
outp.Write("\"");
}
}
}
outp.Write(">");
PdfObject alt = k.Get(PdfName.ALT);
if (alt != null && alt.ToString() != null)
{
outp.Write("<alt><![CDATA[");
outp.Write(Regex.Replace(alt.ToString(), "[\\000]*", ""));
outp.Write("]]></alt>");
}
PdfDictionary dict = k.GetAsDict(PdfName.PG);
if (dict != null)
{
ParseTag(tagN, k.GetDirectObject(PdfName.K), dict);
}
InspectChild(k.GetDirectObject(PdfName.K));
outp.Write("</");
outp.Write(tag);
outp.WriteLine(">");
}
else
{
InspectChild(k.GetDirectObject(PdfName.K));
}
}
private bool ObjectsIsEquals(PdfDictionary outDict, PdfDictionary cmpDict)
{
foreach (PdfName key in cmpDict.Keys)
{
if (key.CompareTo(PdfName.PARENT) == 0)
{
continue;
}
if (key.CompareTo(PdfName.BASEFONT) == 0 || key.CompareTo(PdfName.FONTNAME) == 0)
{
PdfObject cmpObj = cmpDict.GetDirectObject(key);
if (cmpObj.IsName() && cmpObj.ToString().IndexOf('+') > 0)
{
PdfObject outObj = outDict.GetDirectObject(key);
if (!outObj.IsName())
{
return(false);
}
String cmpName = cmpObj.ToString().Substring(cmpObj.ToString().IndexOf('+'));
String outName = outObj.ToString().Substring(outObj.ToString().IndexOf('+'));
if (!cmpName.Equals(outName))
{
return(false);
}
continue;
}
}
if (!ObjectsIsEquals(outDict.Get(key), cmpDict.Get(key)))
{
return(false);
}
}
return(true);
}
/**
* Shows the detail of a dictionary.
* @param dic the dictionary of which you want the detail
* @param depth the depth of the current dictionary (for nested dictionaries)
* @return a String representation of the dictionary
*/
public static String GetDictionaryDetail(PdfDictionary dic, int depth)
{
StringBuilder builder = new StringBuilder();
builder.Append('(');
IList <PdfName> subDictionaries = new List <PdfName>();
foreach (PdfName key in dic.Keys)
{
PdfObject val = dic.GetDirectObject(key);
if (val.IsDictionary())
{
subDictionaries.Add(key);
}
builder.Append(key);
builder.Append('=');
builder.Append(val);
builder.Append(", ");
}
builder.Length = builder.Length - 2;
builder.Append(')');
foreach (PdfName pdfSubDictionaryName in subDictionaries)
{
builder.Append('\n');
for (int i = 0; i < depth + 1; i++)
{
builder.Append('\t');
}
builder.Append("Subdictionary ");
builder.Append(pdfSubDictionaryName);
builder.Append(" = ");
builder.Append(GetDictionaryDetail(dic.GetAsDict(pdfSubDictionaryName), depth + 1));
}
return(builder.ToString());
}
/// <summary>
/// Updates the pagination in the footer.
/// </summary>
/// <param name="inputPdf">The pdf to modify.</param>
/// <param name="outputPdf">The pdf created with updated pagination.</param>
public static void UpdateFooterPagination(string inputPdf, string outputPdf)
{
PdfReader reader = new PdfReader(inputPdf);
FileStream fs = new FileStream(outputPdf, FileMode.Create, FileAccess.Write);
int n = reader.NumberOfPages;
for (int i = 1; i <= n; i++)
{
PdfDictionary dict = reader.GetPageN(i);
PdfObject obj = dict.GetDirectObject(PdfName.CONTENTS);
if (obj.GetType() == typeof(PRStream))
{
PRStream stream = (PRStream)obj;
byte[] data = PdfReader.GetStreamBytes(stream);
String oldStr = System.Text.Encoding.UTF8.GetString(data);
//Get the string matching the pagination
String pageString = CommonUtils.MatchRegex(oldStr, @"\[\(Seite \)\]TJ.*\[\(");
//Regex replacement of page string with updated page number
String updatedPageString = Regex.Replace(pageString, @"\[\(\d+\)\]", "[(" + i + ")]");
String newString = Regex.Replace(oldStr, @"\[\(Seite \)\]TJ.*\[\(", updatedPageString, RegexOptions.Singleline);
stream.SetData(System.Text.Encoding.UTF8.GetBytes(newString));
}
}
PdfStamper stamper = new PdfStamper(reader, fs);
stamper.Close();
reader.Close();
}
protected void TestXObject(bool shouldExist, int page, String xObjectName)
{
PdfReader reader = null;
RandomAccessFileOrArray raf = null;
raf = new RandomAccessFileOrArray(pdfContent[output]);
reader = new PdfReader(raf, null);
try
{
PdfDictionary dictionary = reader.GetPageN(page);
PdfDictionary resources = (PdfDictionary)dictionary.Get(PdfName.RESOURCES);
PdfDictionary xobject = (PdfDictionary)resources.Get(PdfName.XOBJECT);
PdfObject directXObject = xobject.GetDirectObject(new PdfName(xObjectName));
PdfObject indirectXObject = xobject.Get(new PdfName(xObjectName));
if (shouldExist)
{
Assert.NotNull(indirectXObject);
Assert.NotNull(directXObject);
}
else
{
Assert.IsNull(indirectXObject);
Assert.IsNull(directXObject);
}
}
finally
{
reader.Close();
}
}
/**
* Shows the detail of a dictionary.
* @param dic the dictionary of which you want the detail
* @param depth the depth of the current dictionary (for nested dictionaries)
* @return a String representation of the dictionary
*/
public static String GetDictionaryDetail(PdfDictionary dic, int depth){
StringBuilder builder = new StringBuilder();
builder.Append('(');
IList<PdfName> subDictionaries = new List<PdfName>();
foreach (PdfName key in dic.Keys) {
PdfObject val = dic.GetDirectObject(key);
if (val.IsDictionary())
subDictionaries.Add(key);
builder.Append(key);
builder.Append('=');
builder.Append(val);
builder.Append(", ");
}
builder.Length = builder.Length-2;
builder.Append(')');
foreach (PdfName pdfSubDictionaryName in subDictionaries) {
builder.Append('\n');
for (int i = 0; i < depth+1; i++){
builder.Append('\t');
}
builder.Append("Subdictionary ");
builder.Append(pdfSubDictionaryName);
builder.Append(" = ");
builder.Append(GetDictionaryDetail(dic.GetAsDict(pdfSubDictionaryName), depth+1));
}
return builder.ToString();
}
/**
* In case an object is an array, a dictionary or a stream,
* we need to loop over the entries and process them one by one.
* @param object the object to examine
*/
protected void LoopOver(PdfObject @object)
{
switch (@object.Type)
{
case PdfObject.ARRAY:
PdfArray array = (PdfArray)@object;
for (int i = 0; i < array.Size; i++)
{
Process(array.GetDirectObject(i));
}
break;
case PdfObject.DICTIONARY:
case PdfObject.STREAM:
PdfDictionary dict = (PdfDictionary)@object;
if (dict.IsPages())
{
break;
}
foreach (PdfName name in dict.Keys)
{
Process(dict.GetDirectObject(name));
}
break;
}
}
/**
* Looks at a StructElem dictionary, and processes it.
* @param dict the StructElem dictionary that needs to be examined
* @param ref the reference to the StructElem dictionary
* @throws DocumentException
*/
protected void ProcessStructElems(PdfDictionary structElem, PdfIndirectReference refa)
{
LOGGER.Info(String.Format("addStructureItems({0}, {1})", structElem, refa));
if (structElem == null)
{
return;
}
ProcessStructElemKids(structElem, refa, structElem.GetDirectObject(PdfName.K));
}
/**
* Creates a StructureObject for an OBJR dictionary.
* @param structElem the parent structure element
* @param ref the reference of the parent structure element
* @param dict the object reference dictionary
*/
public StructureObject(PdfDictionary structElem, PdfIndirectReference refa, PdfDictionary dict) {
this.structElem = structElem;
this.refa = refa;
this.obj = dict.GetDirectObject(PdfName.OBJ);
this.objref = dict.GetAsIndirectObject(PdfName.OBJ);
this.structParent = ((PdfDictionary) obj).GetAsNumber(PdfName.STRUCTPARENT).IntValue;
PdfIndirectReference pg = dict.GetAsIndirectObject(PdfName.PG);
if (pg == null)
pg = structElem.GetAsIndirectObject(PdfName.PG);
this.pageref = pg.Number;
}
public byte[] GetImageAsBytes()
{
if (streamBytes == null)
{
return(null);
}
if (!decoded)
{
PdfName filter = dictionary.GetAsName(PdfName.FILTER);
if (PdfName.DCTDECODE.Equals(filter))
{
fileType = TYPE_JPG;
return(streamBytes);
}
else if (PdfName.JPXDECODE.Equals(filter))
{
fileType = TYPE_JP2;
return(streamBytes);
}
return(null);
}
pngColorType = -1;
width = dictionary.GetAsNumber(PdfName.WIDTH).IntValue;
height = dictionary.GetAsNumber(PdfName.HEIGHT).IntValue;
bpc = dictionary.GetAsNumber(PdfName.BITSPERCOMPONENT).IntValue;
pngBitDepth = bpc;
PdfObject colorspace = dictionary.GetDirectObject(PdfName.COLORSPACE);
palette = null;
icc = null;
stride = 0;
FindColorspace(colorspace, true);
if (pngColorType < 0)
{
return(null);
}
MemoryStream ms = new MemoryStream();
PngWriter png = new PngWriter(ms);
png.WriteHeader(width, height, pngBitDepth, pngColorType);
if (icc != null)
{
png.WriteIccProfile(icc);
}
if (palette != null)
{
png.WritePalette(palette);
}
png.WriteData(streamBytes, stride);
png.WriteEnd();
fileType = TYPE_PNG;
return(ms.ToArray());
}
/**
* Parses a string with structured content.
*
* @param reader
* the PdfReader that has access to the PDF file
* @param os
* the Stream to which the resulting xml will be written
* @param charset
* the charset to encode the data
* @since 5.0.5
*/
public void ConvertToXml(PdfReader reader, Stream os, Encoding encoding)
{
this.reader = reader;
outp = new StreamWriter(os, encoding);
// get the StructTreeRoot from the root obj
PdfDictionary catalog = reader.Catalog;
PdfDictionary struc = catalog.GetAsDict(PdfName.STRUCTTREEROOT);
// Inspect the child or children of the StructTreeRoot
InspectChild(struc.GetDirectObject(PdfName.K));
outp.Flush();
outp.Close();
}
/**
* Creates a StructureObject for an OBJR dictionary.
* @param structElem the parent structure element
* @param ref the reference of the parent structure element
* @param dict the object reference dictionary
*/
public StructureObject(PdfDictionary structElem, PdfIndirectReference refa, PdfDictionary dict)
{
this.structElem = structElem;
this.refa = refa;
this.obj = dict.GetDirectObject(PdfName.OBJ);
this.objref = dict.GetAsIndirectObject(PdfName.OBJ);
this.structParent = ((PdfDictionary)obj).GetAsNumber(PdfName.STRUCTPARENT).IntValue;
PdfIndirectReference pg = dict.GetAsIndirectObject(PdfName.PG);
if (pg == null)
{
pg = structElem.GetAsIndirectObject(PdfName.PG);
}
this.pageref = pg.Number;
}
public static void Test(string inputFile, string outputFile)
{
Console.WriteLine("Test");
String formFile = inputFile;
String newFile = outputFile;
PdfReader reader = new PdfReader(formFile);
PdfDictionary dict = reader.GetPageN(1);
foreach (var v in dict)
{
Console.WriteLine("{0}", v);
}
PdfObject obj = dict.GetDirectObject(PdfName.CONTENTS);
if (obj is PRStream stream)
{
byte[] data = PdfReader.GetStreamBytes(stream);
string dd = System.Text.Encoding.ASCII.GetString(data);
Console.WriteLine(dd);
stream.SetData(Encoding.ASCII.GetBytes(dd));
}
//PdfStamper stamper = new PdfStamper(reader, new FileStream(newFile, FileMode.Create));
//AcroFields fields = reader.AcroFields; // stamper.AcroFields;
//foreach(var v in fields.Fields)
//{
// Console.WriteLine("{0}", v);
//}
// set form fields
//fields.SetField("{TO}", "John Doe");
//fields.SetField("{FROM}", "2 Milky Way, London");
//stamper.FormFlattening = true;
//stamper.Close();
reader.Close();
}
public override PdfObject GetDirectObject(PdfName key)
{
for (int i = resourcesStack.Count - 1; i >= 0; i--)
{
PdfDictionary subResource = resourcesStack[i];
if (subResource != null)
{
PdfObject obj = subResource.GetDirectObject(key);
if (obj != null)
{
return(obj);
}
}
}
return(base.GetDirectObject(key)); // shouldn't be necessary, but just in case we've done something crazy
}
/**
* Parses a string with structured content.
*
* @param reader
* the PdfReader that has access to the PDF file
* @param os
* the Stream to which the resulting xml will be written
* @param charset
* the charset to encode the data
* @since 5.0.5
*/
public virtual void ConvertToXml(PdfReader reader, Stream os, Encoding encoding)
{
this.reader = reader;
outp = new StreamWriter(os, encoding);
// get the StructTreeRoot from the root obj
PdfDictionary catalog = reader.Catalog;
PdfDictionary struc = catalog.GetAsDict(PdfName.STRUCTTREEROOT);
if (struc == null)
{
throw new IOException(MessageLocalization.GetComposedMessage("no.structtreeroot.found"));
}
// Inspect the child or children of the StructTreeRoot
InspectChild(struc.GetDirectObject(PdfName.K));
outp.Flush();
outp.Close();
}
public override PdfObject GetDirectObject(PdfName key)
{
for (int index = _resourceStack.Count - 1; index >= 0; index--)
{
PdfDictionary subResource = _resourceStack[index];
if (subResource != null)
{
PdfObject obj = subResource.GetDirectObject(key);
if (obj != null)
{
return(obj);
}
}
}
return(base.GetDirectObject(key));
}
public static void FixPageNumberOnPage(String src, String dest, int pageIndex, string newPageLabel)
{
using (PdfReader reader = new PdfReader(src))
{
PdfDictionary dict = reader.GetPageN(pageIndex);
PdfObject o = dict.GetDirectObject(PdfName.CONTENTS);
var streams = SearchAllStreams(reader, o);
foreach (var stream in streams)
{
FixPageNumberOnStream(stream as PRStream, pageIndex.ToString(), newPageLabel);
}
using (PdfStamper stamper = new PdfStamper(reader, new FileStream(dest, FileMode.Create)))
stamper.Close();
reader.Close();
}
}
public override void ConvertToXml(PdfReader reader, Stream os, Encoding encoding)
{
this.reader = reader;
outp = new StreamWriter(os, encoding);
outp.Write("<root>");
// get the StructTreeRoot from the root object
PdfDictionary catalog = reader.Catalog;
PdfDictionary str = catalog.GetAsDict(PdfName.STRUCTTREEROOT);
if (str == null)
{
throw new IOException("No StructTreeRoot");
}
// Inspect the child or children of the StructTreeRoot
InspectChild(str.GetDirectObject(PdfName.K));
outp.Write("</root>");
outp.Close();
}
/**
* Searches for a tag in a page.
*
* @param tag
* the name of the tag
* @param obj
* an identifier to find the marked content
* @param page
* a page dictionary
* @throws IOException
*/
public void ParseTag(String tag, PdfObject obj, PdfDictionary page)
{
PRStream stream = (PRStream)page.GetAsStream(PdfName.CONTENTS);
// if the identifier is a number, we can extract the content right away
if (obj is PdfNumber)
{
PdfNumber mcid = (PdfNumber)obj;
RenderFilter filter = new MarkedContentRenderFilter(mcid.IntValue);
ITextExtractionStrategy strategy = new SimpleTextExtractionStrategy();
FilteredTextRenderListener listener = new FilteredTextRenderListener(strategy, new RenderFilter[] { filter });
PdfContentStreamProcessor processor = new PdfContentStreamProcessor(
listener);
processor.ProcessContent(PdfReader.GetStreamBytes(stream), page
.GetAsDict(PdfName.RESOURCES));
outp.Write(SimpleXMLParser.EscapeXML(listener.GetResultantText(), true));
}
// if the identifier is an array, we call the parseTag method
// recursively
else if (obj is PdfArray)
{
PdfArray arr = (PdfArray)obj;
int n = arr.Size;
for (int i = 0; i < n; i++)
{
ParseTag(tag, arr[i], page);
if (i < n - 1)
{
outp.WriteLine();
}
}
}
// if the identifier is a dictionary, we get the resources from the
// dictionary
else if (obj is PdfDictionary)
{
PdfDictionary mcr = (PdfDictionary)obj;
ParseTag(tag, mcr.GetDirectObject(PdfName.MCID), mcr
.GetAsDict(PdfName.PG));
}
}
/**
* Creates a StructureObject for an OBJR dictionary.
* @param structElem the parent structure element
* @param ref the reference of the parent structure element
* @param dict the object reference dictionary
*/
public StructureObject(PdfDictionary structElem, PdfIndirectReference refa, PdfDictionary dict)
{
this.structElem = structElem;
this.refa = refa;
this.obj = dict.GetDirectObject(PdfName.OBJ);
this.objref = dict.GetAsIndirectObject(PdfName.OBJ);
PdfNumber sp = ((PdfDictionary)obj).GetAsNumber(PdfName.STRUCTPARENT);
if (sp == null)
{
throw new InvalidPdfException("structparentid.not.found");
}
this.structParent = sp.IntValue;
PdfIndirectReference pg = dict.GetAsIndirectObject(PdfName.PG);
if (pg == null)
{
pg = structElem.GetAsIndirectObject(PdfName.PG);
}
this.pageref = pg.Number;
}
private List <PdfObject> Do(PdfLiteral oper, List <PdfObject> operands)
{
PdfName xobjectName = (PdfName)operands[0];
//System.Diagnostics.Debug.WriteLine("Opr: " + xobjectName.ToString());
PdfDictionary xobjects = _modifier.ResourceDictionary.GetAsDict(PdfName.XOBJECT);
PdfObject po = xobjects.Get(xobjectName);
int n = (po as PdfIndirectReference).Number;
if (_convertedIndirectList.Contains(n))
{
return(operands);
}
else
{
_convertedIndirectList.Add(n);
}
PdfObject xobject = xobjects.GetDirectObject(xobjectName);
if (xobject.IsStream())
{
PdfStream xobjectStream = (PdfStream)xobject;
PdfName subType = xobjectStream.GetAsName(PdfName.SUBTYPE);
if (subType == PdfName.FORM)
{
this.Do_Form(xobjectStream);
}
else if (subType == PdfName.IMAGE)
{
this.Do_Image(xobjectStream);
}
}
return(operands);
}
/**
* Displays an XObject using the registered handler for this XObject's subtype
* @param xobjectName the name of the XObject to retrieve from the resource dictionary
*/
private void DisplayXObject(PdfName xobjectName)
{
PdfDictionary xobjects = resources.GetAsDict(PdfName.XOBJECT);
PdfObject xobject = xobjects.GetDirectObject(xobjectName);
PdfStream xobjectStream = (PdfStream)xobject;
PdfName subType = xobjectStream.GetAsName(PdfName.SUBTYPE);
if (xobject.IsStream())
{
IXObjectDoHandler handler;
xobjectDoHandlers.TryGetValue(subType, out handler);
if (handler == null)
{
handler = xobjectDoHandlers[PdfName.DEFAULT];
}
handler.HandleXObject(this, xobjectStream, xobjects.GetAsIndirectObject(xobjectName));
}
else
{
throw new InvalidOperationException(MessageLocalization.GetComposedMessage("XObject.1.is.not.a.stream", xobjectName));
}
}
/// <summary>
/// Replace texts.
/// TODO: Los textos con formato creado en word introducen caracteres entre las palabras y no lo localiza.
/// </summary>
public void Replace()
{
using (FileStream stream = new FileStream(_pathSource, FileMode.Open))
{
using (PdfReader pdfReader = new PdfReader(stream))
{
for (int x = 1; x <= pdfReader.NumberOfPages; x++)
{
using (FileStream streamDest = new FileStream(_pathDest, FileMode.Create))
{
PdfDictionary dict = pdfReader.GetPageN(x);
PdfObject obj = dict.GetDirectObject(PdfName.CONTENTS);
if (obj.GetType() == typeof(PRStream))
{
ReplacePRStream(obj);
}
if (obj.GetType() == typeof(PdfArray))
{
foreach (var r in (PdfArray)obj)
{
PRIndirectReference ir = (PRIndirectReference)r;
PdfObject refdObj = pdfReader.GetPdfObject(ir.Number);
if (refdObj.IsStream())
{
ReplacePRStream(refdObj);
}
}
}
using (PdfStamper stamper = new PdfStamper(pdfReader, streamDest))
{
}
}
}
}
}
}
/**
* Creates a list of StructuredItem objects.
* @param reader the reader holding the PDF to examine
*/
public StructureItems(PdfReader reader)
{
PdfDictionary catalog = reader.Catalog;
structTreeRoot = catalog.GetAsDict(PdfName.STRUCTTREEROOT);
if (structTreeRoot == null)
{
throw new DocumentException(MessageLocalization.GetComposedMessage("can.t.read.document.structure"));
}
// Storing the parent tree
parentTree = PdfNumberTree.ReadTree(structTreeRoot.GetAsDict(PdfName.PARENTTREE));
structTreeRoot.Remove(PdfName.STRUCTPARENTS);
// Examining the StructTreeRoot
PdfObject objecta = structTreeRoot.GetDirectObject(PdfName.K);
if (objecta == null)
{
return;
}
switch (objecta.Type)
{
case PdfObject.DICTIONARY:
LOGGER.Info("StructTreeRoot refers to dictionary");
ProcessStructElems((PdfDictionary)objecta, structTreeRoot.GetAsIndirectObject(PdfName.K));
break;
case PdfObject.ARRAY:
LOGGER.Info("StructTreeRoot refers to array");
PdfArray array = (PdfArray)objecta;
for (int i = 0; i < array.Size; i++)
{
ProcessStructElems(array.GetAsDict(i), array.GetAsIndirectObject(i));
}
break;
}
}
/**
* If the child of a structured element is a dictionary, we inspect the
* child; we may also draw a tag.
*
* @param k
* the child dictionary to inspect
*/
virtual public void InspectChildDictionary(PdfDictionary k, bool inspectAttributes) {
if (k == null)
return;
PdfName s = k.GetAsName(PdfName.S);
if (s != null) {
String tagN = PdfName.DecodeName(s.ToString());
String tag = FixTagName(tagN);
outp.Write("<");
outp.Write(tag);
if (inspectAttributes) {
PdfDictionary a = k.GetAsDict(PdfName.A);
if (a != null) {
Dictionary<PdfName, PdfObject>.KeyCollection keys = a.Keys;
foreach (PdfName key in keys) {
outp.Write(' ');
PdfObject value = a.Get(key);
value = PdfReader.GetPdfObject(value);
outp.Write(XmlName(key));
outp.Write("=\"");
outp.Write(value.ToString());
outp.Write("\"");
}
}
}
outp.Write(">");
PdfDictionary dict = k.GetAsDict(PdfName.PG);
if (dict != null)
ParseTag(tagN, k.GetDirectObject(PdfName.K), dict);
InspectChild(k.GetDirectObject(PdfName.K));
outp.Write("</");
outp.Write(tag);
outp.WriteLine(">");
} else
InspectChild(k.GetDirectObject(PdfName.K));
}
/**
* decodes the bytes currently captured in the streamBytes and replaces it with an image representation of the bytes
* (this will either be a png or a tiff, depending on the color depth of the image)
* @throws IOException
*/
private void DecodeImageBytes()
{
if (streamContentType != null)
{
throw new ArgumentException(MessageLocalization.GetComposedMessage("Decoding.can't.happen.on.this.type.of.stream.(.1.)", streamContentType.FileExtension));
}
pngColorType = -1;
PdfArray decode = dictionary.GetAsArray(PdfName.DECODE);
width = dictionary.GetAsNumber(PdfName.WIDTH).IntValue;
height = dictionary.GetAsNumber(PdfName.HEIGHT).IntValue;
bpc = dictionary.GetAsNumber(PdfName.BITSPERCOMPONENT).IntValue;
pngBitDepth = bpc;
PdfObject colorspace = dictionary.GetDirectObject(PdfName.COLORSPACE);
if (colorspace is PdfName && colorSpaceDic != null)
{
PdfObject csLookup = colorSpaceDic.GetDirectObject((PdfName)colorspace);
if (csLookup != null)
{
colorspace = csLookup;
}
}
palette = null;
icc = null;
stride = 0;
FindColorspace(colorspace, true);
MemoryStream ms = new MemoryStream();
if (pngColorType < 0)
{
if (bpc != 8)
{
throw new UnsupportedPdfException(MessageLocalization.GetComposedMessage("the.color.depth.1.is.not.supported", bpc));
}
if (PdfName.DEVICECMYK.Equals(colorspace))
{
}
else if (colorspace is PdfArray)
{
PdfArray ca = (PdfArray)colorspace;
PdfObject tyca = ca.GetDirectObject(0);
if (!PdfName.ICCBASED.Equals(tyca))
{
throw new UnsupportedPdfException(MessageLocalization.GetComposedMessage("the.color.space.1.is.not.supported", colorspace));
}
PRStream pr = (PRStream)ca.GetDirectObject(1);
int n = pr.GetAsNumber(PdfName.N).IntValue;
if (n != 4)
{
throw new UnsupportedPdfException(MessageLocalization.GetComposedMessage("N.value.1.is.not.supported", n));
}
icc = PdfReader.GetStreamBytes(pr);
}
else
{
throw new UnsupportedPdfException(MessageLocalization.GetComposedMessage("the.color.space.1.is.not.supported", colorspace));
}
stride = 4 * width;
TiffWriter wr = new TiffWriter();
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_SAMPLESPERPIXEL, 4));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_BITSPERSAMPLE, new int[] { 8, 8, 8, 8 }));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_PHOTOMETRIC, TIFFConstants.PHOTOMETRIC_SEPARATED));
wr.AddField(new TiffWriter.FieldLong(TIFFConstants.TIFFTAG_IMAGEWIDTH, width));
wr.AddField(new TiffWriter.FieldLong(TIFFConstants.TIFFTAG_IMAGELENGTH, height));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_COMPRESSION, TIFFConstants.COMPRESSION_LZW));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_PREDICTOR, TIFFConstants.PREDICTOR_HORIZONTAL_DIFFERENCING));
wr.AddField(new TiffWriter.FieldLong(TIFFConstants.TIFFTAG_ROWSPERSTRIP, height));
wr.AddField(new TiffWriter.FieldRational(TIFFConstants.TIFFTAG_XRESOLUTION, new int[] { 300, 1 }));
wr.AddField(new TiffWriter.FieldRational(TIFFConstants.TIFFTAG_YRESOLUTION, new int[] { 300, 1 }));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_RESOLUTIONUNIT, TIFFConstants.RESUNIT_INCH));
wr.AddField(new TiffWriter.FieldAscii(TIFFConstants.TIFFTAG_SOFTWARE, Version.GetInstance().GetVersion));
MemoryStream comp = new MemoryStream();
TiffWriter.CompressLZW(comp, 2, imageBytes, height, 4, stride);
byte[] buf = comp.ToArray();
wr.AddField(new TiffWriter.FieldImage(buf));
wr.AddField(new TiffWriter.FieldLong(TIFFConstants.TIFFTAG_STRIPBYTECOUNTS, buf.Length));
if (icc != null)
{
wr.AddField(new TiffWriter.FieldUndefined(TIFFConstants.TIFFTAG_ICCPROFILE, icc));
}
wr.WriteFile(ms);
streamContentType = ImageBytesType.CCITT;
imageBytes = ms.ToArray();
return;
}
else
{
PngWriter png = new PngWriter(ms);
if (decode != null)
{
if (pngBitDepth == 1)
{
// if the decode array is 1,0, then we need to invert the image
if (decode.GetAsNumber(0).IntValue == 1 && decode.GetAsNumber(1).IntValue == 0)
{
int len = imageBytes.Length;
for (int t = 0; t < len; ++t)
{
imageBytes[t] ^= 0xff;
}
}
else
{
// if the decode array is 0,1, do nothing. It's possible that the array could be 0,0 or 1,1 - but that would be silly, so we'll just ignore that case
}
}
else
{
// todo: add decode transformation for other depths
}
}
png.WriteHeader(width, height, pngBitDepth, pngColorType);
if (icc != null)
{
png.WriteIccProfile(icc);
}
if (palette != null)
{
png.WritePalette(palette);
}
png.WriteData(imageBytes, stride);
png.WriteEnd();
streamContentType = ImageBytesType.PNG;
imageBytes = ms.ToArray();
}
}
public static MemoryStream RemoveText(MemoryStream streamInput, string watermarkText)
{
string content;
PRStream stream;
PdfArray contentarray;
PdfReader reader = new PdfReader(streamInput.ToArray());
string DEST = "D:\\WebMes Projects\\NewYueWebMES\\YueWebMES.Service\\WebMES.API\\Attachments\\fa1.pdf";
//reader.RemoveUnusedObjects();
for (var i = 1; i <= reader.NumberOfPages; i++)
{
PdfDictionary dictionary = reader.GetPageN(i);//获取页
try
{
PdfObject pdfObj = dictionary.GetDirectObject(PdfName.CONTENTS);
if (pdfObj.GetType() == typeof(PRStream))
{
PRStream pst = (PRStream)pdfObj;
byte[] data = PdfReader.GetStreamBytes(pst);
content = System.Text.Encoding.ASCII.GetString(data);//获取pdf页内的文字内容
content = content.Replace(watermarkText, "");
pst.SetData(System.Text.Encoding.Default.GetBytes(content));
}
}
catch (Exception ex)
{
}
//contentarray = page.GetAsArray(PdfName.CONTENTS);
//if (contentarray != null)
//{
// //Loop through content
// for (int j = 0; j < contentarray.Size; j++)
// {
// //Get the raw byte stream
// stream = (PRStream)contentarray.GetAsStream(j);
// //Convert to a string. NOTE, you might need a different encoding here
// content = System.Text.Encoding.ASCII.GetString(PdfReader.GetStreamBytes(stream));//获取pdf页内的文字内容
// //Look for the OCG token in the stream as well as our watermarked text
// if (content.IndexOf("/OC") >= 0 || content.IndexOf(watermarkText) >= 0)//如果pdf内容包含水印文字
// {
// //Remove it by giving it zero length and zero data
// content = content.Replace(watermarkText, "");//替换水印文字为空
// byte[] byteArray = System.Text.Encoding.Default.GetBytes(content);//转换为byte[]
// stream.Put(PdfName.LENGTH, new PdfNumber(byteArray.Length));//重新指定大小
// stream.SetData(byteArray);//重新赋值
// }
// }
//}
}
//PdfStamper stamper = new PdfStamper(reader, new FileStream(DEST, FileMode.Create, FileAccess.ReadWrite));
MemoryStream mstream = new MemoryStream();
PdfStamper stamper = new PdfStamper(reader, mstream);
stamper.Close();
reader.Close();
return(mstream);
}
public byte[] GetImageAsBytes()
{
if (streamBytes == null)
{
return(null);
}
if (!decoded)
{
// if the stream hasn't been decoded, check to see if it is a single stage JPG or JPX encoded stream. If it is,
// then we can just use stream as-is
PdfName filter = dictionary.GetAsName(PdfName.FILTER);
if (filter == null)
{
PdfArray filterArray = dictionary.GetAsArray(PdfName.FILTER);
if (filterArray.Size == 1)
{
filter = filterArray.GetAsName(0);
}
else
{
throw new UnsupportedPdfException("Multi-stage filters not supported here (" + filterArray + ")");
}
}
if (PdfName.DCTDECODE.Equals(filter))
{
fileType = TYPE_JPG;
return(streamBytes);
}
else if (PdfName.JPXDECODE.Equals(filter))
{
fileType = TYPE_JP2;
return(streamBytes);
}
throw new UnsupportedPdfException("Unsupported stream filter " + filter);
}
pngColorType = -1;
width = dictionary.GetAsNumber(PdfName.WIDTH).IntValue;
height = dictionary.GetAsNumber(PdfName.HEIGHT).IntValue;
bpc = dictionary.GetAsNumber(PdfName.BITSPERCOMPONENT).IntValue;
pngBitDepth = bpc;
PdfObject colorspace = dictionary.GetDirectObject(PdfName.COLORSPACE);
palette = null;
icc = null;
stride = 0;
FindColorspace(colorspace, true);
MemoryStream ms = new MemoryStream();
if (pngColorType < 0)
{
if (bpc != 8)
{
return(null);
}
if (PdfName.DEVICECMYK.Equals(colorspace))
{
}
else if (colorspace is PdfArray)
{
PdfArray ca = (PdfArray)colorspace;
PdfObject tyca = ca.GetDirectObject(0);
if (!PdfName.ICCBASED.Equals(tyca))
{
return(null);
}
PRStream pr = (PRStream)ca.GetDirectObject(1);
int n = pr.GetAsNumber(PdfName.N).IntValue;
if (n != 4)
{
return(null);
}
icc = PdfReader.GetStreamBytes(pr);
}
else
{
return(null);
}
stride = 4 * width;
TiffWriter wr = new TiffWriter();
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_SAMPLESPERPIXEL, 4));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_BITSPERSAMPLE, new int[] { 8, 8, 8, 8 }));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_PHOTOMETRIC, TIFFConstants.PHOTOMETRIC_SEPARATED));
wr.AddField(new TiffWriter.FieldLong(TIFFConstants.TIFFTAG_IMAGEWIDTH, width));
wr.AddField(new TiffWriter.FieldLong(TIFFConstants.TIFFTAG_IMAGELENGTH, height));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_COMPRESSION, TIFFConstants.COMPRESSION_LZW));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_PREDICTOR, TIFFConstants.PREDICTOR_HORIZONTAL_DIFFERENCING));
wr.AddField(new TiffWriter.FieldLong(TIFFConstants.TIFFTAG_ROWSPERSTRIP, height));
wr.AddField(new TiffWriter.FieldRational(TIFFConstants.TIFFTAG_XRESOLUTION, new int[] { 300, 1 }));
wr.AddField(new TiffWriter.FieldRational(TIFFConstants.TIFFTAG_YRESOLUTION, new int[] { 300, 1 }));
wr.AddField(new TiffWriter.FieldShort(TIFFConstants.TIFFTAG_RESOLUTIONUNIT, TIFFConstants.RESUNIT_INCH));
wr.AddField(new TiffWriter.FieldAscii(TIFFConstants.TIFFTAG_SOFTWARE, Document.Version));
MemoryStream comp = new MemoryStream();
TiffWriter.CompressLZW(comp, 2, streamBytes, height, 4, stride);
byte[] buf = comp.ToArray();
wr.AddField(new TiffWriter.FieldImage(buf));
wr.AddField(new TiffWriter.FieldLong(TIFFConstants.TIFFTAG_STRIPBYTECOUNTS, buf.Length));
if (icc != null)
{
wr.AddField(new TiffWriter.FieldUndefined(TIFFConstants.TIFFTAG_ICCPROFILE, icc));
}
wr.WriteFile(ms);
fileType = TYPE_TIF;
return(ms.ToArray());
}
PngWriter png = new PngWriter(ms);
png.WriteHeader(width, height, pngBitDepth, pngColorType);
if (icc != null)
{
png.WriteIccProfile(icc);
}
if (palette != null)
{
png.WritePalette(palette);
}
png.WriteData(streamBytes, stride);
png.WriteEnd();
fileType = TYPE_PNG;
return(ms.ToArray());
}