public static PDFStream MakeStream(PDFDictionary StreamDictionary, byte[] Data, int StartingIndex, out int EndingIndex)
{
int TokEnd;
int StreamLength = 0;
if (StreamDictionary.ContainsKey("Length"))
{
StreamLength = (int)(Decimal.Parse(StreamDictionary["Length"].Description));
}
else
{
StreamLength = PDF.FirstOccurance(Data, Encoding.UTF8.GetBytes("endstream"), StartingIndex);
}
EndingIndex = StartingIndex;
if ("stream".Equals(PDFObjectParser.GetTokenString(Data, EndingIndex, out _, out TokEnd)))
{
if ((StreamLength > 0) && (Data.Length >= TokEnd + 1 + StreamLength))
{
byte[] StreamBytes = new byte[StreamLength];
Array.Copy(Data, TokEnd + 1, StreamBytes, 0, StreamLength);
EndingIndex = TokEnd + 1 + StreamLength;
if (!"endstream".Equals(PDFObjectParser.GetTokenString(Data, EndingIndex, out _, out EndingIndex)))
{
} // Error - missing "endstream"
return(new PDFStream(StreamDictionary, StreamBytes));
}
}
// Unable to make a stream from this data
return(null);
}
/// <summary>
/// Attempt to parse the given data stream, returning an indicator of parse progress
/// </summary>
/// <param name="StartingToken">The token immediately preceeding the starting index in Data stream</param>
/// <param name="Data">Raw byte stream to parse</param>
/// <param name="StartingIndex">0-based starting index into Data where StartingToken appears</param>
/// <param name="EndingIndex">Index into data stream where parsing ended (either successfully or unsuccessfully)</param>
/// <returns>Object parsed from data stream, or NULL if unable to parse. If NULL and EndingIndex is equal to Data.Length, parsing may be successful with more data</returns>
public static IPDFObject TryParse(string StartingToken, byte[] Data, int StartingIndex, out int EndingIndex)
{
int StartIndex;
int Number;
EndingIndex = StartingIndex;
if (int.TryParse(StartingToken, out Number))
{
string Declaration = PDFObjectParser.GetTokenString(Data, StartingIndex + StartingToken.Length, out StartIndex, out EndingIndex, 2);
if (!String.IsNullOrEmpty(Declaration))
{
Match objMatch = Regex.Match(Declaration, @"([+-]?\d+) R");
if (objMatch.Success)
{
int Generation = int.Parse(objMatch.Groups[1].Value);
if (Declaration.Length != objMatch.Length)
{
EndingIndex = EndingIndex - (Declaration.Length - objMatch.Length) - 1;
}
return(new PDFIndirectObject(Number, Generation));
}
}
}
return(null);
}
/// <summary>
/// Load a single object from the PDF data stream, starting at a paritcular offset
/// </summary>
/// <param name="Offset">Offset into data stream of object to read</param>
/// <returns>Object as read out of data stream, or null if no object could be parsed</returns>
private IPDFObject LoadObjectAtOffset(int Offset)
{
IPDFObject objectDefinition = PDFObjectParser.Parse(RawData, out _, Offset);
if (objectDefinition.Type == PDFObjectType.ObjectDefinition)
{
return(((PDFObjectDefinition)objectDefinition).Object);
}
return(objectDefinition);
}
/// <summary>
/// Read the trailer from a PDF data file
/// </summary>
/// <param name="Data">Data to read</param>
/// <param name="StartIndex">Starting index of where to look for trailer, or -1 to look from end of file (default: -1)</param>
/// <returns>TRUE if a trailer was successfully read, FALSE otherwise</returns>
public static PDFTrailer ReadTrailer(byte[] Data, int StartIndex = -1)
{
int EndIndex = StartIndex;
if (EndIndex < 0)
{
EndIndex = PDF.FindEOF(Data, Data.Length - 1);
}
if (EndIndex < 0)
{
return(null);
}
int EndOfLineIndex;
byte[] LineData = PDF.ExtractPreviousPDFLine(Data, EndIndex, out EndIndex, out EndOfLineIndex);
while (LineData != null)
{
if ("trailer".Equals(Encoding.UTF8.GetString(LineData).Trim()))
{
int TokenStartIndex = 0;
string Token = PDFObjectParser.GetTokenString(Data, EndOfLineIndex, out TokenStartIndex, out EndIndex);
PDFDictionary TrailerDictionary = (PDFDictionary)PDFDictionary.TryParse(Token, Data, TokenStartIndex, out EndIndex);
if (TrailerDictionary != null)
{
LineData = PDF.ExtractPDFLine(Data, EndIndex, out EndIndex);
if ("startxref".Equals(Encoding.UTF8.GetString(LineData).Trim()))
{
Token = PDFObjectParser.GetTokenString(Data, EndIndex, out TokenStartIndex, out _);
PDFNumber Offset = (PDFNumber)PDFNumber.TryParse(Token, Data, TokenStartIndex, out EndIndex);
PDFCrossReference CrossRef = PDFCrossReference.ReadCrossReference(Data, Offset, out _);
return(new PDFTrailer(TrailerDictionary, CrossRef));
}
else
{
return(null);
}
}
else
{
return(null);
}
}
else
{
LineData = PDF.ExtractPreviousPDFLine(Data, EndIndex, out EndIndex, out EndOfLineIndex);
}
}
return(null);
}
public List <IPDFObject> GetObjects(int MaxObjects)
{
byte[] AllData = File.ReadAllBytes(Filepath);
int EndingIndex = 0;
List <IPDFObject> objects = new List <IPDFObject>();
IPDFObject nextObject;
while ((objects.Count < MaxObjects) && ((nextObject = PDFObjectParser.Parse(AllData, out EndingIndex, EndingIndex)) != null))
{
objects.Add(nextObject);
}
return(objects);
}
/// <summary>
/// Attempt to parse the given data stream, returning an indicator of parse progress
/// </summary>
/// <param name="StartingToken">The token immediately preceeding the starting index in Data stream</param>
/// <param name="Data">Raw byte stream to parse</param>
/// <param name="StartingIndex">0-based starting index into Data where StartingToken appears</param>
/// <param name="EndingIndex">Index into data stream where parsing ended (either successfully or unsuccessfully)</param>
/// <returns>Object parsed from data stream, or NULL if unable to parse. If NULL and EndingIndex is equal to Data.Length, parsing may be successful with more data</returns>
public static IPDFObject TryParse(string StartingToken, byte[] Data, int StartingIndex, out int EndingIndex)
{
Dictionary <string, IPDFObject> KeyValuePairs = new Dictionary <string, IPDFObject>();
EndingIndex = StartingIndex;
if (StartingToken.Equals("<<"))
{
EndingIndex += StartingToken.Length;
while (EndingIndex < Data.Length)
{
int TokEnd;
if (">>".Equals(PDFObjectParser.GetTokenString(Data, EndingIndex, out _, out TokEnd)))
{
EndingIndex = TokEnd;
if ("stream".Equals(PDFObjectParser.GetTokenString(Data, EndingIndex, out _, out TokEnd)))
{
return(PDFStream.MakeStream(new PDFDictionary(KeyValuePairs), Data, EndingIndex, out EndingIndex));
}
else
{
return(new PDFDictionary(KeyValuePairs));
}
}
IPDFObject Key = PDFObjectParser.Parse(Data, out EndingIndex, EndingIndex);
if (Key == null)
{
return(null); // No key found
}
if (Key.Type != PDFObjectType.Name)
{
return(null); // Invalid key type found
}
IPDFObject Value = PDFObjectParser.Parse(Data, out EndingIndex, EndingIndex);
if (Value == null)
{
return(null); // No value found
}
KeyValuePairs.Add(Key.Description, Value);
}
}
// Didn't find a dictionary delimiter
return(null);
}
/// <summary>
/// Attempt to parse the given data stream, returning an indicator of parse progress
/// </summary>
/// <param name="StartingToken">The token immediately preceeding the starting index in Data stream</param>
/// <param name="Data">Raw byte stream to parse</param>
/// <param name="StartingIndex">0-based starting index into Data where StartingToken appears</param>
/// <param name="EndingIndex">Index into data stream where parsing ended (either successfully or unsuccessfully)</param>
/// <returns>Object parsed from data stream, or NULL if unable to parse. If NULL and EndingIndex is equal to Data.Length, parsing may be successful with more data</returns>
public static IPDFObject TryParse(string StartingToken, byte[] Data, int StartingIndex, out int EndingIndex)
{
int Number;
EndingIndex = StartingIndex;
if (int.TryParse(StartingToken, out Number))
{
string Declaration = PDFObjectParser.GetTokenString(Data, StartingIndex + StartingToken.Length, out _, out EndingIndex, 2);
if (!String.IsNullOrEmpty(Declaration))
{
Match objMatch = Regex.Match(Declaration, @"([+-]?\d+) obj");
if (objMatch.Success)
{
if (Declaration.Length != objMatch.Length)
{
EndingIndex = EndingIndex - (Declaration.Length - objMatch.Length) - 1;
}
int Generation = int.Parse(objMatch.Groups[1].Value);
// Parse the indirect object content (should be a single object)
IPDFObject PDFObject = PDFObjectParser.Parse(Data, out EndingIndex, EndingIndex);
if (PDFObject != null)
{
int ObjectEndIndex = PDF.FirstOccurance(Data, Encoding.UTF8.GetBytes("endobj"), EndingIndex);
if (ObjectEndIndex > 0)
{
// Format error - tokens after object definition but before "endobj"
}
EndingIndex = ObjectEndIndex + "endobj".Length;
return(new PDFObjectDefinition(Number, Generation, PDFObject));
}
}
}
}
return(null);
}
/// <summary>
/// Attempt to parse the given data stream, returning an indicator of parse progress
/// </summary>
/// <param name="StartingToken">The token immediately preceeding the starting index in Data stream</param>
/// <param name="Data">Raw byte stream to parse</param>
/// <param name="StartingIndex">0-based starting index into Data where StartingToken appears</param>
/// <param name="EndingIndex">Index into data stream where parsing ended (either successfully or unsuccessfully)</param>
/// <returns>Object parsed from data stream, or NULL if unable to parse. If NULL and EndingIndex is equal to Data.Length, parsing may be successful with more data</returns>
public static IPDFObject TryParse(string StartingToken, byte[] Data, int StartingIndex, out int EndingIndex)
{
EndingIndex = StartingIndex;
if (!String.IsNullOrEmpty(StartingToken) && (StartingToken[0] == '['))
{
PDFArray ObjectArray = new PDFArray();
EndingIndex = StartingIndex + 1;
while (EndingIndex < Data.Length)
{
IPDFObject nextObject = PDFObjectParser.Parse(Data, out EndingIndex, EndingIndex);
if (nextObject != null)
{
ObjectArray.Add(nextObject);
}
byte nextChar = PDF.Whitespace[0];
while (PDF.IsWhitespace(nextChar))
{
nextChar = Data[EndingIndex++];
}
if (nextChar == ']')
{
return(ObjectArray);
}
else
{
EndingIndex--;
}
}
}
return(null);
}
