/// <summary>
/// Attempt to parse a sequence of bytes as a PDF object
/// </summary>
/// <param name="Data">Bytes to parse</param>
/// <param name="EndingIndex">0-based index into data indicating the byte after the last parsed byte</param>
/// <param name="StartingIndex">0-based index into data at which to start parsing (default: 0)</param>
/// <returns>PDF object successfully parsed from the data, or NULL if no complete object was parsable</returns>
public static IPDFObject Parse(byte[] Data, out int EndingIndex, int StartingIndex = 0)
{
int StartTokenIndex;
int EndTokenIndex;
IPDFObject ParsedObject = null;
string Token = GetTokenString(Data, StartingIndex, out StartTokenIndex, out EndTokenIndex);
if (!string.IsNullOrEmpty(Token))
{
if ((ParsedObject = PDFDictionary.TryParse(Token, Data, StartTokenIndex, out EndingIndex)) != null)
{
return(ParsedObject);
}
if ((ParsedObject = PDFArray.TryParse(Token, Data, StartTokenIndex, out EndingIndex)) != null)
{
return(ParsedObject);
}
if ((ParsedObject = PDFString.TryParse(Token, Data, StartTokenIndex, out EndingIndex)) != null)
{
return(ParsedObject);
}
if ((ParsedObject = PDFHexString.TryParse(Token, Data, StartTokenIndex, out EndingIndex)) != null)
{
return(ParsedObject);
}
if ((ParsedObject = PDFObjectDefinition.TryParse(Token, Data, StartTokenIndex, out EndingIndex)) != null)
{
return(ParsedObject);
}
if ((ParsedObject = PDFIndirectObject.TryParse(Token, Data, StartTokenIndex, out EndingIndex)) != null)
{
return(ParsedObject);
}
if ((ParsedObject = PDFNumber.TryParse(Token, Data, StartTokenIndex, out EndingIndex)) != null)
{
return(ParsedObject);
}
if ((ParsedObject = PDFName.TryParse(Token, Data, StartTokenIndex, out EndingIndex)) != null)
{
return(ParsedObject);
}
if ((ParsedObject = PDFComment.TryParse(Token, Data, StartTokenIndex, out EndingIndex)) != null)
{
return(ParsedObject);
}
}
// Could not parse next object
EndingIndex = StartingIndex;
return(null);
}
/// <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);
}
/// <summary>
/// Attempt to read a PDF cross reference subsection from raw data starting at a given offset
/// </summary>
/// <param name="Data">PDF File Data</param>
/// <param name="StartIndex">Index into Data where cross reference subsection should start</param>
/// <param name="EndIndex">Index into Data where parsing of the cross reference subsection stopped (successfully or not)</param>
/// <returns>Instance of PDF Cross Reference Subsection on success, NULL on failure to parse</returns>
private static PDFCrossReference ReadCrossReferenceSubsection(byte[] Data, int StartIndex, out int EndIndex)
{
string Declaration = Encoding.UTF8.GetString(PDF.ExtractPDFLine(Data, StartIndex, out EndIndex));
string[] Definitions = Declaration.Split(' ');
if (Definitions.Length == 2)
{
int StartObject = PDFNumber.TryParse(Definitions[0], -1);
int NumObjects = PDFNumber.TryParse(Definitions[1], -1);
if ((StartObject >= 0) && (NumObjects >= 0))
{
PDFCrossReference Subsection = new PDFCrossReference(StartObject, NumObjects);
for (int i = 0; i < NumObjects; i++)
{
// Parse cross reference table entries
string Next = Encoding.UTF8.GetString(PDF.ExtractPDFLine(Data, EndIndex, out EndIndex));
Match objMatch = Regex.Match(Next, @"(\d\d\d\d\d\d\d\d\d\d) (\d\d\d\d\d) ([n|f])");
if (objMatch.Success)
{
int Generation = int.Parse(objMatch.Groups[2].Value);
char type = objMatch.Groups[3].Value[0];
if (type == 'n')
{
// Add active object to table
int Offset = int.Parse(objMatch.Groups[1].Value);
Subsection.ObjectOffsets[i] = Offset;
Subsection.ObjectGenerations[i] = Generation;
}
else if (type == 'f')
{
// Add deleted object to table
Subsection.ObjectOffsets[i] = 0;
Subsection.ObjectGenerations[i] = Generation;
}
else
{
return(null); // Should never get here; invalid cross reference entry
}
}
else
{
return(null); // Invalid Cross Reference Section
}
}
return(Subsection);
}
}
return(null);
}