Pdfer

A basic C# library meant to make accessing and manipulating PDFs complicated but extremely powerful.

Usage

Pdfer uses Streams to read and write PDFs. It's best to open a Stream with your PDF content, like a FileStream, to reduce memory usage while loading and parsing the PDF.

For a basic example, see the TestConsole Project.

What, you really want more detail? Fine.

Parsing

To parse a PDF, you can create a PdfDocumentParser with the PdfDocumentParserFactory. If you want to adjust behaviour of the parser, you can of course make your own and overwrite the behaviour of the various helper classes, but I recommend not doing that unless extremely necessary. If you need another feature, just make a PR and make the world better for everyone.

using var stream = File.OpenRead("test.pdf");
var parser = PdfDocumentParserFactory.Create();
var document = parser.Parse(stream);

You can also load the PDF into memory and parse a byte array:

byte[] pdf = File.ReadAllBytes("test.pdf");
var parser = PdfDocumentParserFactory.Create();
var document = parser.Parse(pdf);

Manipulating

Currently, manipulation of PDFs is very limited. You can access and edit the objects on the parsed level, like changing the Creator in the dictionary:

var infoDictionary = pdfDocument.DocumentParts[0].Trailer.TrailerDictionary["Info"] switch
{
  IndirectObject indirectObject => pdfDocument.DocumentParts[0].Body[indirectObject.ObjectIdentifier] as DictionaryObject
                                   ?? throw new InvalidOperationException("Info dictionary not found"),
  DictionaryObject dictionaryObject => dictionaryObject,
  _ => throw new InvalidOperationException("Info dictionary not found")
};

infoDictionary.Value["Producer"] = new StringObject(PdfStringHelper.AsHexString("PDFer"));
infoDictionary.Value["Title"] = new StringObject(PdfStringHelper.AsHexString("My PDFer Specification!!!"));

What you currently can't do is changing the raw data, even though there's a RawValue on DocumentObject, this is currently ignored. This might change at some point, if you need it you can make an issue. It's just not a priority for me right now.

Writing

Writing is done with the PdfDocumentWriter and you guessed it, you can make one with the PdfDocumentWriterFactory.

var writer = PdfDocumentWriterFactory.Create();
var stream = File.OpenWrite("test.pdf");
writer.Write(stream, document);

Known Issues

• The parser doesn't currently support multilayer PDFs.
• Things like signed PDFs with multiple trailers don't work.
• Writing PDFs currently only really works for extraordinarily simple and conform PDFs.
• Pdfs with \r\n delimiters are not supported
• Encoding is all over the place. Best to just use ASCII in your PDFs.
• Fuck streamed xref tables, PDF1.5 is evil

Why

I and someone I know want a library that allows us to easily manipulate PDFs on a object level, so I decided to parse PDFs.

Help

Make an issue and pray I have the time to help

I want to give help

Yes!