6.2.0 Preview 2 cannot read certain files

[packdat]

Testing the latest version 6.2.0 Preview 2 with my list of test-files, i observed some small issues that prevented the library from reading certain files.

• When reading objects from object-streams, you create instances of PdfObjectStream and Parser to read objects from the stream.
  This can fail when the stream is compressed with the deflate-filter, because you use objectStream.Stream.Value instead of objectStream.Stream.UnfilteredValue. Instances of this here and here.
• There are files out there, where an object is referenced in more than one object-stream.
  This causes an exception here because you try to add an already added object-id to the dictionary.

I attached a ZIP containing 2 documents which could not be read by the library because of the mentioned issues.
Tests were performed with my usual test-case (mentioned for example in #152 )

Issue_Documents.zip

Out of curiosity:
What kind of tests do you run before pushing a new version ?
If you're not doing this already, i highly suggest collection a bunch of files and running a simple import-test against them.
(like the one in #152 )
The issues described here would immediately pop up.

[StLange]

The bugs will be fixed in the next release (6.2.0 Preview 3).

Out of curiosity

We have an internal repository PDFsharp.Tests where we test all ever submitted PDFfiles which have issues with PDFsharp. Periodically and before each release we use it to ensure that all issues ever fixed with reading PDF files are still fixed. Unfortunately, we did something wrong with some of the unit tests. So, we overlooked some issues. Should work correctly in the next release.

[packdat]

Great news, thank you ! 👍

[PDFsharp-Team]

Fixed with PDFsharp 6.2.0 Preview 3.