Can not open some jbig2 image

[Charltsing]

IndexOutOfRangeException
or
NullReferenceException

error.zip

[Matti-Koopa]

Same here. File see attachment.
Stacktrace:
at JBig2Decoder.NETStandard.TextRegionSegment.ReadSegment() in JBig2Decoder.NETStandard\TextRegionSegment.cs:line 51

Basically, segment is null here but not checked for null:

Segment segment = decoder.FindSegment(referredToSegments[i]);
if (segment.GetSegmentHeader().GetSegmentType() == 0)

pdf-stream11-0.zip

[nicholsab]

We fixed this problem in: https://www.nuget.org/packages/JBig2Decoder.NETStandard/
Your test file pdf-stream11-0.jbig2 works correctly with it.

[Matti-Koopa]

This doesn't seem fixed to me. I extract the image from a PDF. This is the PDF:
1082193-2021-10.pdf

System.NullReferenceException: Object reference not set to an instance of an object.
   at JBig2Decoder.NETStandard.TextRegionSegment.ReadSegment()
   at JBig2Decoder.NETStandard.JBIG2StreamDecoder.ReadSegments()
   at JBig2Decoder.NETStandard.JBIG2StreamDecoder.DecodeJBIG2(Byte[] data, Int32& width, Int32& height)

How I use it (with UglyToad PdfPig NuGet):

using PdfDocument pdfDocument = PdfDocument.Open(bytes, new ParsingOptions());
pdfDocument.GetPages()
page.GetImages()
image.RawMemory
byte[] output = new JBIG2StreamDecoder().DecodeJBIG2(bytes, out int width, out int height);

It's the second picture in the PDF.

[nicholsab]

This is the picture that is created from that pdf-stream11-0.jbig2, using ImageSharp to save as a png, and there are no exceptions. Maybe the pdf is doing something with rotation and color changes.

Also, in that example you have variable bytes in the PDFDocument.Open and bytes in the DecodeJBIG2, is that on purpose?