Created attachment 117281 [details]
Resulting file after using pdftocairo (pages 1 to 13)
When I open the PDF textbook provided by the following link:
some images and pages are shown as black boxes/ black shapes. See, for example, the following pages: 1, 8, 12, 13, 28, 33, 37. Compare results by opening the same file with Adobe Reader, or even with Firefox's built-in PDF reader.
I first reported the issue to the Evince development team:
but they couldn't reproduce the issue, neither with poppler 0.30 nor with 0.32. So, it's probably a regression in 0.33.0-1. I can reproduce the bug with pdftocairo (see the attached resulting file).
PDF viewer: Evince 3.16.1
Operating System: Antergos
Created attachment 117282 [details]
pdftocairo output after converting pages 1 to 13 to a set of png images
Created attachment 117283 [details]
pdftocairo output after converting pages 1 to 13 to a single pdf file
It's because those developers are probably using poppler without linking to openjpeg and you are.
openjpeg is a better backend for jpeg2000 decoding but seems to choke on this file sadly.
Then, should I file a bug against Openjpeg? If so, which info should I provide?
Yes and no, openjpeg is a complicated project, it has a newer version we don't use because it regresses on our use case and then the older version is understandably unmaintained.
We need some time to get them to fix their newer version, move to it and then we can start reporting bugs against them.
-- GitLab Migration Automatic Message --
This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.
You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/poppler/poppler/issues/28.