Comment by kargo

Interesting... I see it now. I assume some issue during the PDF to image conversion in the web app. PDF support is just a few days old.

The OCR library itself supports only image formats as input and is "innocent" with regards to this issue ;)