I have an open source project (haven't chosen a license yet) that uses tesseract for OCR. This is a windows only project. My program requires the end user to have a copy of tesseract on their system and I would rather they didn't have to install it themselves

I'm wondering if I can use the windows installer (https://github.com/UB-Mannheim/tesseract/wiki) to get the files, and then include those compiled files in my own project and upload to GitHub?

To be clear, I'm not talking about including the tesseract source code (which I believe is licensed under apache 2.0). I'd like to include the files installed by an installer, which was created by Mannheim University Library (see link above)


The beauty of open code is that bundling can be done alright.

In your case there are various parts in play: - tesseract proper which seems to be overall under the Apache license and some extras BSD - additions by UB-Mannheim: the actual installer -if any- and possibly extra fonts from a quick look at the repos

If you want to bundle a pre-built installer using the UB-Mannheim fors as a base, you would need to ensure that you comply with the licensing of all these parts. My assumption would be that any installer script is like to be under the same license as tesseract. But the fonts may have different licenses that you need to research.

In general if we are talking about Apache and BSD, the key requirement for compliance is attribution (and possibly some change notifications) when you redistribute binaries which should be pretty easy to produce.

Keep us posted!

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.