OCR - make folder "tessdata" writeable

[odie]

Hello

The OCR feature is very useful. But I need to add addtional languages (Norwegian and Swedish). According to Striling documentation (here:), it should be possible to upload new xx.traineddata files containing more languages.

I ran into a problem trying to do this. The /usr/share/tessdata folder is read-only, so I cannot add these languages. Can this be fixed? Either by finding a solution to make the tessdata folder wrtieable for end users (probably the best solution?), or by adding these lanuages to the standard pack?

Thank you.

[nebulon]

Indeed such folders are read-only on Cloudron. Instead of attempting to make them read/write with some merging strategy for future updates, I have just added those two languages to the package. The initial selection was a bit random, so we can just add them one-by-one as requested here.

The package update just got our with Swedish and Norwegian.

[odie]

@nebulon Thank you very much, @nebulon! Much appreciated!

[Algaweb]

Hello @nebulon I was trying to do the same for Czech language. Would you please add the language to the package? Thank you.

[scooke]

I had been using this on a free Crunchbits server that has now been cancelled, so I'm going to install it on my Cloudron. @nebulon, could also add these languages please (most likely they aren't by default included!):
Kazak - https://github.com/tesseract-ocr/tessdata/blob/main/kaz.traineddata
Uighur - https://github.com/tesseract-ocr/tessdata/blob/main/uig.traineddata

I'll install once these are included. Is it safe to assume languages like German and French are included?