Hi,
I found two problems in the OCR functions (at the moment in the "Production"):
the Dockerfile it was missing the package 'tesseract-ocr'
the ocr function when return text contains some html characters, the edit_message function receive from Telegram an error parsing because the parse_mode is html.
In the Dockerfile, now the scripts run with a no-root user.
I created a workflow action for automatic Docker image build and publish in the package manager (for now only on push).
After some test, I will add an test task for test the dockerfile build when there is a pull request.
At the moment, is available on my repository a docker image builded with this action system
Hi, I found two problems in the OCR functions (at the moment in the "Production"):
In the Dockerfile, now the scripts run with a no-root user.
I created a workflow action for automatic Docker image build and publish in the package manager (for now only on push). After some test, I will add an test task for test the dockerfile build when there is a pull request.
At the moment, is available on my repository a docker image builded with this action system