Robert Sachunsky

> with ..._dta50 there is missing some punctation, but there is almost no word diff ... I've expected the dictionary to have greater impact. me too. But the averages do...

> - this has to be done in an external ocrd-core copy (of the PR), it does not work as a submodule IIUC > - Reason: when building, `make assets`...

> * docker-compose could be changed to download processor-images (if made available with publishing images for the processors) instead of building with path to Dockerfiles I agree, but ideally the...

I believe we should try to generate the docker-compose files per module from their respective ocrd-tool.json (or the ocrd-all-tool.json) and then include all sub-compose files into a single top-level one,...

> * docker-compose currently uses for the core-test-container, which is used to run the tests, a self created Image because the test are not yet available in the published image...

> I believe we should try to generate the docker-compose files per module from their respective ocrd-tool.json (or the ocrd-all-tool.json) and then include all sub-compose files into a single top-level...

My suggested fix would be to use the same strategy as in #417 – assume core (via target `bin/ocrd`) is already installed right from the start, i.e. `FROM` stage. So...

Dup of #42. See esp. [my comment about parameters which might help](https://github.com/OCR-D/ocrd_anybaseocr/issues/42#issuecomment-1077502826). But basically broken and not supported. (Remember we only keep this repo up for the cropper ATM, will...

Note: this is not covered by `make check` (which succeeds for current master, despite my l10n).

Thanks @stweil for your analysis! Just two points: - Tesseract uses `pixaConvertToPdf` in two places for debugging. (The PDF renderer does not seem to use Leptonica directly, but it does...