I have used bash to write an OCR processor that called a python wrapper around tesseract, and then turned the pdf output into json to go into a solr search data

I have used bash to write an OCR processor that called a python wrapper around tesseract, and then turned the pdf output into json to go into a solr search database by parsing the output with sed. No, it wasn't smart to do so. Yes, it still works. No, I don't know how it works.