Actually, converting PDF to a file format is very non-trivial. The problem is that the PDF typically retains none of the semantic structure of the original file

Actually, converting PDF to a file format is very non-trivial. The problem is that the PDF typically retains none of the semantic structure of the original file. Yes, I know there are converters to Word, none of which are perfect, but in each case the people who wrote the converters had to do a lot of work. PDF is all about geometry. So even the characters in a word do not have to be together in the PDF content stream. (I worked extensively on semantic reconstruction of PDF documents before I retired.)