This is a single-purpose script - not only useless by itself, but even when integrated (or planned to be integrated) into a larger pipeline, it's of interest only to people who 1) speak Czech, and 2) are interested in Czech Open data and/or public procurement.
www.hlidacstatu.cz indexes and curates data made available (compulsorily and not very conveniently) by Czech public administration - public contracts, companies supplying them, politicians on their boards etc. Among other things, it publishes a developer API (described at https://hlidacstatu.docs.apiary.io/) allowing registered 3rd parties to upload structured data they consider relevant to the site's mission.
One such dataset is the list of public electromobile charging stations, available from https://www.mpo.cz/cz/energetika/statistika/statistika-cerpacich-stanic-pohonnych-hmot/seznam-verejnych-dobijecich-stanic-_-stav-k-31--10--2019--250631/ as a PDF file containing a simple table. The make-json.py script converts it into a machine-readable format.
make-json.py depends on PDFMiner (www.github.com/euske/pdfminer). For configuration, it requires a config.json file in the current directory, with a mandatory setting "sourceFile" for path to the downloaded PDF. The "targetDir" setting specifies the output directory, by default "json". A separate file is generated there for every table row, in a format suitable for upload by www.github.com/vbar/krmic-hlidace-statu . Note the script overwrites files in "targetDir" if they already exist - if you want to preserve results of a specific run, copy them somewhere else.
Generated JSON conforms to the included schema; the full JSON for registration of a new dataset (in a format documented at https://hlidacstatu.docs.apiary.io/) is also included, as an example.