Incorrect escaping in utils.js_to_json

[LukeLR]

Checklist

• I'm reporting a broken site support issue
• I've verified that I'm running youtube-dl version 2020.03.24
• I've checked that all provided URLs are alive and playable in a browser
• I've checked that all URLs and arguments with special characters are properly quoted or escaped
• I've searched the bugtracker for similar bug reports including closed ones
• I've read bugs section in FAQ

Description

Consider a JavaScript dictionary that has a key-value pair like this:

{
link: encodeURI('https://mediathek.hhu.de/watch/6ea779d9-a2c1-4150-8d26-0653832e9d67/'),
}

utils.js_to_json is not able to escape this value correctly. The expected result would be:

{
"link": "encodeURI('https://mediathek.hhu.de/watch/6ea779d9-a2c1-4150-8d26-0653832e9d67/')",
}

However, the actual result is:

"link": "encodeURI"("https://mediathek.hhu.de/watch/7db1e695-b4c2-46a7-9227-1b22d7c7c05f/")

As the value is not enclosed in ', the value gets split into two seperate parts, which is not valid JSON anymore. But as the input was valid JavaScript, js_to_json should be able to produce valid JSON from it as well.

I think, the regex in youtube_dl/utils.py#L4022 could be adapted to parse a single word bevor a : as a key, and anything after the : until the next , as a value. This would allow parsing of JavaScript like this and similar cases, and maybe even simplify the regex a bit.

I tried implementing such a fix for the past hour now, but I guess, I'm not enough of a regex magician for that. (although I'll keep looking into it). What do you think?

Any other ideas? How can js_to_json be improved to cover JavaScript like this?