amet wrote: ↑20 Mar 2019, 20:34
Webscrape is purely based on regex and changes when the website changes.... we have played this game in Kodi days and it was not a good idea then and is not a good idea now
I'm a bit confused by this, and why we're focusing on the web scraping part. What I have been proposing is to let me take care of the implementation details. Web scrape via regex is of course one potential implementation but so is e.g. controlling a headless Chrome, which is not limited by JS-heavy sites, for example. That would also allow usage of JS to query the DOM. I realize it's not a fit for integration into MrMC, but again, that's not what I was proposing.
I'm not really interested in debating about what's technologically possible and what is not. This argument does not really help me see any limitations to my proposal. It would require a bit of work to replicate the OpenSubtitles API, but once that's in place there aren't really any limitations since it's running as an external service, pretending to be OpenSubtitles. Thanks for the links.