More and more sites are using frontend rendering in JS. Wallabag can't handle this contents as the parser (curl-like) doesn't run any kind of script.

It would be great that Wallabag can parse these pages.
We could achieve this by fetching problematic sites with a headless Chromium.

This project could help in getting this to work: https://github.com/gildas-lormeau/single-file-cli
I've made experiments (on the command-line) and lynx -dump shows the page contents properly.

The next steps I'm going to experiment are:

• make a Web API from single-file-cli
• manually feed my Wallabag instance with pages from this API

If it works correctly, we could imagine this as a fallback when internal parser fails to retrieve the page body.

We'll have to deal with security issues: it would be dangerous to allow any URL to be handled by a headless Chromium. The fallback API should restrict the pages it accepts to handle.
This will probably lead to user expectations vs. admin restrictions difficulties.