Frozen-soup

This now properly handles including URLs based on a `<base>` tag, and there's a new `knockouts` parameter (or repeatable `--knockout` command line option) that allows specifying CSS selectors to be removed from the output HTML. This could be useful for removing popups or ads, for example.

(Forgot to bump the version number in `pyproject.toml`, so this is a re-do of 0.4 with that fixed.)

This now properly handles including URLs based on a `<base>` tag, and there's a new `knockouts` parameter (or repeatable `--knockout` command line option) that allows specifying CSS selectors to be removed from the output HTML. This could be useful for removing popups or ads, for example.

Added a timeout parameter for freezing a URL, stopped mangling pre-existing `data:` URLs, and added a quick-and-dirty server that makes it a little easier to experiment.

(The GitHub Action for publishing releases still had Python 3.8 testing in it, so this is a re-do of 0.2 with that removed.)

Added a timeout parameter for freezing a URL, stopped mangling pre-existing `data:` URLs, and added a quick-and-dirty server that makes it a little easier to experiment.

This is a initial proof of concept that will download a URL, assume it is HTML, run it through Beautiful Soup and process every `<link>`, `<script>`, and `<img>` to either include external resources (CSS and JS) or turn them into a data: URL. It also further processes CSS to turn `url(...)` into `url(data:...)`.

This pulls in everything, without timeouts, and fails hard if anything doesn't work as expected. There is a bunch of basic functionality that should be implemented before this is used for anything other than playing around.