Wikilivres is a wiki similar to Wikisource but which conforms to the more permissive copyright laws of New Zealand. The structure of pages is somewhat similar to that of Wikisource, and so wsexport should be able to support it.
Actually, this wasn't my idea, but I forgot to attribute it correctly when I copied the issue over from Github as part of T255512. The original issue was https://github.com/wsexport/tool/issues/22 by Tpt on Mar 22, 2013.
There's already code in WS Export for connecting to Wikilivres: just use the language code wl or wikilivres. We can't test this at the moment though, because the domain name it links to is wrong (wikilivres.ca) and the new one (wikilivres.org) isn't responding to API requests. Or, actually, working for anything other than the main page.
At the moment, I don't know if there's any way to export books from Wikilives. I would suspect not.
As far as I know, there's no formal relationship between Wikilivres and Wikimedia, but I think its founding idea is to only accept works that are not in the public domain in the US but which are in New Zealand (where it's hosted; it used to be Canada I think, for the same reasons of more permissive IP laws).
Oh, yeah and I meant to add: if the wikilivres site doesn't start working again soon, I think we should just remove the small bit of code that supports it from the tool. It can easily be added back again later if required.
I've made a PR for at least fixing the domain name (.org instead of .ca): https://github.com/wsexport/tool/pull/260
Did you get any feedback from Arthur?
No, nothing yet.
The above patch has been merged, so as soon as wikilivres.org starts serving MediaWiki correctly things should be fine again.
It looks like all requests are now redirecting to the homepage (probably this was the case before but I didn't check):
$ curl -I https://wikilivres.org/wiki/The_Widow_in_the_Bye_Street HTTP/2 301 date: Mon, 30 Nov 2020 10:22:20 GMT content-type: text/html; charset=UTF-8 location: / server: cloudflare […etc.]
That's true, but the site is actually still up, so someone is still paying the bills! It seems like a reasonably small configuration error (maybe with Cloudflare?) is breaking it at the moment, so maybe someone can fix that.