I am entirely dependent on JUnrar to extract files from RAR archives. And as you have noticed, JUnrar is not maintained anymore (and I couldn't find any alternative).

So there is not much I can do, apart from fixing it myself (which I won't do) or advise you to convert your files to a more sensible format, like cbz.

I opened your file with an hex editor: there is no character (BOM or other) before the usual XML content.

I also launched Ubooquity using your file, without triggering any error.

So I still have no idea of the cause of the problem.

Ok, just found it: the problem comes from the "#" character in your folder name.

I don't know why (problem is inside Jaxb, the internal XML library used by Java), but I was able to reproduce the problem that way.

Just remove the "#" and it will work fine

Yes, sorry.

Sumatra is written in a completely different language, so it wouldn't help, but thanks. :)

As for the work required for the modification: it's done in Javascript (not Java), a language I'm not an expert in and which is one of the most painful to use (but when you write code for browser you don't have a choice).

Breaking things written in Javascript is quite easy, that's why even simple features take time when they are done in this language (although Angular helps a bit, for those who are familiar with the Javascript ecosystem).

Reader settings are supposed to be stored in a cookie, so it should stay on "fit to height" once you've selected it.

It's not working ?

Already on my todo list.

Not at the top, but I'll do it eventually since I'd like to have this feature for myself too.

In any case, you're right about the meaning of the error: there is some unauthorized character at the beginning of the file, before the XML content

So you can either open the file with an hexadecimal editor to take a look, or you can send it to me (tom at vaemendis dot net) and I'll take a look.

Ubooquity is not supposed to write a BOM at the beginning of the preferences file.

Are you positive there is one in yours ?

No, Ubooquity is focused on collection browsing.

Updating said collection is out of its scope.

About documentation and a full client (something that have been on my mind too but that I won't start before several years, if ever): the best way to programmatically get data from Ubooquity is through the OPDS feed (explained in this post).

Ubooquity implements the OPDS specification, and you'll find everything you need (metadata, covers, files, links to extract individual pages) in the feed.

The individual links for page extraction is an extension of the OPDS format described in this document.