Thanks for your feedback !

That's an intentional limitation of the free version of the JPedal library Ubooquity uses to render PDF books (comics are rendered by a different one which just extract images without rendering anything else). Despite a very long search, I was not able to find a free Java library that was capable of correctly rendering all PDF files, so I settled for this one.

I have plans to try to use Mozilla's PDF.js library to do PDF books rendering. It's quite good and would greatly improve quality. But I don't know yet when I'll integrate it (and if it will work).

Yes, you can fix that by editing the theme you use (or create a new one if you use the default theme).
Add the following lines in the comics.css file:

.thumb img
{
    max-width: [[COMIC_W]]px;
    max-height: [[COMIC_H]]px;
}

I'll add it by default in the next release.

I have been able to reproduce the problem with the image you provided, I'll investigate.
It does not seem new though, as I had the same result with version 1.7.6.

Your first error is a different one, I'll need the image file to investigate this one too.
If you want you can send it to me using the "private message" option (just under the search field).

You have to set a regular expression (in the "Scan exclusion pattern" option) that will determine which files and folders will be ignored.
The expression is tested against the whole path of each scanned file and directory.
I usually use this tool to quickly test regexp when I need one.

Done.

Good. :)

Did you try to do a force refresh in your browser (usually shift + "refresh" button), or to clear your cache ?