If I fill in the metadata for a book in Libprs500, save the book to disk and then add the LRF file to a Libprs500 database on another computer, the following fields of metadata seem to be lost (though they are in the opf file which is saved at the same time as the LRF file):

Rating
Tags
Series
Book number (in series)
ISBN
Comments
Book cover

Is this what should happen? If so, is there a way for the metadata to be picked up from the opf file?

When I implement the recursive import feature it will pick up metadata from the opf file as well.

So does that mean those fields don't form part of the embedded LRF metadata, they are extra fields held only in the Libprs500 database (but accessible through the opf file)?

yes .

Thanks.

On a related note, is there a way of managing collections on the reader via a libprs500 command line?

No there isn't

is there a way of managing collections on the reader via a libprs500 command line?
Well, You can try this Perl script (http://sourceforge.net/projects/prs505media/) - chances are it will work alongside libprs500.

Any chance this Perl script could maybe be rewritten in Python?

Any chance this Perl script could maybe be rewritten in Python? Well, I guess there is no point in such exercise per se - this script just does what it should, that is tries to maintain relation between disk directory structure and collection names. As Perl is (more or less) as cross-platform as Python there is no need for porting.
If you want such (or similar) functionality directly in libprs500 - I guess just open a ticket on the site.

I can see how author-sort works in Libprs500. I can also see that Libprs500 sets the author.reading element in the files metadata.

What I cannot understand is what use the reader (prs-505) makes of this information - it doesn't seem to use it which means that the author name really needs to go onto the reader in lastname, firstname order. Can anyone enlighten me?