[BensonBear]

Originally Posted by aliali47

I didn't think bookshelves would work with pdfs because some features work on epubs, like highlighting and note taking, and not on pdfs. Alas, I can't get the shelves working on either.

Yes, but there is an excuse for why he highlighting doesn't work on pdfs. It would require different code since the logic of rendering is totally different for pdf and epub. But it should not require different code to add the books to a shelf, since one is just dealing with the name of the book, which is the same kind of thing in both cases. No real excuse here (except, again, the shelves are not being offered to us by Kobo at this point).

I assumed that shelves were a way to organise files so that you could search for a file in a group rather than from a list of all files on the KT. I have tried the update but it doesn't work that way on my KT. I open the shelves and there is one untitled shelf that says '6 books'. I can edit, rename, delete, but the name I give it doesn't appear after I save it. I can add/remove items from the shelf and save the changes, but it doesn't remember the changes.

I could add epubs fine, just not pdfs.

When I open the shelf, it gives me the same menu as the library. Perhaps because it is the only shelf on the KT. I was hoping for a 'create shelf' option, and to be able to add files to it but I guess that is what you mean by 'trivial cases'?

If you go to home, then choose library, there should be a create shelves option in the menu you get. It is not there if you go to bookshelves from home. Kobo will probably arrange the menus better when it finally adds shelves.

Yes, that is what I meant. But I don't know how to run a shell script in calibre's epub meta. I tried in the past to edit metadata in calibre so my files are not called things like 'template.p65' but the KT didn't recognise the changes that I thought I had made correctly. I don't want to spend time manually changing file data, even for 50 files that is a lot of time.

Calibre has a command line program "ebook-meta". If you say

ebook-meta -t "The title" -a "The author" filename

it will set the title and author accordingly. Now if you can just strip that data out of the filename, you can set the metadata automatically. For example, I extract the title from various all my converted SEP html files and stuff them into the epubs with the one line

for $x in `echo *`; do ebook-meta $x/$x.epub -t "`sed -nr 's/<title>(.*)<.title>/\1/p' $x/index.html`"; done

If you tell me the format of the file name containing the info, and maybe give me a list of filenames for testing, I can probably whip you up a command in a few minutes that will work on your entire collection. (Which reminds me: before you mess around with shelves very much, I would recommend backing up your sqlite database )

Because of the poor organising options, I can only use the KT for leisure reading and not work.

Well you can read a document for work on it. I don't see why one needs organising options just to read a book.
That is something at least.

When I heard about a device that ran a version of linux, I hoped that it would have the capabilities to be a researcher's companion (or be more open for programmers to be creative).

Yeah, bummer. But, it is no different from your desktop if it runs Linux, and which you bought a proprietary program for. There is no need to give the source of that program, or plugin capacity, or anything, and that is what Kobo has done. It is an argument that Stallman and others have for not allowing the so called Lesser GNU Public License but that horse has left the barn long ago.

Reading pdfs compared to epubs is tedious, yes I think fixing it is a priority. But I would like to be able to find pdfs as well as read them easily!

Let us collect together a list of people who want pdf reading to be as easy as epub. THIS IS EASY TO ADD! Also, the search I mentioned would be easy to add as well.

Amazon offer a conversion service for Kindle, which I hear works very well on text only pdf articles, which is what I mostly have.

I highly doubt it! I will believe it when (if) I see it.

I have very little programming knowledge but I assumed that nearly all of the above mentioned features would be straightforward to add. It is very frustrating.

Decent pdf reading is EASY TO ADD. The search I mentioned is EASY TO ADD. Decent "shelves" would be a little more work but still pretty easy. The hypertextual collection of excerpts a little harder still, but still not too hard.

But only if one has the source.