[Device Plugin] Kobo Touch Extended Driver

[dcamp]

I convert all my ebook in kepub.
I open an ebook (kepub) and I read some pages.
Then I open another ebook (kepub) and I read some pages
I return on the first ebook (kepub) but it automaticaly open on different page and not on the last read.
Sorry for my english.
Diego
Italy

[jgoguen]

Originally Posted by dcamp

I convert all my ebook in kepub.
I open an ebook (kepub) and I read some pages.
Then I open another ebook (kepub) and I read some pages
I return on the first ebook (kepub) but it automaticaly open on different page and not on the last read.
Sorry for my english.
Diego
Italy

This is a well-known issue, mentioned by a few people already. I've highlighed the relevant portion of my first post in red below.

Originally Posted by jgoguen

Known Issues
Not everything works perfectly all the time. Here are some of the issues I know about that are likely to stick around for a while.

• Bookmarks, annotations, highlighting, and anything else that's highly dependent on your current position may work fine, may have slight issues, or may not work very well.
• The hyphenation feature may cause the Kobo device to fail to properly recognize words when you tap and hold a word to search for it in the dictionary. This may manifest as the device only selecting part of a word, in which case you may drag the selection to encompass the entire word as a workaround. It may also manifest as the Kobo saying that "word" could not be found and the closest match is "word".

[bookwurm70]

Originally Posted by davidfor

This, and shelves, is part of the main Kobo driver.



Have a look at https://www.mobileread.com/forums/sho...d.php?t=193184 for how to do shelves. If there is still a problem, ask away.

Thanks, I think the above link s what I needed. I didn't have the # in front of my custom field to get the shelf information to work. Also had the update set to manual. All fixed now

[bookwurm70]

Thanks for everyone who helped me set up my Aura! The Kobo boards are awesome. So very helpful. I appreciate all the work everyone has done to make Kobo readers even better than the Kobo engineers ever imagined they could be

[Fursty Ferret]

Hi jgoguen - do you know why the Kobo struggles with bookmarks in converted kepubs?

[jgoguen]

Originally Posted by Fursty Ferret

Hi jgoguen - do you know why the Kobo struggles with bookmarks in converted kepubs?

Yes, it's because Kobo kepubs come with an extra JavaScript file that I can't distribute or add myself. One of the functions in that JavaScript file is for determining the current position.

There is also a possibility that I need to add additional kobo tags to the generated kepubs. When I get some quiet time I'm going to take apart one of the free books they provide with the Kobo Glo and compare their internal structure with mine again. This won't affect annotations and such things, but if other peoples' suspicions are right it should help the bookmarks and remembering where you left a book.

The other thing that throws it off is, if you have it enabled, the hyphenation functionality. Hyphenation inserts the "soft hyphen" character at the appropriate places in each word, but the definition of a "soft hyphen" (a hyphen which is only displayed if it is the last character on a line, otherwise not displayed) causes problems for character parsing code that doesn't expect them to be present. For example, the English word "image" in the middle of a line is displayed as "image" but is represented as "im-age" internally.

Adding soft hyphens to a kepub also causes display issues (most likely because the renderer counts the soft hyphens as characters even when it shouldn't) which I'm hoping to try and address for the next release.

[PeterT]

One thing I have wondered.... Have you considered having people download a "free" Kobo eBook, and then having the plugin extract the js file from it, and add it into the books you convert?

That way you are not distributing the file, you are merely providing a tool that permits the user to extract the file and add it to their books?

[jgoguen]

I don't see a fundamental difference between that and having the JS file itself somewhere. I'll read the copyright law again, maybe I can do that. Could you open a Github bug for this, asking to investigate if it's allowed to have the plugin pick up a file for inclusion? I'm on my phone and the interface isn't as nice for phones

EDIT: Alright, I'll give it a shot based on sections 29.21 and 29.22 of the Copyright Act.

[buffaloseven]

Would writing a whole new JS to duplicate functionality through a different method be legal? I have a ton of JS under my belt and would be willing to take a look at something like that if it would get around the problem.

EDIT: Reading some more, it does sound like PeterT's solution could work. Then you're not distributing the file and the law moves into personal-use areas.

[jgoguen]

That's my thought as well.

To answer your question though, I'm fairly certain that writing a new JS file would be OK as long a whoever writes it has never seen the original file and doesn't reference Kobo materials. A "clean room" or "black box" implementation.

[samhy]

Originally Posted by jgoguen

New hyphenation idea was a bust, turns out it didn't play nicely with languages using non-Latin characters. So, it's back to the drawing board to see if I can make the current hyphenation code play nicely.

I was going to report my experience with a French book
It's working but not always following the hyphenation rules (e.g. cutting the words at the wrong place). However, it was nice to try it

[Thalinor]

Just wanted to say thank you for your work on this project. Everything is working great with the plugin accept the known bug where it does not remember the correct page.

Also, for some reason a few books have missing chapters; for some reason when the Glo uses the kepub software to display an epub, they don't show up. If I change the extension back to just .epub the chapters are there and show correctly. This is another known bug as I have read about it here on the forums, but the "fix" someone posted was cryptic at best.

FYI: I have played around with converting the original epubs to remove margins, etc as well as playing around with the internal margin settings on the Glo; unfortunately nothing fixed the skipping pages. Skipping pages (while slightly annoying) is not a huge deal as I just use the internal bookmark feature on the glo by pressing the upper right hand corner of the screen, but missing chapters really makes it hard to read books.

Anyway, again thanks for your hard work.

[buffaloseven]

So I'm a little confused about the new addition to extract the *.js file...

I just put reference.kepub.epub into the directory before I compile the zip, or is there a Calibre folder I put it into after the plugin has been installed?

And does it work?

[jgoguen]

Originally Posted by Thalinor

his is another known bug as I have read about it here on the forums, but the "fix" someone posted was cryptic at best.

Do you have a link to this fix?

Originally Posted by buffaloseven

So I'm a little confused about the new addition to extract the *.js file...

I just put reference.kepub.epub into the directory before I compile the zip, or is there a Calibre folder I put it into after the plugin has been installed?

Watch for an update to the README file in the next few hours. Once I get home I'll update it. I had this working and tested over the weekend, then the Copy.com service decided on Monday morning to revert everything to the way it was Thursday night. Basically you drop reference.kepub.epub in the KoboTouchExtended configuration directory.

Originally Posted by buffaloseven

And does it work?

Well, that depends. Do you mean "does the plugin do what it's supposed to and always put the Kobo JS file into the books and reference it properly"? If so, then yes it works every time all the time. If you mean "does it always make the book open at the correct spot and make my annotations and bookmarks stay in the right place", then yes, sometimes, but not always. There's more work for that unrelated to the JavaScript file. I'm still working on getting text split correctly. It's greatly improved based on the books I have, but not perfect. I commonly saw differences of 5+ pages before, now I haven't seen more than a page difference and the extra tags I know I need would probably address that.

[Xenophon]

Originally Posted by jgoguen

[SNIP]

This plugin will not work with the Aura HD, if you want to use the Aura HD (as far as I'm aware, which is admittedly very poorly informed) you will need to leave the KoboTouch driver enabled. I don't have an Aura to test with, nor will I be purchasing one in the foreseeable future. Anyone who's interested in Aura support and has Python experience could add the Aura as a recognized device and see what needs to change and submit patches, or anyone with a lot of extra money sitting around can donate an Aura.
[SNIP]

I recently broke down and ordered an Aura. I'm not prepared to just donate an Aura, but I'd happily contribute some $$ towards an Aura for the devs to work with. Way back when, we used ChipIn to raise money for a prs500 for Kovid. ChipIn is gone now, but if someone suggests a reasonable crowdfunding alternative, I'd happily contribute. I suspect that other Aura owners would probably do the same.

Xenophon