kepubify 2.0 - A standalone tool to convert ePubs to Kepubs

[konstantinus]

Originally Posted by geek1011

The advantage is you don't have to connect multiple times.

In order not to do this, I will have to reboot the device. Doubtful benefit. IMHO
P.S. It seemed to me that the process has become much faster.

[geek1011]

Originally Posted by konstantinus

In order not to do this, I will have to reboot the device. Doubtful benefit. IMHO
P.S. It seemed to me that the process has become much faster.

The benefit I had from doing it was: the intermittent connection issues on the AuraE2v1, my automated scripts (it's easier to do the same thing every time than to switch each time it's connected), the speed improvements you noticed, and the ability to do more complex things with the metadata (I have a custom build to be able to add metadata ahead of time for when I don't transfer all my books at once). In addition, the code is a bit simpler this way.

And yes, it is a bit faster in this version because of the triggers and improved metadata scanning.

I might be able to patch nickel to re-read the series metadata on import, but I'm not totally certain if it's possible yet.

[rtiangha]

Originally Posted by geek1011

I might be able to patch nickel to re-read the series metadata on import, but I'm not totally certain if it's possible yet.

Now THAT would be awesome. Or a standalone daemon that monitors nickel for when it does its import then kicks in using the same logic nickel does to determine which books are new and rereads them could work too.

One thing I noticed is that seriesmeta will scan through hidden directories. I know that the logic changed in newer firmwares for nickel to read through hidden folders too, but would it be possible to have a command switch to enable or disable that ability in seriesmeta? It seems inefficient to do if someone is storing epubs in hidden folders to be used in Plato or koreader never intending for it to be used in nickel (in my case, it's problematic epubs that choke nickel, like some Delphi Classics collections).

[konstantinus]

Originally Posted by geek1011

The benefit I had from doing it was: the intermittent connection issues on the AuraE2v1, my automated scripts (it's easier to do the same thing every time than to switch each time it's connected), the speed improvements you noticed, and the ability to do more complex things with the metadata (I have a custom build to be able to add metadata ahead of time for when I don't transfer all my books at once). In addition, the code is a bit simpler this way.

And yes, it is a bit faster in this version because of the triggers and improved metadata scanning.

I might be able to patch nickel to re-read the series metadata on import, but I'm not totally certain if it's possible yet.

I see. Thanks. So far, the need to reconnect / reboot and complete scanning of all books each time make seriesmeta an interesting but time-consuming program, unfortunately. For me, naturally.

[geek1011]

Originally Posted by konstantinus

I see. Thanks. So far, the need to reconnect / reboot and complete scanning of all books each time make seriesmeta an interesting but time-consuming program, unfortunately. For me, naturally.

No problem. Also keep in mind that you don't have to reboot right away unless you absolutely need the series metadata before your next reboot (it won't cause any issues if you do other stuff in between or run seriesmeta again). Out of curiosity, how large are your books, how many do you have, and how long does it take to scan? I have around 600 on my device at once, and it takes ~30 seconds to scan everything.

[konstantinus]

Originally Posted by geek1011

Out of curiosity, how large are your books, how many do you have, and how long does it take to scan?

I can’t answer now about the time of the scan - I cleaned the device, the number of books - in different ways, from 200 to 2000.

[rtiangha]

Originally Posted by konstantinus

I can’t answer now about the time of the scan - I cleaned the device, the number of books - in different ways, from 200 to 2000.

I've got 12,000 and a 2.2GB Kobo database running on a 32-bit Intel Clovertrail Atom machine and it takes forever, lol. It's a special CPU that executes slightly differently, I believe, and most of it seems to be CPU bound so that probably explains it.

[pat34]

Hi,

Just a simple question.
There is a way to have the .kebub extension (not .kebub.epub) ? I know, i can write a script to convert and rename ...

Thanks.

[davidfor]

Originally Posted by pat34

Hi,

Just a simple question.
There is a way to have the .kebub extension (not .kebub.epub) ? I know, i can write a script to convert and rename ...

Another simple question:

Why? Unless you are going to add the kepub to calibre, nothing will know what to do with the books.

[pat34]

Originally Posted by davidfor

Another simple question:

Why? Unless you are going to add the kepub to calibre, nothing will know what to do with the books.

I just want to try the kepubify one, before to import in my calibre database.

[geek1011]

Originally Posted by pat34

I just want to try the kepubify one, before to import in my calibre database.

I don't know what you're asking, but I you want to try kepubify, drag the folder with your books onto kepubify.exe, then copy the folder name_converted to your Kobo.

[pat34]

Originally Posted by geek1011

I don't know what you're asking, but I you want to try kepubify, drag the folder with your books onto kepubify.exe, then copy the folder name_converted to your Kobo.

Ok, i know about the d&d process to convert directories ...
I just want kepubify generate a file myfile.kepub in place of myfile.kepub.epub
maybe i didn't see a command line option ?

[geek1011]

Originally Posted by pat34

Ok, i know about the d&d process to convert directories ...
I just want kepubify generate a file myfile.kepub in place of myfile.kepub.epub
maybe i didn't see a command line option ?

Oh, I see. I think I'm going to add this feature for the next version, as it has been a frequent request.

[rashkae]

Originally Posted by geek1011

Oh, I see. I think I'm going to add this feature for the next version, as it has been a frequent request.

I can see the benefit of this, for people who want both 'pure' original epub and upload with Calibre. You can import the books in Calibre and send to Kobo with the Extended driver, and get much the same results. However, this does take time. Barely noticeable for 1 or 2 average size books, but significant if you are trying to send 100's at a time.

The solution to this is to keep the kepub version directly in the library... But Calibre itself can't do this without milting the epubs through the full conversion process, which will irrevocably change the actual formatting.

So an option in your standalone converter to make the files import into Calibre can be of great usability for some folk.

[pat34]

Originally Posted by geek1011

Oh, I see. I think I'm going to add this feature for the next version, as it has been a frequent request.

So, wait and see