Find Obok Failed DeDRM

[bw777]

Hello. Loading epub files downloaded from Kobo into Adobe Digital Editions, then dragging the resultant epub from the Digital Editions folder into Calibre, imports books correctly. However, I started using Obok DeDRM for its convenience, and spot checking the results seemed successful. However, now I've discovered that it's been occasionally failing, leaving me with some sort of encoded book about once in every dozen. So as I find the failed books I have manually replaced those via ADE.

So, first, is there a known issue with Obok failing?

And, second, is there a way to find these files that failed import without having to view every book? I have several thousand books, so searches take some time, and i haven't found a reliable search yet (Quality Check - Check DRM doesn't find them, so i assume Obok did something to them).

Thank you.

[Sirtel]

Try running the Count Pages plugin - it should fail with the DRMed books.

[bw777]

Originally Posted by Sirtel

Try running the Count Pages plugin - it should fail with the DRMed books.

Interesting... it succeeds, and by way of example, for "The Age of Caesar: Five Roman Lives" by Plutarch (one of the books in question) reports 501 pages in the Obok result epub. Via the ADE route, it reports 464 pages. The book actually has 432 pages. It would appear Obok is removing DRM metadata, but failing to decrypt. Count Pages is "succeeding" the same way it "succeeds" with image-only books (i.e. its guesses fail).

The Obok variant retains a Table of Contents.

Any other ideas here?

Thank you.

[Sirtel]

Originally Posted by bw777

Interesting... it succeeds, and by way of example, for "The Age of Caesar: Five Roman Lives" by Plutarch (one of the books in question) reports 501 pages in the Obok result epub. Via the ADE route, it reports 464 pages. The book actually has 432 pages. It would appear Obok is removing DRM metadata, but failing to decrypt. Count Pages is "succeeding" the same way it "succeeds" with image-only books (i.e. its guesses fail).

The Obok variant retains a Table of Contents.

Any other ideas here?

Thank you.

In my experience, the Count Pages plugin is unable to count the pages in a DRMed book. I just tried and it couldn't. If it could with your books, then something else is going on, not merely a failure to remove the DRM.

What happens if you try to convert the book?

[BetterRed]

Originally Posted by bw777

Count Pages is "succeeding" the same way it "succeeds" with image-only books (i.e. its guesses fail).

For image only PDFs (e.g. magazine) and comic formats Count Pages counts the images.

BR

[DNSB]

Originally Posted by bw777

Hello. Loading epub files downloaded from Kobo into Adobe Digital Editions, then dragging the resultant epub from the Digital Editions folder into Calibre, imports books correctly. However, I started using Obok DeDRM for its convenience, and spot checking the results seemed successful. However, now I've discovered that it's been occasionally failing, leaving me with some sort of encoded book about once in every dozen. So as I find the failed books I have manually replaced those via ADE.

So, first, is there a known issue with Obok failing?

And, second, is there a way to find these files that failed import without having to view every book? I have several thousand books, so searches take some time, and i haven't found a reliable search yet (Quality Check - Check DRM doesn't find them, so i assume Obok did something to them).

Thank you.

Did you insert the serial number of your Kobo device into the OBOK customization if you are copying the books from your Kobo device? When I look at my Kobo device or the Kobo App library, about 20% of the books show as having DRM, the other 80% are not DRMmed. For whatever reason, Kobo does not encrypt the ePub3 navigation document, the ePub2 NCX document or the CSS stylesheet(s).

As far as I am aware Quality Check does not detect Kobo's DRM scheme so it's DRM check is not going to work however since the cover is encrypted, you should be able to use Quality Check's Missing covers to detect them.

At one point, I was playing with a utility to check if there was a rights.xml file in the root of the ePub's .zip container but never finished it.

[JSWolf]

You are best to download ePub and if you want to read KePub, use the Kobo Touch Extended driver to convert to KePub. I've not had an ePub fail to have the DRM removed.

Which version of DeDRM are you using?

This link is to the alpha version with the latest code.

https://github.com/noDRM/DeDRM_tools...947bf338d3ffda

[bw777]

Originally Posted by Sirtel

In my experience, the Count Pages plugin is unable to count the pages in a DRMed book. I just tried and it couldn't. If it could with your books, then something else is going on, not merely a failure to remove the DRM.

What happens if you try to convert the book?

Results in a MOBI version with the same encoded content: =���OO��L etc.

Yes, i'm pretty sure the result is not an actual normal DRM-ed file, but some partially-processed-by-Obok DRM-ed file.

[bw777]

Originally Posted by BetterRed

For image only PDFs (e.g. magazine) and comic formats Count Pages counts the images.

BR

Makes sense. Thanks.

[bw777]

Originally Posted by DNSB

Did you insert the serial number of your Kobo device into the OBOK customization if you are copying the books from your Kobo device? When I look at my Kobo device or the Kobo App library, about 20% of the books show as having DRM, the other 80% are not DRMmed. For whatever reason, Kobo does not encrypt the ePub3 navigation document, the ePub2 NCX document or the CSS stylesheet(s).

I don't have a Kobo device, so no serial number. I download them to my PC for reading.

Originally Posted by DNSB

As far as I am aware Quality Check does not detect Kobo's DRM scheme so it's DRM check is not going to work however since the cover is encrypted, you should be able to use Quality Check's Missing covers to detect them.

Well, that doesn't actually work, since i've already gone through and added metadata (incl. cover) to these books.

Originally Posted by DNSB

At one point, I was playing with a utility to check if there was a rights.xml file in the root of the ePub's .zip container but never finished it.

I see. I'm getting curious about how to write plug-ins for calibre. Seems useful in this type of situation.

Thank you!

[bw777]

Originally Posted by JSWolf

You are best to download ePub and if you want to read KePub, use the Kobo Touch Extended driver to convert to KePub. I've not had an ePub fail to have the DRM removed.

Which version of DeDRM are you using?

This link is to the alpha version with the latest code.

https://github.com/noDRM/DeDRM_tools...947bf338d3ffda

10.0.9

[bw777]

Well, I appreciate all the help. It looks like the root problem is somehow resolved; re-downloading the errant books with Obok seems to work.

So, I still need a reliable way to search for books that failed to decode completely, so I know to re-download them. But I've already added metadata and covers to them (just part of my normal process).

Maybe learning how to write a plug-in is the way to go...? Any tips?

[DNSB]

Originally Posted by bw777

I don't have a Kobo device, so no serial number. I download them to my PC for reading.

So you are using the Kobo Desktop app to download the files? One odd issue a while back with one friend of mine's PC and OBOK was that he had his Windows WiFi set to random MAC addresses & change daily so he kept getting new random MAC addresses.