Project Gutenberg DRM'd ebook..?

[Chris Jones]

I'm currently reading David Hume's History of England. There are six epub's that I downloaded from Project Gutenberg when I was building my personal library.

Last night, I was checking something in my Kobo Glo's Library->Books display and I noticed that alongside with the version I am reading, I had also uploaded to the device a copy of all six volumes in a single epub. I thought I'd take a look and compare the contents and much to my surprise.. as soon as I tried to access the epub I was greeted with an "Oops! This document couldn't be opened... This document is protected by Adobe Digital Rights Management (DRM) and is not currently authorized.. bla bla bla..." popup on the Kobo Glo..!

Now, this rather came as a surprise since I do not "own" any DRM'd stuff, books or otherwise..!

Well, I fired up Calibre and checked my original download.. Sure enough, I got this epub from Project Gutenberg...! So, I tried to open it with Calibre's built-in viewer and it did take an unusual 30 seconds or so to open, but I was able to access it.

The only potential problem with this epub is that it is quite large -- Mr Hume's history adds up to c. 4,000 pages and there are a number of jpeg's as well. All in all the epub weighs in at a whopping 17.3 MB..

I was able to unzip the epub and browse its contents, so I very much doubt DRM is the issue here..!

Is this a bug with the Kobo viewer software failing to open the epub (not enough RAM, perhaps) and misdiagnosing the cause of the failure..?

Has anybody else run into this problem and possibly found a fix or workaround..?

Oh, in case that matters to anybody, the Kobo's software version is 2.6.1 and was last updated 3 months ago.

Thanks,

CJ

[jiminrussell]

Could you link to the ebook in question? I tried looking for it but all I see is volume 1 with no images or parts A to F with images and I have no issue opening them with my Kobo Aura HD.

[anacreon]

Originally Posted by Chris Jones

I'm currently reading David Hume's History of England. There are six epub's that I downloaded from Project Gutenberg when I was building my personal library.

Last night, I was checking something in my Kobo Glo's Library->Books display and I noticed that alongside with the version I am reading, I had also uploaded to the device a copy of all six volumes in a single epub. I thought I'd take a look and compare the contents and much to my surprise.. as soon as I tried to access the epub I was greeted with an "Oops! This document couldn't be opened... This document is protected by Adobe Digital Rights Management (DRM) and is not currently authorized.. bla bla bla..." popup on the Kobo Glo..!

Now, this rather came as a surprise since I do not "own" any DRM'd stuff, books or otherwise..!

Well, I fired up Calibre and checked my original download.. Sure enough, I got this epub from Project Gutenberg...! So, I tried to open it with Calibre's built-in viewer and it did take an unusual 30 seconds or so to open, but I was able to access it.

The only potential problem with this epub is that it is quite large -- Mr Hume's history adds up to c. 4,000 pages and there are a number of jpeg's as well. All in all the epub weighs in at a whopping 17.3 MB..

I was able to unzip the epub and browse its contents, so I very much doubt DRM is the issue here..!

Is this a bug with the Kobo viewer software failing to open the epub (not enough RAM, perhaps) and misdiagnosing the cause of the failure..?

Has anybody else run into this problem and possibly found a fix or workaround..?

Oh, in case that matters to anybody, the Kobo's software version is 2.6.1 and was last updated 3 months ago.

Thanks,

CJ

It is a Kobo problem with some very big files. I had a similar one with a detailed catalogue of my whole library (27.5 MB), which neither Glo nor Aura HD accepted, saying I wasn't allowed DRMs (though I am), but which worked fine on the Sony PRS-T2.

So I limited the detailed catalogue to only those books on the Aura HD (some 13MB), which the Kobos accept.

FYI, I navigated on the Aura without problem in a 20.1 MB dictionary, and I read a 21.4 MB picture book. So it is not only a problem of size, though it does seem to concern only very big files. I have the latest versions of everything, but I first had the problem some time ago, with I don't remember which versions.

[EDIT] I forgot to mention I thought at first it was a Calibre bug, so submitted one. David Goyal wrote that it was a Kobo problem and I should submit it to them. Considering the very poor automated answers I had with deficient bought books, I gave up.

[Uschiekid]

You're right, it's not actually a DRM-problem. It IS an adobe problem. Occasionally, Adobe gets it into its little computer head that some DRM-less book has DRM. I usually solve it by turning off and on my device, or deleting, restarting and re-adding said book. There are several things that can trigger this, from what I can tell. One being related to library books that DO have DRM (they can cause problems with other books, if opened after expiration). But since you indicate you don't have any such books, I'm guessing it is another issue. Such as it didn't load onto your device properly in the first place. Anyway, for me, one of the two above solutions worked, no matter what the trigger was.

[Chris Jones]

Wow.. cool forum..! I didn't even expect any replies.. and certainly not this quickly..!

Sorry for not replying individually, but thanks for confirming my hunch(es).. I will try switching off the reader as Uschiekid recommends.. and removing the offending epub and uploading a fresh copy just in case.

As to opening a ticket with Kobo.. well.. I think that even though they claim to be more open — i.e. general-purpose — than Nook or Kindle.. I have a feeling they're only interested in selling their e-books and they're not really interested in fixing any bugs that may affect stuff downloaded from elsewhere and didn't pay for. Fine with me.. except if that's the way they look at it, well.. they should let you have the reader for free..!

That's why I am contemplating switching to a 7" android tablet + fbreader or such. A little heavier.. a lot less battery life.. perhaps lower quality font rendering.. but a lot more capable, and that includes debugging and bug-fixing capabilities..!

Thanks,

CJ

[BobC]

Another approach might be to install the Kobo Extended driver in Calibre then upload your book as a kepub meaning that it would use the ACCESS rather than the Adobe engine for display.

You may have a better experience with that.

Another point is that stuff from Gutenberg can usually be improved by re-splitting the book as Gutenberg seems to split on file size rather than chapters meaning that opening and moving within a book can be slow. I merge and re-split all my Gutenberg sourced epubs using Sigil before loading them onto my Glo (along with other tidying up) - it enhances the reading experience as each new chapter will start on a fresh screen.

BobC

[Chris Jones]

Originally Posted by BobC

Another approach might be to install the Kobo Extended driver in Calibre then upload your book as a kepub meaning that it would use the ACCESS rather than the Adobe engine for display.

You may have a better experience with that.

Another point is that stuff from Gutenberg can usually be improved by re-splitting the book as Gutenberg seems to split on file size rather than chapters meaning that opening and moving within a book can be slow. I merge and re-split all my Gutenberg sourced epubs using Sigil before loading them onto my Glo (along with other tidying up) - it enhances the reading experience as each new chapter will start on a fresh screen.

BobC

Ah, interesting.. I'd heard/read about k-epub's and the corresponding Calibre plugin. I'll give that a try if all else fails.

I do have a copy of Sigil installed on another system but I've only used it for a couple of hours to address one very specific need.

I'm a little unclear about what you mean by merging/splitting epubs. When you say "files" do you mean the html files inside the zipped collection..? What I mean is that I use Vim for editing and if I point Vim to a Project Gutenberg epub, I'm first presented with a list of files and directories with the mimetype file right at the top, naturally, and then a directory whose name is a number, presumably the PG reference of the book in their system.. something like 14324/ or 22315/ for instance. This directory contains a number of .htm files (and optionally .jpg's or .png's).

Do you mean you use Sigil to rearrange the contents of these files so that you end up with one .html file per chapter..?

Would that also address this annoying problem I have noticed where for no good reason you occasionally have a page with a bunch of blank lines at the bottom, even though you're not at the end of a chapter.. even not at the end of a paragraph for that matter..?

Is there any hands-on documentation that explains how you would go about doing this merging/splitting..?

Thanks,

CJ

[DNSB]

Originally Posted by Chris Jones

Ah, interesting.. I'd heard/read about this those k-epub's and the corresponding Calibre plugin. I'll give that a try if all else fails.

I do have a copy of Sigil installed on another system but I've only used it for a couple of hours to address one very specific need.

I'm a little unclear about what you mean by merging/splitting epubs. When you say "files" do you mean the html files inside the zipped collection..? What I mean is that I use Vim for editing and if I point Vim to a Project Gutenberg epub, I'm first presented with a list of files and directories with the mimetype file right at the top, naturally, and then a directory whose name is a number, presumably the PG reference of the book in their system.. something like 14324/ or 22315/ for instance. This directory contains a number of .htm files (and optionally .jpg's or .png's).

Do you mean you use Sigil to rearrange the contents of these files so that you end up with one .html file per chapter..?

Would that also address this annoying problem I have noticed where for no good reason you occasionally have a page with a bunch of blank lines at the bottom, even though you're not at the end of a chapter.. even not at the end of a paragraph for that matter..?

Is there any hands-on documentation that explains how you would go about doing this merging/splitting..?

The end of a file is treated as the end of page. So if the chapters are not aligned with the files, you get some rather odd page breaks. Admittedly, Adobe's Reader Mobile renderer can also split pages at some odd spots itself.

You can check Sigil's on-line help at http://web.sigil.googlecode.com/git/...roduction.html. There is also a MobileRead forum which does offer some help.

What I've had to do on several occasions is to select all the chapter files and then right-click and merge. Then go back through and split at the chapter markers. A bit of pain but it does give you a much better looking result than mindlessly splitting files at fixed file sizes which seems to be popular with some publishers. So yes, this does give 1 chapter/1 file.

Unfortunately at this time, Sigil's development is pretty much on life support and Google's changes to it's Google Code are not going to help.

Regards,
David

[BobC]

Originally Posted by DNSB

What I've had to do on several occasions is to select all the chapter files and then right-click and merge. Then go back through and split at the chapter markers. A bit of pain but it does give you a much better looking result than mindlessly splitting files at fixed file sizes which seems to be popular with some publishers. So yes, this does give 1 chapter/1 file.

Yes, that is what I meant - I also take the opportunity to rename the .html files from the ridiculously long names Gutenberg uses to shorter ones.

The only real problem with doing this is that it can mangle internal hyperlinks used for footnotes etc but a bit of judicious examination and the use of regex S&R will sort that out.

There are two ways to split - either manually or again by examining the files and using regex S&R to "insert" the needed split markers before each chapter (if you are lucky they may be h1 or h2 - if someone has done it based on class then it can be a bit more messy) then tell Sigil to "Split on markers". If you have a book with 117 chapters like one I did this week the second way is definitely best.

The reason I use merge then split it to "heal" any already split chapters before performing the chapter split, you could just split the existing stuff then go through manually and merge any "broken" chapters.

I've found Sigil an excellent tool for simply examining and understanding the structure of EPUBS. I used to stay away from them, preferring .fb2 format files which are monolithic in structure but incredibly easy to edit with any old text editor, but needed to get my head round EPUBs when I bought a Glo.

Your early breaking pages may just be a widows and orphans issue. I usually add :

body {widows : 0 ; orphans: 0;}

to my css to help with this.

BobC

[Chris Jones]

Thanks much for all the comments..!

I tried pretty much all the above-recommended methods to no avail.. including Calibre conversions from epub to epub and epub to kepub.. The conversion jobs appeared to be dead in the water — progress bar stuck for something like 40-45 minutes, which didn't sound plausible even for an oversized ebook.. and since it looked as if those conversion jobs were intent on frying my laptop's CPU I decided to play it safe and killed them.

I eventually figured it might be of interest to ask for a second opinion.. so I booted debian unstable and fired up Sigil, pointing it to a copy of the epub. After what felt like ages I saw a popup that basically (can't remember the actual wording..) informed me that the .epub was broken.. but that Sigil was willing to try to fix it at my own risk. Since I had nothing to lose, I did just that and after another fairly long interval.. Sigil was eventually able to display the book. So, I did a "Save As" under a different name.. uploaded this new file to Calibre... proceeded to convert the .epub to a .kepub.. and this time the conversion was successful (mind you, this took about ten minutes.. which is rather surprising on this rather powerful laptop of mine — hey, that's about the time it takes it to build a linux kernel) ...

Finally, I uploaded the book to the Glo' and at long last, I was able to open it.

So it looks like this everything-in-one-file version of Hume's History of England (that suspiciously doesn't seem to be available from PG any longer...) was badly broken in the first place.

Moral of this story from my uneducated perspective is that short of manually rebuilding the epub, if you really have to do something like this (I only went to all this trouble out of curiosity..) your best bet is to use Sigil's built-in epub-fixing routines and hope for the best.

Thanks again for help..!

CJ