KindleUnpack (MobiUnpack): Extracts text, images and metadata from Kindle/Mobi files

[Hitch]

Originally Posted by tacitus

Yeah, I guess my terminology was inaccurate. Really what I mean (in MS Word terms) is for the "before" paragraph spacing attribute to be honored after a page-break (i.e. at the start of a new chapter).

That's because Word uses pages. There aren't any "pages" in an ebook, not really. There's no top-of-page, and thus...

KF8 seems to handle that just fine, and by the time my next book is ready, I guess it may be time ditch the old format altogether.

I'm really not trying to be argumentative, but it doesn't, actually. To get any type of "top-margin" to work for K8, you need separate HTML files for each chapter and/or section of frontmatter, which then accepts the top of the HTML file as the "top of the page," (essentially--I'm mis-speaking, but I'm trying to explain) and then moves the content to the correct place, e.g., top-margin: 2em or whatever the setting is.

(Does Amazon keep stats on the number of old-format-only devices out there, in the same way Google tracks the different versions of Android?)

No, but we know that there are still literally millions of K-1's, K-2's, DX's and early K3's (that didn't upgrade) out there.

FWIW, this is what Calibre generates for mobi7:

 <p height="0pt" width="0pt" align="center">
  <font size="2">*</font>
 </p>

The * is actually 0xa0 (which is the ASCII version of correspond to   )

Well, as you say it works on the devices, I can't argue--I have to assume that Calibre subsequently parses that and makes it work. If I put that in an ePUB I was going to convert via KG or KP, it would display asterisks with a 2x font-size, not blank paragraphs. I assume that you'll download a copy of the on-sale book ASAP, and check it in real devices, etc.? (Sorry: I don't recall which devices you said you had or don't have).

I assume if you specify KF8 instead, it will generate something different, but I suspect they will try to preserve the spacing in some way. They did a passable job of faking small caps for mobi7 (not good enough for me to want to use, but still...) by capitalizing the text and setting the appropriate font sizes.

Those aren't small-caps; they're Raised Initials, and it's the only sort of faux-caps available for K7. They're extensively used for K7 and in K7 fallback coding for more-advanced books using media-queries.

Best of luck to you on your book sales.

Hitch

[pdurrant]

Originally Posted by Hitch

Those aren't small-caps; they're Raised Initials, and it's the only sort of faux-caps available for K7.

You're using some terminology I don't understand, or you've misunderstood.

SMALLCAPS IS WHAT WAS DESCRIBED, I thought.

Raised Initials is surely something different?

[Hitch]

Originally Posted by pdurrant

You're using some terminology I don't understand, or you've misunderstood.

SMALLCAPS IS WHAT WAS DESCRIBED, I thought.

Raised Initials is surely something different?

PDurrant:

I likely misspoke. I was making an assumption, based upon the fact that he used a K7 mobi generated by Calibre. He said something along the lines of using caps and setting a font-size. What I most commonly see from Calibre-made MOBI K7 files is a Raised Initial in lieu of a Dropcap.

If he meant a faux small-cap...actually, given the extremely limited font-size manipulation for K7, I can't see that working at all. So, I give up on that topic, without actually seeing it (the code/book).

Yes, I concur, what you are displaying as a Raised Initial and Smallcaps are what I mean when I say, "Raised Initial" versus "smallcaps." I vaguely recall trying to force a faux small-cap in K7, upon a client's unrelenting insistence, about....2 years ago, and while it worked "somewhat" in K2 and K3, it didn't work worth dog's bollocks in Kindle for X (anything) and was fairly catastrophic in Kindle for iPad, if memory serves.

Or it seemed to work in one reader, but was unreadably small in the other...really don't recall. As you would know better than almost anyone, when you try to use incremental font-sizes in K7, it simply rounds up or down to the next-available font-size (e.g., small, very small, medium, large, very-large) or one of the 8 available font sizes that already display as font-size options on the device.

So, yes, I stand corrected on this poster's intent; and yes, we are talking about the same things when we use the same words.

Hitch

[tacitus]

Well, as you say it works on the devices, I can't argue--I have to assume that Calibre subsequently parses that and makes it work. If I put that in an ePUB I was going to convert via KG or KP, it would display asterisks with a 2x font-size, not blank paragraphs. I assume that you'll download a copy of the on-sale book ASAP, and check it in real devices, etc.? (Sorry: I don't recall which devices you said you had or don't have).

The HTML was post-Calibre, though pre-KindleGen, and I tested it on a Kindle Touch, the Android Kindle App (Nexus 10), and the web preview. They all looked fine. I'll be asking my friends to report any formatting issues back to me on their devices, so I should hear if anything is too badly amiss. It wouldn't be terrible if I had to change something -- it only affects a couple of pages.

No, but we know that there are still literally millions of K-1's, K-2's, DX's and early K3's (that didn't upgrade) out there.

Nice to know, but it would be very helpful if Amazon would publish actual usage stats. They wouldn't even have to publish numbers, just a piechart showing the proportional share of each device type. At some point, K7 device usage will shrink to the point where it won't be worth maintaining both formats, and it and its quirks will go the way of IE6.

Anyway, I guess we're wandering way off topic here...

[DiapDealer]

Originally Posted by tacitus

Nice to know, but it would be very helpful if Amazon would publish actual usage stats. They wouldn't even have to publish numbers, just a piechart showing the proportional share of each device type. At some point, K7 device usage will shrink to the point where it won't be worth maintaining both formats, and it and its quirks will go the way of IE6.

In my humble opinion, there's really no decision for the ebook creator/designer/author/whatever to make on that subject. So there's really no need to publish stats. Amazon will simply stop delivering the original mobi7 format when it reaches that point of diminishing returns. With that in mind, I'll feel obligated to make sure that any ebook I design will gracefully degrade to the mobi7 format until such time as a user cannot BUY it for a device that needs that format.

In short: if Amazon is still selling my ebook to customers in the mobi7 format, it's my responsibility to see that they've purchased quality, and not just a "let the chips fall where they may" vestigial appendage of the KF8 conversion process.

[Hitch]

Originally Posted by DiapDealer

In my humble opinion, there's really no decision for the ebook creator/designer/author/whatever to make on that subject. So there's really no need to publish stats. Amazon will simply stop delivering the original mobi7 format when it reaches that point of diminishing returns. With that in mind, I'll feel obligated to make sure that any ebook I design will gracefully degrade to the mobi7 format until such time as a user cannot BUY it for a device that needs that format.

In short: if Amazon is still selling my ebook to customers in the mobi7 format, it's my responsibility to see that they've purchased quality, and not just a "let the chips fall where they may" vestigial appendage of the KF8 conversion process.

Not to mention, but:

which book's "usage stats" would they provide? A bestseller by Dan Brown, or a romance novel? I imagine that the stats vary wildly. Amazon would have to take an aggregate figure, over some time period, and put out the stats.

To me, it also does not matter; if even a mere two people are going to buy a client's book in K7, they deserve as much attention as the others who are buying it in K8. I've mentioned here on MR before that I am a little disturbed at a fairly cavalier dismissal (this is not directed at Tacitus) of the K7 users and the formatting needed for graceful degradation for those devices. I'm pretty constantly surprised that the vast majority of my clients (now numbering over 2,000, which amazes me) have K2 devices or original or 2nd-Gen Nooks. So...if we take that as anything to go by...

Hitch

[DiapDealer]

Originally Posted by Hitch

To me, it also does not matter; if even a mere two people are going to buy a client's book in K7, they deserve as much attention as the others who are buying it in K8. I've mentioned here on MR before that I am a little disturbed at a fairly cavalier dismissal (this is not directed at Tacitus) of the K7 users and the formatting needed for graceful degradation for those devices.

Oh absolutely. I feel the same way (and I too, would like to make it clear that I'm not targeting Tacitus--who was merely asking questions).

To me, it's quite simple really: the two people who buy your client's book in the K7 format are paying the exact same price as those who buy it in the KF8 format. If it's my work, I OWE those two people the same quality experience (or as close as I can possibly get to it) that the others are getting. That's it.

[nva1991]

Hi!
Thanks for the great plugin. I have unpacked some PDFs from AZW4 files
But today, when i tried with a file >300Mb, the plugin did not work!
Here is the error:
calibre, version 1.2.0
ERROR: Unhandled exception: <b>Exception</b>:Problem locating unpacked pdf.

calibre 1.2 Portable isfrozen: True is64bit: False
Windows-7-6.1.7601-SP1 Windows ('32bit', 'WindowsPE')
('Windows', '7', '6.1.7601')
Python 2.7.4
Windows: ('7', '6.1.7601', 'SP1', 'Multiprocessor Free')
Traceback (most recent call last):
  File "calibre_plugins.kindleunpack_plugin.extraction", line 225, in print_replica
  File "calibre_plugins.kindleunpack_plugin.utilities", line 275, in getPDFFile
Exception: Problem locating unpacked pdf.

Could you please tell me what is the problem? And how can i fix it? If you need, i will send you the AZW4 file via PM.
I use Windows 7 64 bit, 8Gb RAM.
Thank you!

[DiapDealer]

This might be a plugin-only issue ... in which case, there's a separate thread in the calibre forum for it. We'll see. I'm not sure memory management was ever a very big priority with KindleUnpack, so the huge filesize itself could definitely be the culprit.

I suggest trying the standalone KindleUpack program available in the first post of this thread. If that successfully unpacks the AZW4, then clearly I need to see what the plugin is doing wrong/different.

Let's go from there. I frankly don't want to try and deal with a 300+Mb test file.

Also, try just unpacking the AZW4 to an external folder (with the plugin) rather than "Extracting" the PDF. See what shows up in that folder.

[nva1991]

I have figured it out!
I downloaded KindleUnpack_v62.zip, then run the .pyw file (with Active Python installed), it successfully unpack the PDF file!
Thank you so much!
I have another problem that:
When i try to remove DRM from [print replica] book, for books >400 Mb, my program (EPUBEE) does not works. It asks the Kindle Serial Number, but i do not have Kindle, i download them on PC. Books <400Mb can be decrypted normally!
Opening books >400Mb by kindle for PC is a pain, it loads very slow, sometimes hangs the computer.
Is there any python script to remove DRM from [print replica] without using Callibre?

[pdurrant]

Originally Posted by nva1991

When i try to remove DRM from [print replica] book, for books >400 Mb, my program (EPUBEE) does not works. It asks the Kindle Serial Number, but i do not have Kindle, i download them on PC. Books <400Mb can be decrypted normally!
Opening books >400Mb by kindle for PC is a pain, it loads very slow, sometimes hangs the computer.
Is there any python script to remove DRM from [print replica] without using Callibre?

Detailed help with DRM removal is not permitted at MobileRead.

As you've been fooled into paying for a DeDRM program, you might at least try to get support from them.

Otherwise, I suggest trying the latest (free!) tools from Apprentice Alf.

[nva1991]

Thank you!
Is anyone successful in decrypt KNO textbook? It is a PDF file but the common program can not decrypt it...
I have been waiting for so long but still no program can do it!

[pdurrant]

Originally Posted by nva1991

Thank you!
Is anyone successful in decrypt KNO textbook? It is a PDF file but the common program can not decrypt it...
I have been waiting for so long but still no program can do it!

As I mentioned before, detailed help with DRM removal may not be given on MobileRead. I suggest you ask elsewhere.

[nva1991]

But is there anyone successful?

[DiapDealer]

Originally Posted by nva1991

But is there anyone successful?

That's not relevant here. KindleUnpack is not a DRM removal tool. Even if this discussion were within the bounds of Mobile Read rules, this thread would not be the place to ask your question.