KindleUnpack (MobiUnpack): Extracts text, images and metadata from Kindle/Mobi files - Page 43

DiapDealer · 11-08-2013, 03:06 PM

I would suggest a quick and dirty test to make sure that changing the KF8 EXTH data in the hybrid mobi/KF8 even has the desired effect on Kindle devices (when unsplit).

With regard to the PDOC/EBOK cdetype (EXTH 501), there would need to not only be a way to change the record but also to add the record entirely. Because so far as I can tell, Kindlegen/Previewer doesn't create the 501 EXTH record in the output it creates in the first place. That record doesn't get added until delivery from Amazon to a user's device.

Hitch · 11-08-2013, 03:19 PM

Quote:

Originally Posted by DiapDealer

I would suggest a quick and dirty test to make sure that changing the KF8 EXTH data in the hybrid mobi/KF8 even has the desired effect on Kindle devices (when unsplit).

With regard to the PDOC/EBOK cdetype (EXTH 501), there would need to not only be a way to change the record but also to add the record entirely. Because so far as I can tell, Kindlegen/Previewer doesn't create the 501 EXTH record in the output it creates in the first place. That record doesn't get added until delivery from Amazon to a user's device.

Diap:

Clarify for me: why are we doing this in the first place? I don't mean to be dense, and admittedly just skimmed over most of this, but...what's the point?

Hitch

DiapDealer · 11-08-2013, 03:20 PM

Quote:

Originally Posted by AcidWeb

EDIT:
In files created by KindleGen all three are empty.

Correction: in files created by Kindlegen, all three records are non-existent.

DiapDealer · 11-08-2013, 03:30 PM

Quote:

Originally Posted by Hitch

Diap:

Clarify for me: why are we doing this in the first place? I don't mean to be dense, and admittedly just skimmed over most of this, but...what's the point?

Hitch

I think we're away from Kindle(Mobi)Unpack now and discussing a separate (and new) tool. My understanding is that it would simply make personal, Kindlegen created output not look like personal, created content (i.e ugly cover) on certain Kindle devices.

Basically, a meta-data tweaker capable of modifying specific KF8 EXTH records (which would be in the second header after the MOBI7 header) in Kindlegen's dual-format output.

Hitch · 11-08-2013, 04:01 PM

Quote:

Originally Posted by DiapDealer

I think we're away from Kindle(Mobi)Unpack now and discussing a separate (and new) tool. My understanding is that it would simply make personal, Kindlegen created output not look like personal, created content (i.e ugly cover) on certain Kindle devices.

Basically, a meta-data tweaker capable of modifying specific KF8 EXTH records (which would be in the second header after the MOBI7 header) in Kindlegen's dual-format output.

Ooooooooooooohkay. I don't think I've ever used the PDS, so I suspect I don't "get" the idea, or the purpose. This is all to save something from having an ugly cover on the bookshelf--is that right? I cannot think of another distinction, other than the destination folder, for a PDOC over a book?

Hitch

DiapDealer · 11-08-2013, 04:05 PM

Quote:

Originally Posted by Hitch

This is all to save something from having an ugly cover on the bookshelf--is that right?

That is my understanding. Yes.

AcidWeb · 11-08-2013, 04:06 PM

Quote:

Originally Posted by DiapDealer

That is my understanding. Yes.

Yep. That is correct.

DiapDealer · 11-08-2013, 04:14 PM

We should probably get a new thread going for this discussion and/or proposed tool.

KevinH · 11-08-2013, 06:10 PM

Hi,

Okay created a "Need for a dual mobi metadata tweaker" thread to move this discussion to.

Kevin

DiapDealer · 11-17-2013, 10:09 AM

Hey, does anybody remember the script that could take the output from mobiunpack (or mobi2html, I forget) and process all the inline styling and put it into a simple external style file and output the cleaned-up html?

I remember it vividly (and remember using it), but I can't find it anywhere, or remember what it was called.

EDIT: Well, after a very nostalgic romp through the 3 year comments section of a certain "New Blog" (alliances made, friendships forged, achievement unlocked), I found the name of the script: "mobiml2xhtml.py." Now to see if I can find a copy on local storage or online.

DaleDe · 12-18-2013, 10:02 PM

has anyone taken a look at the AZK format that is now being used on the iOS devices and readers from Amazon? The Previewer 2.9.2 will make one of those if you select an iOS device.

Dale

Hitch · 12-19-2013, 04:24 AM

Quote:

Originally Posted by DaleDe

has anyone taken a look at the AZK format that is now being used on the iOS devices and readers from Amazon? The Previewer 2.9.2 will make one of those if you select an iOS device.

Dale

Grrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr rrrrrrrrrrrrrrrrrrrrrrrrrr.

Yes. I have. Don't get me started on that bloody thing.

Hitch

pdurrant · 12-19-2013, 05:02 AM

Quote:

Originally Posted by Hitch

Grrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr rrrrrrrrrrrrrrrrrrrrrrrrrr.

Yes. I have. Don't get me started on that bloody thing.

Hitch

Ummm.. but has anyone looked at the internals? Presumably it's another variant on the Palm Database/Mobipocket format?

(I don't have time at the moment. Perhaps in the New Year.)

Hitch · 12-19-2013, 05:18 AM

Quote:

Originally Posted by pdurrant

Ummm.. but has anyone looked at the internals? Presumably it's another variant on the Palm Database/Mobipocket format?

(I don't have time at the moment. Perhaps in the New Year.)

I haven't dug that far into it. Ditto on the New Year. I'm just irritated about the whole "other format" thing, it's ludicrous. I have yet to see any difference between the mobis we make, on the iPad reader (K4iPad) and the AZK files. It just seems like a lot of extra steps for very little gain.

Hitch

KevinH · 12-19-2013, 05:31 PM

Hi,
I have a bit. It is a zip archive. Inside you will find gzipped json pieces and scripts and something that looks like the skeleton and fragments found in KF8 mobi internals. Basically the skeletons are the file frameworks and the fragments are json objects. The proper fragment for where the user is reading seems to be loaded on demand using a javascript.

So it appears to be an html5 based web application container. Try renaming the .azk to .zip, unzipping it and then gunzipping the gzipped json object files. Looks like fun!

I gave up at that point as I simply did not care enough about things. I am sure it is related to the skeletons and fragments found in the KF8.

11-17-2013, 10:09 AM	#640
DiapDealer Grand Sorcerer Posts: 28,040 Karma: 199464182 Join Date: Jan 2010 Device: Nexus 7, Kindle Fire HD	Hey, does anybody remember the script that could take the output from mobiunpack (or mobi2html, I forget) and process all the inline styling and put it into a simple external style file and output the cleaned-up html? I remember it vividly (and remember using it), but I can't find it anywhere, or remember what it was called. EDIT: Well, after a very nostalgic romp through the 3 year comments section of a certain "New Blog" (alliances made, friendships forged, achievement unlocked), I found the name of the script: "mobiml2xhtml.py." Now to see if I can find a copy on local storage or online. Last edited by DiapDealer; 11-17-2013 at 10:50 AM.

12-19-2013, 05:31 PM	#645
KevinH Sigil Developer Posts: 8,160 Karma: 5450818 Join Date: Nov 2009 Device: many	Hi, I have a bit. It is a zip archive. Inside you will find gzipped json pieces and scripts and something that looks like the skeleton and fragments found in KF8 mobi internals. Basically the skeletons are the file frameworks and the fragments are json objects. The proper fragment for where the user is reading seems to be loaded on demand using a javascript. So it appears to be an html5 based web application container. Try renaming the .azk to .zip, unzipping it and then gunzipping the gzipped json object files. Looks like fun! I gave up at that point as I simply did not care enough about things. I am sure it is related to the skeletons and fragments found in the KF8. Last edited by KevinH; 12-19-2013 at 05:33 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Can i rotate text and insert images in Mobi and EPUB?	JanGLi	Kindle Formats	5	02-02-2013 05:16 PM
PDF to Mobi with text and images	pocketsprocket	Kindle Formats	7	05-21-2012 08:06 AM
Mobi files - images	DWC	Introduce Yourself	5	07-06-2011 02:43 AM
pdf to mobi... creating images rather than text	Dumhed	Calibre	5	11-06-2010 01:08 PM
Transfer of images on text files	anirudh215	PDF	2	06-22-2009 10:28 AM

11-08-2013, 03:06 PM	#631
DiapDealer Grand Sorcerer Posts: 28,040 Karma: 199464182 Join Date: Jan 2010 Device: Nexus 7, Kindle Fire HD	I would suggest a quick and dirty test to make sure that changing the KF8 EXTH data in the hybrid mobi/KF8 even has the desired effect on Kindle devices (when unsplit). With regard to the PDOC/EBOK cdetype (EXTH 501), there would need to not only be a way to change the record but also to add the record entirely. Because so far as I can tell, Kindlegen/Previewer doesn't create the 501 EXTH record in the output it creates in the first place. That record doesn't get added until delivery from Amazon to a user's device.

11-08-2013, 04:14 PM	#638
DiapDealer Grand Sorcerer Posts: 28,040 Karma: 199464182 Join Date: Jan 2010 Device: Nexus 7, Kindle Fire HD	We should probably get a new thread going for this discussion and/or proposed tool.

11-08-2013, 06:10 PM	#639
KevinH Sigil Developer Posts: 8,160 Karma: 5450818 Join Date: Nov 2009 Device: many	Hi, Okay created a "Need for a dual mobi metadata tweaker" thread to move this discussion to. Kevin

12-18-2013, 10:02 PM	#641
DaleDe Grand Sorcerer Posts: 11,470 Karma: 13095790 Join Date: Aug 2007 Location: Grass Valley, CA Device: EB 1150, EZ Reader, Literati, iPad 2 & Air 2, iPhone 7	has anyone taken a look at the AZK format that is now being used on the iOS devices and readers from Amazon? The Previewer 2.9.2 will make one of those if you select an iOS device. Dale