Problem, Kindle doesn't open my mobi file

moji3000 · 11-14-2012, 09:37 PM

Hi
I am developer and i decided to write a export odt to mobi plugin for the software (Calligra Author) that i am working on it. After attempts i succed to write it or better to say not succed because now calibre can open and show the files but kindle can't.

I check the differences between my file and other files ( mobi files that i had created them with mobi packet creator and calibre). I don't know where is the problem and where i am working wrong.

I attached my file. I really like to do to make my file work because i am working on it for month, and i realy need help.

and sorry if i write bad because my english is not very good.

Regards moji

Nate the great · 11-14-2012, 10:16 PM

Hi! Welcome to MobileRead!

I can't help you with the file validation, but it did also break my copy of Kindle 4PC.

But we do have a number of developers around here who have worked on the guts of Mobi files. One of them should be able to help you.

JSWolf · 11-14-2012, 10:18 PM

You could have a look at the code of Calibre and see if that helps.

fjtorres · 11-14-2012, 10:35 PM

Have you looked at the Mobiperl tools?
https://dev.mobileread.com/trac/mobiperl

Mobils might help you compare your output to a known-good mobi file.
Plus the source code for the HTML to Mobi converter might be useful.

DiapDealer · 11-14-2012, 10:47 PM

The problem is that you've not converted the html from your original source (whatever that may be) into any kind of valid mobi markup. It's a MOBI by all external appearances (the binary database structure is valid), but that's about it. Your internal mobi markup frankly looks like straight-up HTML4 -- complete with inline CSS. That'll never fly on any MOBI app/device. Calibre views it because calibre's viewer doesn't really render MOBIs. It extracts the contents of all ebooks into an OEBPS-compliant format for viewing.

Hit up the MOBI WIKI here, and use something like Mobi_Unpack.py to see what the markup looks like inside some valid MOBIs.

HarryT · 11-15-2012, 02:57 AM

Moved to the Mobi forum.

caleb72 · 11-15-2012, 03:14 AM

Given that Kindle is my platform of choice at the moment I really should get my a** into gear and learn the mobi format a bit. I could possibly put together a couple of utilities to help me.

moji3000 · 11-15-2012, 05:48 AM

Thanks for all of your answers.
DiapDealer i will use Mobi mark up and will back again.
but i have used Mobi_unpack to unpack my mobi files its html is ok but in its debug
ther are some thing that i think should be the problem.
These are the part of unpack debug:

Section is 7
First 4 bytes: 46 4c 49 53
Skipping FLIS section
Section is 8
First 4 bytes: 46 43 49 53
Skipping FCIS section
Section is 9
Skip section 9 as it contains the EOF record.

I don't have any idea !!

Thank all again.

DiapDealer · 11-15-2012, 07:07 AM

Quote:

but i have used Mobi_unpack to unpack my mobi files its html is ok

Quite frankly, it's not "OK". The raw markup contained in your mobi is quite invalid (for MOBIPOCKET).

Use calibre or kindlegen to convert your source material to a MOBI. Then use Mobi_Unpack to examine the raw markup they produce as compared to the raw markup in your mobi. The difference will be quite apparent. MOBI-markup is more akin to HTML3.2 -- all inline attributes and no styles/classes.

moji3000 · 11-22-2012, 06:54 PM

Hi DiapDealer
Thanks for your help

As you told me i put a standard mobi mark up in mobi and now kindle can open my ebooks but an uncmpress one.

Thank you

moji3000 · 11-25-2012, 08:14 AM

Hi all
again me and my mobi converter problem.
The problem is now about image, its html is standard as i know and think header is ok too.

But Kindle desktop previewer can not open it. :/
Again no idea why!!

I have attached file.

Moji

pdurrant · 11-25-2012, 09:08 AM

Quote:

Originally Posted by moji3000

Hi all
again me and my mobi converter problem.
The problem is now about image, its html is standard as i know and think header is ok too.

But Kindle desktop previewer can not open it. :/
Again no idea why!!

I have attached file.

Moji

I don't know precisely what you've got wrong or missing that Kindle wants, but here's your file after running through mobiunpack and kindlegen 2.7 and then mobiunpack again to get the non-KF3 format portion.

By comparing the files (use mobiunpack in dump mode) and tweaking the file you generate, you should be able to find out which bits you need.

But bear in mind that there's no guarantee. The Mobi file format is officially undocumented.

moji3000 · 11-26-2012, 12:42 PM

Thank you pdurrant

But it didn't help me, any idea?

when i get diff from my ebook and one of calibre ebook (mobi) in headers
There is an attribute "Extra Record Data Flags"
( https://wiki.mobileread.com/wiki/MOBI#MOBI_Header) in mobi header
That for calibre is "1". This attribute is important? and it can be the problem?

pdurrant · 11-26-2012, 01:31 PM

Quote:

Originally Posted by moji3000

Thank you pdurrant

But it didn't help me, any idea?

when i get diff from my ebook and one of calibre ebook (mobi) in headers
There is an attribute "Extra Record Data Flags"
( https://wiki.mobileread.com/wiki/MOBI#MOBI_Header) in mobi header
That for calibre is "1". This attribute is important? and it can be the problem?

The Extra Record Data Flags indicate whether there is extra info at the end of each text section in the book. The multibyte bytes flag indicates that in utf8 files the text sections include the extra bytes making up a utf-8 character if it happens to break between sections, so that complete characters can still be obtained without decoding two sections at once. (The mobiunpack code handles these).

The other two bits indicate other stuff that can be at the end of the text sections. Since you're creating the file, and it only currently has one small text section, this is not your problem.

moji3000 · 11-26-2012, 02:09 PM

Hi

I fixed the problem and now i support images too.

The problem was about two bytes zero.
After the text i should add two bytes zero.
This two bytes zero called "Magic Records ".

Thanks from all.

11-14-2012, 10:47 PM	#5
DiapDealer Grand Sorcerer Posts: 28,891 Karma: 207182180 Join Date: Jan 2010 Device: Nexus 7, Kindle Fire HD	The problem is that you've not converted the html from your original source (whatever that may be) into any kind of valid mobi markup. It's a MOBI by all external appearances (the binary database structure is valid), but that's about it. Your internal mobi markup frankly looks like straight-up HTML4 -- complete with inline CSS. That'll never fly on any MOBI app/device. Calibre views it because calibre's viewer doesn't really render MOBIs. It extracts the contents of all ebooks into an OEBPS-compliant format for viewing. Hit up the MOBI WIKI here, and use something like Mobi_Unpack.py to see what the markup looks like inside some valid MOBIs. Last edited by DiapDealer; 11-14-2012 at 10:50 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Kindle Fire doesn't display some images in MOBI file	RobFreundlich	Devices	2	03-25-2012 08:15 AM
Can I get a .mobi file to open on the cover image?	Todd Young	Conversion	2	01-17-2012 03:25 PM
Java open source api required to convert from PDF to MOBI file	tmpnkvee	Kindle Formats	5	12-05-2011 08:39 AM
Problem sending mobi file to Kindle via email	readslow	Amazon Kindle	5	07-29-2011 09:45 PM
sigil doesn't open epub file	cyberbaffled	Sigil	8	11-10-2009 04:28 PM

11-14-2012, 10:16 PM	#2
Nate the great Sir Penguin of Edinburgh Posts: 12,375 Karma: 23555235 Join Date: Apr 2007 Location: DC Metro area Device: Shake a stick plus 1	Hi! Welcome to MobileRead! I can't help you with the file validation, but it did also break my copy of Kindle 4PC. But we do have a number of developers around here who have worked on the guts of Mobi files. One of them should be able to help you.

11-14-2012, 10:18 PM	#3
JSWolf Resident Curmudgeon Posts: 80,782 Karma: 150249619 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	You could have a look at the code of Calibre and see if that helps.

11-14-2012, 10:35 PM	#4
fjtorres Grand Sorcerer Posts: 11,732 Karma: 128354696 Join Date: May 2009 Location: 26 kly from Sgr A* Device: T100TA,PW2,PRS-T1,KT,FireHD 8.9,K2, PB360,BeBook One,Axim51v,TC1000	Have you looked at the Mobiperl tools? https://dev.mobileread.com/trac/mobiperl Mobils might help you compare your output to a known-good mobi file. Plus the source code for the HTML to Mobi converter might be useful.

11-15-2012, 02:57 AM	#6
HarryT eBook Enthusiast Posts: 85,560 Karma: 93980341 Join Date: Nov 2006 Location: UK Device: Kindle Oasis 2, iPad Pro 10.5", iPhone 6	Moved to the Mobi forum.

11-15-2012, 03:14 AM	#7
caleb72 Indie Advocate Posts: 2,863 Karma: 18794463 Join Date: Sep 2010 Location: Melbourne, Australia Device: Kindle	Given that Kindle is my platform of choice at the moment I really should get my a** into gear and learn the mobi format a bit. I could possibly put together a couple of utilities to help me.

11-15-2012, 05:48 AM	#8
moji3000 Junior Member Posts: 7 Karma: 10 Join Date: Aug 2012 Device: none	Thanks for all of your answers. DiapDealer i will use Mobi mark up and will back again. but i have used Mobi_unpack to unpack my mobi files its html is ok but in its debug ther are some thing that i think should be the problem. These are the part of unpack debug: Section is 7 First 4 bytes: 46 4c 49 53 Skipping FLIS section Section is 8 First 4 bytes: 46 43 49 53 Skipping FCIS section Section is 9 Skip section 9 as it contains the EOF record. I don't have any idea !! Thank all again.

11-22-2012, 06:54 PM	#10
moji3000 Junior Member Posts: 7 Karma: 10 Join Date: Aug 2012 Device: none	Hi DiapDealer Thanks for your help As you told me i put a standard mobi mark up in mobi and now kindle can open my ebooks but an uncmpress one. Thank you

11-26-2012, 12:42 PM	#13
moji3000 Junior Member Posts: 7 Karma: 10 Join Date: Aug 2012 Device: none	Thank you pdurrant But it didn't help me, any idea? when i get diff from my ebook and one of calibre ebook (mobi) in headers There is an attribute "Extra Record Data Flags" ( https://wiki.mobileread.com/wiki/MOBI#MOBI_Header) in mobi header That for calibre is "1". This attribute is important? and it can be the problem?

11-26-2012, 02:09 PM	#15
moji3000 Junior Member Posts: 7 Karma: 10 Join Date: Aug 2012 Device: none	Hi I fixed the problem and now i support images too. The problem was about two bytes zero. After the text i should add two bytes zero. This two bytes zero called "Magic Records ". Thanks from all.