Using dictionaries purchased from Kindle Store

[qys]

I wonder if it's possible to use a dictionary purchased from the Amazon Kindle Store (for example, this one) in KOReader. It looks like the dictionary downloaded to my Kindle is in some "azw" format, not the "stardict" format KOReader supports. However, the stock reader in Kindle supports the "azw" dictionaries in "Settings > Language & Dictionaries > Dictionaries".

How do I convert "azw" to "stardict"? What are these "azw" files anyway, a generic format for e-books, or a specialized format for dictionaries?

[Ebookus]

you can try pyglossary

[DJUNGELSKOG]

I've actually managed to pull it off and I have written a manual for my future self. However I won't be at home until next week. If you remind me about it next weel then I'll post the instructions here.

I've also managed to get Rakuten Serif font to work in Koreader lol.

[qys]

Thanks Ebookus for suggesting pyglossary; that looks promising!

DJUNGELSKOG, please post the instructions once you have time! I'm sure many people like me will benefit from this.

[DJUNGELSKOG]

Ok here we go.

I was looking to achieve the same thing and eventually I've found these instructions: https://gist.github.com/Fef0/913caa8...0ed1eb9d0f11e7 . They didn't work for me initially so I have amended them.

Step 1:
Download the KindleUnpack program from here: https://github.com/kevinhendricks/KindleUnpack
Start GUI with python ./KindleUnpack.pyw
Select .mobi file as input and choose your output directory
(optional) I don't know if it is necessary, but select ePub Output type: Auto-Detect
Click the Start button and wait
A file called mobi7/book.html is now present in the output directory

Step 2
Download the mobi2stardict.py script from here: https://github.com/anezih/mobi2stardict
Place the script in the same folder as the book.html file
Run it with
python mobi2stardict.py --fix-links --dict-name "vandale-nl-nl" --textual --html-file "book.html"
A file called book_stardict_textual.xml will appear

Step 3
Download PyGlossary https://github.com/ilius/pyglossary
Select input file book_stardict_textual.xml
Select input format stardict_textual_file (.xml)
Specify the ouput file /home/schatje/Downloads/dictionaries_to_convert/name_of_ur_file
Specify the output format StarDict.ifo

Step 4 (optional, KOReader on Kindle)

(optional) If you want to change the dictionary name (it is book by default) then open book.ifo with a text editor and update "bookname" entry to bookname=wordnet3
Copy ALL the files (book.dict, book.idx, book.ifo, book.syn) to koreader/data/dict/
Restart KOReader
From KOReader select the magnifying glass > Dictionary settings > Manage dictionaries > and select wordnet3

I'm very happy with the result! Maybe there is an easier way to achieve the conversion but I didn't look for it.

[Frenzie]

Based purely on that message step 2 (or 3) sounds a bit odd given that step 2 is running something called "to stardict"?

[DJUNGELSKOG]

I'm sure that it's possible to achieve in a more effective way, but this is what has worked for me I didn't check whether the .xml file would already work with Koreader and just followed all the steps from that gist.

[Frenzie]

You're probably right that the intermediary is required if PyGlossary can't do mobi.

[nezih]

Originally Posted by DJUNGELSKOG

I'm sure that it's possible to achieve in a more effective way, but this is what has worked for me I didn't check whether the .xml file would already work with Koreader and just followed all the steps from that gist.

I'm the author of https://github.com/anezih/mobi2stardict. I plan to add direct StarDict output to the script via PyGlossary. However, starting from next week I won't be available for approximately 1 month.

Apart from that, I have started to write a library in C# that can directly read MOBI dictionary files which will again produce binary StarDict files without any intermediary format.

When any of the above is completed I can update this thread.

[Thorns]

Originally Posted by DJUNGELSKOG

Ok here we go.

I was looking to achieve the same thing and eventually I've found these instructions: https://gist.github.com/Fef0/913caa8...0ed1eb9d0f11e7 . They didn't work for me initially so I have amended them.



I'm very happy with the result! Maybe there is an easier way to achieve the conversion but I didn't look for it.

Hey I followed your steps but when I get to the second stage I keep getting the message that it's "Not a dictionary file" - when I open the HTML on the web browser it definitely looks like a dictionary! and it looks to have unpacked it properly. No idea what to do, any ideas?

[Thorns]

Originally Posted by nezih

I'm the author of https://github.com/anezih/mobi2stardict. I plan to add direct StarDict output to the script via PyGlossary. However, starting from next week I won't be available for approximately 1 month.

Apart from that, I have started to write a library in C# that can directly read MOBI dictionary files which will again produce binary StarDict files without any intermediary format.

When any of the above is completed I can update this thread.

Hey I followed your steps but when I get to the second stage I keep getting the message that it's "Not a dictionary file" - when I open the HTML on the web browser it definitely looks like a dictionary! and it looks to have unpacked it properly. No idea what to do, any ideas?

[nezih]

Originally Posted by Thorns

Hey I followed your steps but when I get to the second stage I keep getting the message that it's "Not a dictionary file" - when I open the HTML on the web browser it definitely looks like a dictionary! and it looks to have unpacked it properly. No idea what to do, any ideas?

It may render like a dictionary on a browser but if it lacks the necessary idx tags the script can't parse the file.

[Thorns]

Originally Posted by nezih

It may render like a dictionary on a browser but if it lacks the necessary idx tags the script can't parse the file.

Is there any way around it? Anyway to get it to have the idx tags somehow?

[nezih]

If it was a dictionary file, KindleUnpack should have added the necessary tags already. Can't tell much without seeing the original mobi file.

[Thorns]

Originally Posted by nezih

If it was a dictionary file, KindleUnpack should have added the necessary tags already. Can't tell much without seeing the original mobi file.

Have sent you a PM with the file if you wouldn't mind looking at it? Just the MOBI file. I wasn't sure if allowed to post here publicly. Would really appreciate any advice either way