[PoP]

Originally Posted by baf

I didn't find any tool able to decompile this kind of index.
Why do you want to unpack it?

To shamelessly copy good structures

Originally Posted by baf

The format is well described in Amazon publishing guidelines, which you quoted. Doesn't it work for you?

I finaly tried it and it works great

Spoiler:

H:\>mobigen nadsat.opf

*****************************************
* Mobipocket mobigen.exe V6.2 build 43  *
* A command line e-book compiler        *
* Copyright Mobipocket.com 2003-2008    *
*****************************************

opt version: try to minimize (default)
Info(prcgen): Added metadata dc:Title        "NADSAT chelloveks multilingual dic
tionary v0.4"
Info(prcgen): Added metadata dc:Date         "2013-02-16"
Info(prcgen): Added metadata dc:Creator      "PoP"
Info(prcgen): Added metadata dc:Publisher    "Pop via MobileRead"
Info(prcgen): Added metadata BASICCode       "REF008000"
Info(prcgen): Added metadata dc:Subject      "Dictionaries"
Info(prcgen): Added metadata dc:Description  "Translations FRench: ENglish: ARgo
t: RUssian: from Internet resources"
Info(prcgen): Parsing files  0000001
Info(prcgen): Resolving hyperlinks
Info(prcgen): Building table of content     URL: H:\nadsat.ncx
Info(prcgen/inflections): Number of new <idx:infl> inflection rules:  0000003
Info(prcgen/inflections): Of which rules used only once or twice:     0000003
Info(prcgen/inflections): Number of inflection rule groups:           0000002
Info(prcgen): Computing UNICODE ranges used in the book
Info(prcgen): Found UNICODE range: Basic Latin [20..7E]
Info(prcgen): Found UNICODE range: Latin-1 Supplement [A0..FF]
Info(prcgen): Found UNICODE range: Basic Arabic [600..6FF]
Info(prcgen): Found UNICODE range: Arabic Presentation Forms-B [FE70..FEFE]
Info(prcgen): Found UNICODE range: Mathematical Operators [2200..22FF]
Info(prcgen): Found UNICODE range: Basic Greek [370..3FF]
Info(prcgen): Found UNICODE range: Latin Extended-B [180..24F]
Info(prcgen): Found UNICODE range: Cyrillic [400..4FF]
Info(prcgen): Found UNICODE range: Latin Extended-A [100..17F]
Info(prcgen): Building MOBI file, record count:   0000019
Info(prcgen): Final stats - text compressed to (in % of original size):  035.30%

Info(prcgen): The document identifier is: "NADSAT_chell-ictionary_v0.4"
Info(prcgen): The file format version is V6
Info(prcgen): Saving MOBI file
Info(prcgen): MOBI File successfully generated!

H:\>

Interestingly, now that my dictionary has inflections, mobiunpack fails on it, as expected.

The tags are lengthy to code by hand and probably meant to be programmatically generated. Well, that concludes my quest: there are no magical builtin rules in amazon code, only tediously adding infl tags will bring inflections support.

Thank you again for your helpful posts.