Pocketbook dictionary format revisted

[Eva_19]

Thank you so much, Markismus! You're literally my saviour, the dictionary works fully! My gratitude is immense! I'm not a tech material but I made a vain attempt to covert that file and I nearly lost hope. I tapped random korean words and they were searched and translated instantly with your file. More million thanks :-)

[Markismus]

@AvijitBis The dictionary is converted. However, it is a glorious html mess. It seems to be the furthest I am willing to stretch the code. It's on pCloud.

[mandelgado58]

Originally Posted by Markismus

@mandelgado58 Diccionario Esencial Français Espagnol is reconstructed and uploaded here. It was a badly converted ePub, so there is a repetitive string inflexiones in the definitions. In the ePub there is even a missing image in front of it. So maybe you shouldn't let Calibre convert it and put the unencrypted, unconverted dictionary in a dropbox link. (Calibre unencrypts the documents when they are first added, no consecutive conversion is needed. )

This one is how the file is in Calibre library:

https://www.dropbox.com/s/29fulcrpv5...ocido.azw?dl=0

Thankyou!

[Vi42per]

Dear Markismus

I would also like to ask for a help with a conversion of DE-EN dictionary from MOBI to Stardict format (allegedly natively understood by ONYX BOOX when placed properly into file structure) as I failed converting myself.

https://anonfiles.com/bcWdSdT9y3/German-English_mobi

Thank you in advance.

[mandelgado58]

Originally Posted by Markismus

@mandelgado58 Diccionario Esencial Français Espagnol is reconstructed and uploaded here. It was a badly converted ePub, so there is a repetitive string inflexiones in the definitions. In the ePub there is even a missing image in front of it. So maybe you shouldn't let Calibre convert it and put the unencrypted, unconverted dictionary in a dropbox link. (Calibre unencrypts the documents when they are first added, no consecutive conversion is needed. )

Thankyou very much it works well, only the inflexions

Here is how Calibre keep it in the librery:

https://www.dropbox.com/s/29fulcrpv5...ocido.azw?dl=0

[Markismus]

@Vi42per Nice! I already converted the EN-DE from the same publisher. This one doesn't convert, though. That one was probably an older version. Odd that you have a mobi-format from 2016. Seems Kindle was using azw-format by then.

@Mandelgado58 This file is also a mess. Would need a lot of extra work to convert. You're stuck with the string inflexions. However, you can download the xdxf-file, manually remove all those strings and convert it.

[San Zamoyski]

Hi!

How to add language information to xdxf so convert.exe see it correctly? Currently my dictionary (cc-sjp.zabałaganionemiejsce.pl) is listed as 'other'.
See

Best regards!

[eMole]

I think that language grouping is hardwired for factory dicts. User dicts always appear under Other.

In my experience convert.exe ignores XDXF lang_from and lang_to fields, only reads the dictionary name field.

[nezih]

Originally Posted by Vi42per

Dear Markismus

I would also like to ask for a help with a conversion of DE-EN dictionary from MOBI to Stardict format (allegedly natively understood by ONYX BOOX when placed properly into file structure) as I failed converting myself.

https://anonfiles.com/bcWdSdT9y3/German-English_mobi

Thank you in advance.

Hi, I was able to convert your dictionary to StarDict format successfully. I unpacked it via mobitool and fed the resulting html to my script. Afterwards, I converted the textual xml format via StarDict editor but you can also use PyGlossary. If you want to try it yourself here is the link to my script's github repository: mobi2stardict

[DenS]

Originally Posted by nezih

If you want to try it yourself here is the link to my script's github repository: mobi2stardict

Hi @nezih. I ran your script at the windows prompt and was able to convert a .html dictionary to .xml. Next I used pyglossary to convert the .xml to stardict(.ifo). It worked great, Thanks!
But there is a dictionary, actually what I needed most, which I can't convert to .xml. The command I use at the prompt is this:

mobi2stardict.py --html-file "book.html" --fix-links --dict-name "Grande Dicionário de Português" --author "Porto Editora" --textual --chunked

And the prompt gives me this error:

Traceback (most recent call last):
  File "D:\Downloads\mobi2stardict\mobi2stardict.py", line 160, in <module>
    convert(args.html_file, args.dict_name, args.author, args.fix_links, args.gls, args.textual, args.chunked)
  File "D:\Downloads\mobi2stardict\mobi2stardict.py", line 115, in convert
    key     = ET.SubElement(article, "key").text = entry.HW
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "src\lxml\etree.pyx", line 1042, in lxml.etree._Element.text.__set__
  File "src\lxml\apihelpers.pxi", line 748, in lxml.etree._setNodeText
  File "src\lxml\apihelpers.pxi", line 736, in lxml.etree._createTextNode
  File "src\lxml\apihelpers.pxi", line 1541, in lxml.etree._utf8
ValueError: All strings must be XML compatible: Unicode or ASCII, no NULL bytes or control characters

It might be useful to say that to extract the .mobi dictionary to .html I used the KindleUnpack caliber plugin.
To install BeautifulSoup and lxml I used the commands "pip install beautifulsoup4" and "pip install lxml". The Python version I'm using is 3.11.2.
Could you help me figure out what I'm doing wrong?

[nezih]

Originally Posted by DenS

Hi @nezih. I ran your script at the windows prompt and was able to convert a .html dictionary to .xml. Next I used pyglossary to convert the .xml to stardict(.ifo). It worked great, Thanks!
But there is a dictionary, actually what I needed most, which I can't convert to .xml. The command I use at the prompt is this:

mobi2stardict.py --html-file "book.html" --fix-links --dict-name "Grande Dicionário de Português" --author "Porto Editora" --textual --chunked

And the prompt gives me this error:

Traceback (most recent call last):
  File "D:\Downloads\mobi2stardict\mobi2stardict.py", line 160, in <module>
    convert(args.html_file, args.dict_name, args.author, args.fix_links, args.gls, args.textual, args.chunked)
  File "D:\Downloads\mobi2stardict\mobi2stardict.py", line 115, in convert
    key     = ET.SubElement(article, "key").text = entry.HW
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "src\lxml\etree.pyx", line 1042, in lxml.etree._Element.text.__set__
  File "src\lxml\apihelpers.pxi", line 748, in lxml.etree._setNodeText
  File "src\lxml\apihelpers.pxi", line 736, in lxml.etree._createTextNode
  File "src\lxml\apihelpers.pxi", line 1541, in lxml.etree._utf8
ValueError: All strings must be XML compatible: Unicode or ASCII, no NULL bytes or control characters

It might be useful to say that to extract the .mobi dictionary to .html I used the KindleUnpack caliber plugin.
To install BeautifulSoup and lxml I used the commands "pip install beautifulsoup4" and "pip install lxml". The Python version I'm using is 3.11.2.
Could you help me figure out what I'm doing wrong?

Hi, If I remember correctly, I came across this problem recently. Most probably headwords include control characters. If you choose to convert to gls format(--gls) only, it will probably run fine. However, you would still need to substitute those with what they actually intended to show.
Open gls file via vscode, look for control chars. such as

BEL

,

ACK

etc. (You can use \p{C} in Find) Replace those with the intended characters. For example, in my problematic file, I replaced

BEL

with

ll

,

ACK

with

ch

.

[DenS]

Originally Posted by nezih

You can use \p{C} in Find) Replace those with the intended characters.

Sorry, do I have to put \p{C} here?

[nezih]

Originally Posted by DenS

Sorry, do I have to put \p{C} here?

Use \p{C} to glance at different control characters. Then, replace those on a case by case basis. Sorry if I can't be of help but without access to dictionary my assistance will be limited.

[DenS]

Originally Posted by nezih

Use \p{C} to glance at different control characters. Then, replace those on a case by case basis. Sorry if I can't be of help but without access to dictionary my assistance will be limited.

I sent a dm with the link (dropbox) of the dictionary

[nezih]

Originally Posted by DenS

I sent a dm with the link (dropbox) of the dictionary

I think there might be a problem with KindleUnpack because when unpacked via mobitool, the whole process completed without problem. I'm attaching the Windows mobitool binaries if you want to try it yourself.(Use at your own risk!)

mobitool -s filename.mobi