Index of Custom Dictionaries for Kobo eReader

[oren64]

Originally Posted by rmreader

Hi orren64,

I've converted the file. I got a few errors, so I'm not sure if this works. Please report back if it works. Then Alberto can add it to the list.

Kind regards,
Ruud

Thanks a lot.

But it's not working i rename existing dictionary restart, it's show no translation or only symbols .
maybe it's because hebrew it's right to left.

maybe if you can the convert the GBL to stardict to dicthtm.

http://www.babylon.com/free-dictiona...ebrew/905.html

Thanks

Oren

[AlPe]

Originally Posted by rmreader

I've converted the file. I got a few errors

When using Penelope for Python 2, I get dicthtml-en-he.zip of size 6297832 bytes.

When using Penelope for Python 3, I get a

UnicodeEncodeError: 'ascii' codec can't encode character u'\xf3' in position 1: ordinal not in range(128)

(which might be a bug or a problem with the input file, I do not know yet), and no file dicthtml-en-he.zip is output. What was the error you mentioned?

[tatung2112]

Originally Posted by AlPe

As far as I know, there is no way to embed a font inside a Kobo dictionary, for the reader/app to use it when displaying the dictionary definition.

Hence, all you have is the default "Kobo" font: if it does not have glyphs for certain (or even all) Vietnamese characters, you are out of luck.

(Maybe custom firmwares do have the possibility of changing the dictionary font --- I do not know for sure, try searching the Kobo Developer part of this forum.)

I have tried reaching the Kobo devs several times, asking them for the dictionary format documentation and official support for custom dictionaries, but with zero results.

===

Is the dictionary linked above still usable? Would you like me to list it in the index in the first post?

Thank you very much for your replying. I have searched and can display Vietnamese well in Kobo dictionary (thread Modifications for bad eyes). The dictionary that I attached in the another post is English - Vietnamese, hope it will help someone when you put it to the first post dictionary list. Cheers!

[johnnyb]

Any chance someone here has a French-German dictionary (and maybe vice versa) lying around?

[AlPe]

@tatung2112 : added to the first post, thank you.

[rmreader]

@Oren64:
The file you mentioned is the same as I tried to convert.
@Alberto:
I used stardict-editor to compile the BGL file and got:
=======
Building...
Default charset: cp1252
Source Charset: cp1252
TargetCharset: cp1255
.................................................. ...........................Error in iconv conversion
.................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .............................................Error in iconv conversion
Error in iconv conversion
.................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .............
Write file: /home/ruud/dictionaries/dicts/Babylon_English_Hebrew.babylon

Bookname: Babylon English-Hebrew
Word count: 163443
Author: Babylon Ltd.
Email: linguistic-support@babylon.com
Website:
Description: This comprehensive English-Hebrew dictionary contains words, phrases, abbreviations & acronyms. It includes terms from a vast variety of subjects, such as Medicine, Electronics, Zoology, Business, Computers, Religion, etc., and features both the American and British forms of spelling.
Done!
=======
In the terminal window I also got three error messages:
(stardict-editor:20117): Gtk-CRITICAL **: gtk_text_buffer_emit_insert: assertion `g_utf8_validate (text, len, NULL)' failed

Then compiling the .babylon file I got:
=======
Building...
Line 16, bad word!!!
Line 19, bad word!!!
Line 303938, bad definition!!!
Line 490331, bad definition!!!
Over
wordcount: 163439
synwordcount: 132928
Done!
=======
Then I started penelope:
python ~/penelope_v2.0.0/penelope.py -p Babylon_English_Hebrew -f en -t he --output-kobo
This gave no errors and poutput:
=======
[INFO] Input dictionary has sequence type 'h'.
[INFO] Starting conversion with the following parameters:

[INFO] Dictionary file: new.dicthtml-en-he
[INFO] Index file: words
[INFO] Language from: en
[INFO] Language to: he
[INFO] License: GNU GPL 3
[INFO] Copyright: GNU GPL 3
[INFO] Title: Dictionary en -> he
[INFO] Description: Dictionary en -> he
[INFO] Year: 2012

[INFO] Reading input dictionaries...
[INFO] Parsing the input data...
[INFO] Using the built-in parser...
[INFO] Outputting in Kobo format to file...
[INFO] Creating compressed index file words...
[INFO] Creating zip file new.dicthtml-en-he.zip...
[INFO] File new.dicthtml-en-he.zip created successfully!
[INFO] File new.dicthtml-en-he.zip created successfully!
=======
I don't know anything of what is really going on, I just know this trick to convert BGL to kobo dictionary.
Does it help if I send the input files to penelope?

Kind regards,
Ruud

[oren64]

Hi rmreader

AlPe succeed to convert the BGL to dicthtml, but it didn't work because there's no Hebrew glyphs in kobo, the definition box pops up contains unreadable symbols.
Thanks for trying.

Oren

[rmreader]

Hi johhnyb,

On the Babylon site I found a French to German dictionary and converted it. I could not find the reverse. Please check if it works. Then Alberto can add it to the list.

Kind regards,
Ruud

[johnnyb]

Originally Posted by rmreader

Hi johhnyb,

On the Babylon site I found a French to German dictionary and converted it. I could not find the reverse. Please check if it works. Then Alberto can add it to the list.

Kind regards,
Ruud

Works great, thanks a lot!

[AlPe]

Added, thanks to rmreader again.

[AlPe]

Originally Posted by rmreader

@Oren64:
I don't know anything of what is really going on, I just know this trick to convert BGL to kobo dictionary.
Does it help if I send the input files to penelope?

Kind regards,
Ruud

As Oren commented above, I tried to convert the dictionary from the StarDict file he provided. On Python 2 the conversion to Kobo format succeeded, on Python 3 failed with the error quoted above (I have not investigated the issue).

I privately sent the file generated with Penelope on Python 2 to Oren, and he confirmed that the dictionary seems to be properly installed, but when he looks a word up, the definition panel shows up with unreadable symbols --- which seems to suggest that the font that Kobo uses for the vocabulary has no glyphs for the Hebrew characters, let alone the fact that Hebrew should be displayed right-to-left.

In fact, trying to convert to CSV or XML from the same StarDict, yields a "correct" output file (that is, when opened in Vim, I can see the definitions in Hebrew). Moreover, by manually querying the "words" index, it contains the words in the original StarDict file --- which is consistent with Oren's observation that the dictionary box pops up, but contains unreadable text.

[AlPe]

Important question, as I have just realized that many of the dictionaries listed here come from Babylon dictionaries: are you sure that those conversions are legal? Where can I find the text of the license associated with Babylon dictionaries?

I repeat myself:

Please respect the copyright laws, and NEVER EVER post about converted dictionaries that infringe them. (Also, do not attempt at mailing me asking for converting copyrighted material, I will not help you on that.)

[rmreader]

Hi Alberto,

All the babylon dictionaries I converted came from http://www.babylon.com/free-dictionaries. So I guess from the name that it is legal. ;-)

Kind regards,
Ruud

[AlPe]

Thanks for the link.

However, "free" as in "not paid" does not automatically mean "it is OK to download and use them inside something else than Babylon software".

On that page there is only a generic

The free version of Babylon is for private use only.

while the Terms of Service seem to imply that the "free dictionaries", being provided by third parties (mostly users themselves, judging from the dictionaries' descriptions), can be used according to each dictionary's own licensing status.

AFAIK, I am mostly satisfied that the "free Babylon dictionaries" seem to origin from open/collaborative/freely licensed sources, thus converting to Kobo format seems OK.

=== === ===

BTW, at this point implementing support for the Babylon format in Penelope itself seems worth, doesn't it?

[rmreader]

You're right, I didn't look far enough and thought that since they were contributed it was okay to use them on the Kobo too. Sorry.

The way I convert them is to use the stardict-editor on Linux. You load the .BGL file, convert it to a .babylon file, load that one and convert it to the files needed for Penelope. I think the stardict-editor is open source so if you want to add support to Penelope you might look there. For me the few extra steps is no problem since I only had to convert a few dictionaries.

BTW, thanks for making Penelope otherwise I (and a lot of other people) would be missing a lot of helpful dictionaries. I use a them quite a lot.