Index of Custom Dictionaries for Kobo eReader

[AlPe]

Alright, no problem.

=== === ===

I have just pushed a new version (v2.0.2) of Penelope on GitHub, adding two functions:

1. reading StarDict .syn files, as requested above
2. the possibility of merging more definitions for the same index word, by specifying the switch(es) --merge-definitions and --merge-separator . The latter can be used to specify a string (the default value is a space: " ") to be placed between two definitions in the merged definition string.

Let me know if you find bugs.

[AlPe]

I also add that the current code is a mess, and it should undergo a deep refactoring, which I do not have time to do now. If someone wants to take over, I will be happy to.

[mietek81]

I am having this error when trying to merge several stardict dictionaries into one in stardict format:

michal@konrad ~/penelope/penelope-master/src
$ python penelope.py --sd -p /cygdrive/f/Pobrania/Gotowe/_dict/moje/SJP\ 1.0\ 2006/sjp-nc,/cygdrive/f/Pobrania/Gotowe/_dict/moje/SF\ 1.0\ 2006/sf-nc,/cygdrive/f/Pobrania/Gotowe/_dict/moje/SWB\ 1.0\ 2006/slo-nc,/cygdrive/f/Pobrania/Gotowe/_dict/moje/SWO\ 1.0\ 2006/swo-nc -f pl -t pl --output-sd
[INFO] Input dictionary has sequence type 'h'.
[INFO] Input dictionary does not have a synonym file.
[INFO] Input dictionary has sequence type 'h'.
[INFO] Input dictionary does not have a synonym file.
[INFO] Input dictionary has sequence type 'h'.
[INFO] Input dictionary does not have a synonym file.
[INFO] Input dictionary has sequence type 'h'.
[INFO] Input dictionary does not have a synonym file.
[INFO] Starting conversion with the following parameters:

[INFO] Dictionary file: new./cygdrive/f/Pobrania/Gotowe/_dict/moje/SJP 1.0 2006/sjp-nc.dict
[INFO] Index file:      new./cygdrive/f/Pobrania/Gotowe/_dict/moje/SJP 1.0 2006/sjp-nc.idx
[INFO] Language from:   pl
[INFO] Language to:     pl
[INFO] License:         GNU GPL 3
[INFO] Copyright:       GNU GPL 3
[INFO] Title:           Dictionary pl -> pl
[INFO] Description:     Dictionary pl -> pl
[INFO] Year:            2012

[INFO] Reading input dictionaries...
[INFO] Parsing the input data...
[INFO] Using the built-in parser...
[INFO] Outputting in StarDict format to file...
Traceback (most recent call last):
  File "penelope.py", line 2191, in <module>
    main()
  File "penelope.py", line 2147, in main
    write_to_stardict_format(config, parsed_data, debug)
  File "penelope.py", line 617, in write_to_stardict_format
    dictionary_file = open(dictionary_filename, "wb")
IOError: [Errno 2] No such file or directory: 'new./cygdrive/f/Pobrania/Gotowe/_dict/moje/SJP 1.0 2006/sjp-nc.dict'

Is it because of spaces in directory name? Or maybe rather something else as with the same input dictionaries I created a Kobo version of merged dictionary.

[AlPe]

It is indeed a bug in penelope, due to poor path management. The fact that the tentative output is

'new./cygdrive/f/Pobrania/Gotowe/_dict/moje/SJP 1.0 2006/sjp-nc.dict'

suggests that you already have

'/cygdrive/f/Pobrania/Gotowe/_dict/moje/SJP 1.0 2006/sjp-nc.dict'

(indeed, it is one of your input files!) so the script attempts to create

'new.' + '/cygdrive/f/Pobrania/Gotowe/_dict/moje/SJP 1.0 2006/sjp-nc.dict'

instead of

'/cygdrive/f/Pobrania/Gotowe/_dict/moje/SJP 1.0 2006/new.sjp-nc.dict'

(here is the bug).

=== === ===

A workaround should be the following:

1. cd to the directory containing the python script
2. copy your input dictionaries there
3. run penelope.py from the same directory

(Not elegant, I know...)

$ cd ~/penelope/penelope-master/src

$ cp /cygdrive/f/Pobrania/Gotowe/_dict/moje/SJP\ 1.0\ 2006/sjp-nc* .
$ cp /cygdrive/f/Pobrania/Gotowe/_dict/moje/SF\ 1.0\ 2006/sf-nc* .
$ cp /cygdrive/f/Pobrania/Gotowe/_dict/moje/SWB\ 1.0\ 2006/slo-nc* .
$ cp /cygdrive/f/Pobrania/Gotowe/_dict/moje/SWO\ 1.0\ 2006/swo-nc* .

$ python penelope.py --sd -p sjp-nc,sf-nc,slo-nc,swo-nc -f pl -t pl --output-sd

should generate new.sjp-nc.* files with the merged contents.

[mietek81]

Ah, reliable as always!

michal@konrad ~/penelope/penelope-master/src
$ python penelope.py --sd -p sjp-nc,sf-nc,slo-nc,swo-nc -f pl -t pl --output-sd

...

[INFO] Creating compressed dictionary file new.sjp-nc.dict.dz...
[INFO] File new.sjp-nc.dict.dz created successfully!
[INFO] Files new.sjp-nc.dict.dz, new.sjp-nc.idx, and new.sjp-nc.ifo created successfully!

Another thing. When merging like this - I guess that definitions from scource dictionaries for the same word in merged dictionary are in order of provided to Penelope? So it would be: definition from sjp-nc, definition from sf-nc, definition from slo-nc, definition from swo-nc?

[AlPe]

Yes, when merging more dictionaries into one, order of the (word, definition) is the same as the order of the input dictionaries. That is:

Input D1: (word1, defA), (word1, defB), (word2, defC)
Input D2: (word0, defD), (word1, defE), (word2, defF), (word2, defG)
Output D1+D2: (word0, defD), (word1, defA), (word1, defB), (word1, defE), (word2, defC), (word2, defF), (word2, defG)

=== === ===

Note that with the latest version (2.0.2), you can specify the

--merge-definitions

flag, which will collapse all the (word, def1), ..., (word, defN) into a single index entry (word, def1 ... defN), i.e. it concatenates the definitions from all the input dictionaries. Continuing the example above:

Output D1+D2 (with --merge-definitions):
(word0, defD)
(word1, defA defB defE)
(word2, defC defF defG)

There is also --merge-separator=STRING to specify the separator to be used when merging:

Output D1+D2 (with --merge-definitions and --merge-separator=" | "):
(word0, defD)
(word1, defA | defB | defE)
(word2, defC | defF | defG)

[unaj41]

English -> Czech

[AlPe]

Added, thank you!

[huuduc1kk2]

Here are 2 Dict EV and VE i have convert, it have break line
English-Vietnamese:

Vietnamese-English:


09/01/2022 Update: I Uploaded Files Here

[unaj41]

Originally Posted by AlPe

Added, thank you!

No, thank you AlPe - you helped me couple years back with compiling these ... And while you at it please add the opposite (Czech -> En) dictionary too ... Thanks!

[AlPe]

@unaj41 : sure, no problem. Added also the Czech->English.

@huuduc1kk2 : thank you, I added them to the first post.

[indiffereddie]

English - Serbian

[AlPe]

Added, thank you.

[invention13]

Is there a Swedish-English dictionary.
The ones I saw posted are ...de-en.zip which suggests it is a german-english dictionary.

[PeterT]

Remember a user added dictionary has to replace one of the existing ones; so the Swedish-English one is replacing the German-English one.