[GUI Plugin] WordDumb

[xxyzz]

Originally Posted by beecom

I thought WordDumb plugin gets the information from kll.en.en.klld file, but there is just one definitions per word in this klld file. So, WordDumb gets many other definitions from other file than kindle's kll.en.en.klld file?

The plugin only get Kindle Word Wise definitions from the .klld file. Some lemmas have multiple rows in the "senses" table.

[beecom]

Originally Posted by xxyzz

I'm working on import Anki .apkg file: enable the word of the card front field and set difficulty based on the card type(new, learning, review, relearning).

Import CSV requires a specific file structure. I would assume the first column is word, the second column is difficulty(1-5).

To change the difficulty level of a word, I made a test CSV file that contains a word and difficulty number. It contains only "abode" and "5".
At first when I looked up 'abode' in the 'Customize kindle word wise', the difficulty level was 4.
But even after I imported my CSV file, the difficulty level of 'abode' was still 4 in the 'Customize kindle word wise' window.
Did I do something wrong?
And where does the difficulty information saved?

[xxyzz]

I test the csv file with this content and it works:

adobe,5

Difficulty data is stored in the "kindle_lemmas_dump_v0" file in the "worddumb-lemmas" folder, the default file is in the plugin's data folder.

For Wiktionary, it's in the "worddumb-lemmas/en/wiktionary_en_v0.json" file.

[beecom]

Originally Posted by xxyzz

I test the csv file with this content and it works:

adobe,5

Difficulty data is stored in the "kindle_lemmas_dump_v0" file in the "worddumb-lemmas" folder, the default file is in the plugin's data folder.

For Wiktionary, it's in the "worddumb-lemmas/en/wiktionary_en_v0.json" file.

I found out that the first row of the csv file doesn't have effect.
I made a CSV file with two rows, then the word in the second line was reflected but the word in the first line had no effect. So, if the csv file has only one line, it is not reflected on the worddumb.

One more question is, how can I use Wiktionary for worddumb? I tried hard to find the way in this thread, but coundn't. Would you please give me a clue?

Thank you very much.

[xxyzz]

The csv file I tested only has one row and it's not ignored. The first row of csv file usually is the table header not the actually data, maybe that's why it's ignored.

Using Wiktionary data for Kindle Word Wise needs to create a new .klld file, but I don't know the meaning of some columns of the "senses" table, for example: "sense_number", "synset_id", and "corpus_count". Besides, the short definitions extracted from Wiktionary are not short for many words, they may not display very well anyway.

Update:
You can convert the book to EPUB then use the plugin to add Word Wise <ruby> tags and footnote to the EPUB file. Then you can read the new EPUB file or convert it to KFX or AZW3.

[beecom]

Changing of hint level from csv file works well for the books that is moved via worddumb after the chang. But for the books that were already in the kindle before the hint level change are not affected.
Is there a way to make the change effective for books that were bought from Amazon?
(Now I am building up my own English-Korean word wise with custom hint level. So happy about this)

[xxyzz]

Originally Posted by beecom

Is there a way to make the change effective for books that were bought from Amazon?
(Now I am building up my own English-Korean word wise with custom hint level. So happy about this)

Remove DRM from your book(https://github.com/noDRM/DeDRM_tools), use WordDumb to create Word Wise file, then send the book and created file to Kindle.

Would you mind share the source of your English to Korean definition data if it's free to use like Wiktionary?

[beecom]

Originally Posted by xxyzz

Would you mind share the source of your English to Korean definition data if it's free to use like Wiktionary?

I am making my own English-Korean dictionary in a csv file. Now it contains only about 300 words. After I look up some words while reading a book with kindle, I review the words that I newly learned using the vocabulary builder of Kindle. Then, I add the words and their short definitions of my own with hint level into my csv file.

I made a short python script that replaces short_def field of kll.en.en.klld file with my own short Korean definition and then put this new kll.en.en.klld file in Kindle.
My script uses sqlite3, pandas, and base64 library.

[xxyzz]

Oh, I though you find the short definitions from somewhere.

You can import the Kindle Vocabulary Builder database file, WordDumb'll set the difficulty based on the lookup times of each word.

I'm trying to edit the wiktextract code to parse non-English Wiktionary dump file, currently I have only test the code on the Chinese dump file. Once all Wiktionary dump files can be parsed, I can use these files to add Word Wise short definitions in many languages. You can find my fork repo at here: https://github.com/xxyzz/wiktextract/tree/non-english

[beecom]

Originally Posted by xxyzz

Oh, I though you find the short definitions from somewhere.

You can import the Kindle Vocabulary Builder database file, WordDumb'll set the difficulty based on the lookup times of each word.

I'm trying to edit the wiktextract code to parse non-English Wiktionary dump file, currently I have only test the code on the Chinese dump file. Once all Wiktionary dump files can be parsed, I can use these files to add Word Wise short definitions in many languages. You can find my fork repo at here: https://github.com/xxyzz/wiktextract/tree/non-english

Importing kindle vocabulary builder is a good function for many users. But in my case, once I look up a word, I add that word into my custom vocabulary list, and next time I see the word in a book, the definition appears. So I seldom look up a same word again.

I hope you successfully finish your project about Wiktionary. I am a novice in coding, and there is not much I understand about the process. I hope there would be a chance to help you someday.

[xxyzz]

Are you interested in adding Korean translations for the plugin UI texts? You could use Poedit(https://poedit.net) or any other text editor to create new translation file from the Chinese translation file("zh_*.mo") in the plugin's "translations" folder.

You can find some debug documents at here: https://github.com/xxyzz/WordDumb/bl...d#contributing

[beecom]

Originally Posted by xxyzz

Are you interested in adding Korean translations for the plugin UI texts? You could use Poedit(https://poedit.net) or any other text editor to create new translation file from the Chinese translation file("zh_*.mo") in the plugin's "translations" folder.

You can find some debug documents at here: https://github.com/xxyzz/WordDumb/bl...d#contributing

Are you talking about translations of worddumb UI text?
I can't find 'translations' folder in AppData\Roaming\calibre\plugins folder and no files like "zh_*.mo" is found in my pc.
I found "translations" folder in C:\Program Files\Calibre2\app\translations. But there is no .mo files.
Did I misse installing something?

[xxyzz]

Yeah, I mean the UI text. you can find the .mo files in the project folder: https://github.com/xxyzz/WordDumb/tr...r/translations

The plugin code is in the "WordDumb.zip" file. But for testing and debugging, you'll need to clone the git project. If you are familiar with git, you could send a pull request at GitHub. Otherwise, you could also upload the new translation file and I'll create a git commit for you.

[beecom]

Originally Posted by xxyzz

Yeah, I mean the UI text. you can find the .mo files in the project folder: https://github.com/xxyzz/WordDumb/tr...r/translations

The plugin code is in the "WordDumb.zip" file. But for testing and debugging, you'll need to clone the git project. If you are familiar with git, you could send a pull request at GitHub. Otherwise, you could also upload the new translation file and I'll create a git commit for you.

I checked out the links, but I couldn't figure out what to do. I haven't used poedit and I don't know how to open Chinese translation file, either.
If you could give me the texts in English or Chinese, I would be able to translate them into Korean line by line(In case of Chinese texts, I can use google translation first then smooth it out). If doing so would help, I'd be happy to.

And this is another question.
I didn't know until now that if I 'create word wise' without Kindle plugged in, an epub with definition notes is made.(You explained about epub and wiktionary before, but I didn't understand then. I thought that worddumb is for use with Kindle devices only.)
It is great to have an epub file that I can read in any e-book readers.
Now, I hope I can edit the dictionary that the definition is brought from to make epub file like I did with kll.en.en.klld file.(changing definitions with my custom Korean definition)
For that, which file should I edit?

[xxyzz]

Originally Posted by beecom

but I couldn't figure out what to do. I haven't used poedit and I don't know how to open Chinese translation file, either.

*.po files are plain text files, you could open them in any text editors. The file contains the English text and the translation text, for example:

#: config.py:82 config.py:323     # this line is a comment, it's the text location in source code
msgid "Preferred format order"    # the English text
msgstr "首选格式顺序"                  # the translation text, in this case, Chinese text

Originally Posted by beecom

I didn't know until now that if I 'create word wise' without Kindle plugged in

Now, I hope I can edit the dictionary that the definition is brought from to make epub file like I did with kll.en.en.klld file. For that, which file should I edit?

Only sending file to device and open customize Kindle lemmas dialog the first time requires a connected Kindle device.

For EPUB books, you can double click the gloss text in the customize EPUB Wiktionary dialog to edit them or edit the "wiktionary_*_v0.json" file in the worddumb-lemmas folder. You need to click the save button after editing the JSON file.

I'm still working on parsing the non-english Wiktionary dump files, currently I can use the wiktextract project to parse the Chinese Wiktionary dump file. If anyone is interested, here are the code links:

• https://github.com/xxyzz/wiktextract/tree/non-english
• https://github.com/xxyzz/wikitextpro...unto_submodule