DaKanji-Dictionary

The repository to build and store the dictionary files for DaKanji.

Setup

You can either download the latest release from the releases page, clone this repo and then directly jump to Usage or

Building from source

First install the necessary requirements:

• a python environment with all dependiencies in requirements.txt installed. We recommend python 3.10.12, as this version is tested to work. Currently (19.11.2023), it seems that not everything works on 3.12.
  • Assure this environemnt is set as current environment (python3 defaults to this env.)
• dart
• MeCab with ipadic
  • Windows
  • MacOS - homebrew:
    • brew install mecab
    • brew install mecab-ipadic
• Optional
  • Debugging svgs
    • Assure Cairo is installed and on path for cairocffi (Installation details)

Additionally, the isar binaries for your platform are requried. They can be downloaded here (make sure that the isar library version matches your platform and the version in pubspec.yaml). After downloading the library, copy it to database_builder/example/libisar.{dylilb|so|dll}.

Now download the necessary dictionary files

• Pitch accent data
• Wanikani audio files
  • Releases -> latest
• JLPT data comes from Jonath wallner and is curated by stephenmk
  • Releases -> latest
• JMDict or direct download
• KanjiDic2 or direct download
• Kanjium
  • Antonyms
  • Kanjidict
  • Look alikes
  • particles
  • Synonyms
• KanjiVG (Download the kanjivg-xxx-all.zip)
• KLC order
• Tatoeba
  • Weekly exports -> All languages -> sentences.csv
  • Links -> links.csv
• Radkfile / Kradfile
  • Releases -> latest
• Currently unused
  • ENAMDICT or direct download

and place them in the inputFiles directory. Your folder should look like that

• inputFiles
  • japanese-vocabulary-pronunciation-audio-master
    • lib
      • mp3
      • ogg
  • jlpt
    • n_{1-5}.csv
  • JMdict
    • JMdict
    • JMnedict.xml
  • kanjidic2
    • kanjidic2.xml
  • kanjium
    • antonyms.txt
    • kanjidict.txt
    • lookalikes.txt
    • synonyms.txt
  • kanjiVG
    • 0f9a8.svg
    • 0f9ab.svg
    • ...
  • klc
    • database.js
  • tatoeba
    • links.csv
    • sentences.csv
  • accents.txt
  • radkfile-x.x.x.json
  • kradfile-x.x.x.json

Now, you need to make sure that your python environment has all the requirements from requirments.txt installed. You can use python -m pip install -r requirements.txt if you want to install them on your main environment, or use python -m pip install virtualenv, followed by virtualenv venv. Next, if you use linux, use source venv/bin/activate or venv\Scripts\activate if on windows. At last, just use pip install -r requirements.txt like you would normally.

Afterwards, you can either use execute.sh or manually run the following commands if the shell script doesn't work:

• python utils/database_entries_creator/database_entries_creator.py -a
• ASSURE MeCab is installed in your default python environment
• cd database_builder
• dart run build_runner build
• dart example/database_builder_example.dart

Usage

After building / downloading the dictionary it can be used by including import 'package:database_builder/database_builder.dart'; in your dart app.

Credits and additional information

• Databases:
  • KanjiVG
  • Kanjidic 2
  • JMDict
  • ENAMDICT
  • Tatoeba