JDict is a Japanese-English offline dictionary featuring over 160000 entries over 10000 kanji and almost 60000 example sentences.
It is largely inspired by well-known "[Imi_wa?]" dictionary for [i_O_S] platform both functionally and visually.

Current features:
- works offline
- shows kana/romaji as you type
- supports wildcards in vocabulary
- recent words examples and kanji
- romaji mode in search
- kanji search is divided by ON/KUN readings Nanori and meaning results
- very fast example search
- extract words from sentences feature
- kanji stroke order animations
- furigana is matched everywhere with my own algorithm (still can have some bugs though)
- in component view you can see all kanji with given component ordered by JLPT level and word commonness
- in kanji view you can see all compounds with given kanji ordered like above
- search kanji by components (design inspired by "Japanese" dictionary)
- browse kanji by JLPT and School Grades
- notes
- lists (works like favorites but with folders/sub-folders)
- vocabulary search recognizes some basic grammar when typing
- in word detail view examples are categorized by word meaning if such a connection exists

As you can see features are mostly basic for now.
The database file is quite big (~230MB). The reason for that is that it contains a lot of example sentences data and needs a lot of indexes to run smooth.

My plans:

- make the application more user-friendly (list search)
- learn vocabulary feature (using lists)
- exporting lists to Anki
- external links in vocabulary
- kanji handwriting recognition
- Shiratori game
- improve extract words algorithm (I have some ideas using the statistical approach to make it run faster)

Acknowledgments:

- Electronic Dictionaries Research Groups
- KanjiVG
- Tatoeba
- KanjiCafe
- Tanos (JLPT)