Bushu dictionary files
dev.mtsstudio.bushu · iOS · Android
Every word, reading, meaning, example sentence and character breakdown in Bushu comes from open dictionaries. Bushu's copies are filtered, reformatted, graded into levels and annotated, so they are modified versions of that data.
They are offered here under the same licences as the data they derive from, exactly as they ship inside the app, so you can get them without going through an app store. What was changed in each source is stated in the notices file.
Japanese vocabulary
japanese.json.gz · 3.5 MBAbout 30,000 words with readings, meanings, per-character furigana, pitch accent and a Bushu level per word, with example sentences and translations for the graded levels.
Chinese vocabulary
chinese.json.gz · 6.0 MBAbout 116,000 words in simplified and traditional script with pinyin, meanings, measure words and a Bushu level per word, with example sentences and translations for the graded levels.
Characters
characters.json.gz · 178 KB4,603 characters with their radical, stroke count, Japanese and Mandarin readings, meanings, and the components each one is built from.
Japanese stroke order
strokes-ja.json.gz · 1.1 MBStroke paths, in order, for the Japanese characters and kana Bushu teaches.
Chinese stroke order
strokes-zh.json.gz · 4.6 MBStroke outlines and centrelines, in order, for the Chinese characters Bushu teaches.
Each file is gzip-compressed JSON. The notices file credits every source and says how each was modified, and the Arphic Public License travels with the Chinese stroke data, as that licence requires. Bushu's levels are its own, derived from word frequency; they are not JLPT or HSK levels.