Open data

Daolang took its character strokes, character parts and word meanings from open projects and changed them for the course. The changed files are here, and you can use them under the same terms.

Character strokes

Stroke outlines and stroke order, for writing and for taking characters apart in the app. Source: the graphics.txt file of the Make Me a Hanzi project (Shaunak Kishore), whose outlines come from the Arphic PL KaitiM GB and Arphic PL UKai fonts (Arphic Technology). Licence: Arphic Public License — text, Chinese text.

• strokes.json: 1,089 characters, last changed 2026-10-18.
• strokes-extra.json: 436 more characters, last changed 2026-10-18.

What we changed: we kept only the characters the course needs and rewrote the data as JSON — the character, its stroke outlines and their median lines. The changes were made from 2026-08-10 to 2026-10-18.

Character parts

Which parts a character is made of, for the build-a-character exercises and character breakdowns. Source: the dictionary.txt file of the Make Me a Hanzi project (Shaunak Kishore), based on the Unihan database (Unicode, Inc.) and cjklib. Licence: GNU LGPL 3.0 or any later version, with the Unicode notice for the Unihan data — LGPL 3.0 and the Unicode notice, GPL 3.0, Make Me a Hanzi COPYING.

• components.json: 2,987 characters, last changed 2026-10-18.
• components-extra.json: 57 more characters, last changed 2026-10-18.

What we changed: we kept the characters the course needs; for each one we added its parts and their stroke numbers, a reading, and a meaning in Russian and English; the data is rewritten as JSON. The changes were made from 2026-08-28 to 2026-10-18.

Word meanings

Daolang's English word meanings are derived from the CC-CEDICT dictionary (MDBG and the CC-CEDICT contributors): we selected and shortened them for the course. The Spanish, Japanese and Korean meanings are translated from these English ones. All of them are shared under CC BY-SA 4.0.

• glosses-en.json: English, 4,807 words.
• glosses-es.json: Spanish, 4,807 words.
• glosses-ja.json: Japanese, 4,807 words.
• glosses-ko.json: Korean, 4,807 words.

How to use the files

The files are JSON in UTF-8. The first key of each file, "@notice", says where the data comes from, which licence it is under and what was changed. You may copy, change and share the files under the terms of their licences. Questions: help@daolang.app. All the other sources of the app are listed on the licences page.