HANZI multilingual learning definitions These downloadable JSON dictionaries are adaptations, separate from the application code. English en.v1.json: CC-CEDICT, MDBG and contributors, CC BY-SA 4.0. https://www.mdbg.net/chinese/dictionary?page=cc-cedict https://creativecommons.org/licenses/by-sa/4.0/ French fr.v1.json: CFDICT, David Houstin / Chine Informations and contributors, release 2024-12-14, CC BY-SA 3.0. https://chine.in/mandarin/dictionnaire/CFDICT/ https://creativecommons.org/licenses/by-sa/3.0/ German de.v1.json: HanDeDict, Gabor L Ugray and contributors; original project Dr. Michael Klaus Engel, Jan Hefti and the Chinese-German Society Hamburg. Release header 2026-11-16, CC BY-SA 3.0 (as specified in the actual data file). https://handedict.zydeo.net/ https://github.com/gugray/HanDeDict https://creativecommons.org/licenses/by-sa/3.0/ Spanish es.v1.json: jmjm93/cedict-translations, derived from CC-CEDICT, CC BY-SA 4.0. Upstream Spanish definitions were machine-translated; only a subset was checked by the upstream author. This is not a claim of complete human review. https://github.com/jmjm93/cedict-translations https://creativecommons.org/licenses/by-sa/4.0/ Adaptations by HANZI, 2026-11-16: selected current catalog word/reading pairs; normalized tone-marked/numbered pinyin keys, retained up to four useful senses, removed classifier-only references and German example sentences. Missing exact catalog readings have individually authored definitions; no runtime selection of the first same-spelling reading. Additions are shared under each dictionary's license above. Russian authored materials and their existing sources are unchanged. The complete adapted datasets are downloadable here as en.v1.json, fr.v1.json, es.v1.json and de.v1.json. Provenance, exact upstream hashes, authored additions, coverage and reproducible builder: sources/localization and scripts/build-word-locales.mjs in the project source distribution. The current catalog contains inherited malformed readings and truncated words. This localization does not rewrite those source records. Authored definitions describe the displayed Chinese spelling, including fixed legacy pairs. They are not a certification of the original pinyin or every upstream dictionary sense. Embedded public search-catalog metadata (IDs, Chinese, pinyin and original Russian glosses): HSK30 by Ivan Krasilnikov and contributors, including Shawky and Pleco Inc., MIT. https://github.com/ivankra/hsk30 Russian HSK dictionary by argb/hanzi-data, copyright 2019 argb, MIT. https://github.com/argb/hanzi-data Full copyright and permission notices are distributed in licenses/HSK30-MIT.txt and licenses/HANZI-DATA-MIT.txt. The Russian catalog metadata is newly copied into these public assets for localized searching and lossless assignment saving; its original values and stored user materials are not rewritten. HSK 2026 curriculum selection and word/reading target pairs additionally derive from profesorm/hsk30 (CC BY 4.0) and Zitou (zitou.app), hsk-chinese-uzbek-russian (CC BY 4.0). Local authored corrections are recorded in sources/hsk-2026-ru-overrides.json. https://github.com/profesorm/hsk30 https://zitou.app https://creativecommons.org/licenses/by/4.0/