ila
b1b2e6c6fe
- ECDICT 常用子集（82,721 条，含中文释义）与 CMUdict ARPAbet→IPA 转写音标， 由 scripts/dict_prepare.py 从 sha256 pin 的源显式准备，运行时不联网。 - schema v12：放开 lexgo_dictionaries 单行约束并增加 provider 列，英英与英汉词典 可分别启用；查词合并时中文释义在前、英英释义随后。 - 音标只取可验证来源：优先 CMUdict 转写，否则保留 ECDICT 记法并标注来源， 两者都不可靠时不显示；含 ^ 等丢失首音的记法整条丢弃。 - 屈折形经 WordNet lemma 解析后回查中文词典，但不把词目音标复制到屈折形上。 - 管理端可区分两种词典并分别启停；学习端查词面板显示音标与中文释义。
66 lines
3.0 KiB
JSON
66 lines
3.0 KiB
JSON
{
|
|
"name": "简明英汉字典增强版常用子集 (ECDICT) + CMUdict 音标转写",
|
|
"language": "en",
|
|
"provider": "ecdict",
|
|
"format": "lexgo-zh-dict-v1",
|
|
"version": "ecdict-1.0.28+cmudict-0.7b",
|
|
"sources": [
|
|
{
|
|
"name": "ECDICT StarDict",
|
|
"role": "chinese-gloss",
|
|
"source": "https://github.com/skywind3000/ECDICT/releases/download/1.0.28/ecdict-stardict-28.zip",
|
|
"mirrors": [],
|
|
"sha256": "c707d0f3ded6ec79b96466da4a1574e074703da5af9c120fbad97f9cb08c6f2c",
|
|
"bytes": 70415834,
|
|
"license": "ECDICT-LICENSE.txt",
|
|
"licenseUrl": "https://github.com/skywind3000/ECDICT/blob/master/LICENSE",
|
|
"note": "仓库为 MIT；词典数据由公开资料整理，随项目再分发前需自行评估来源。子集按下方 subset 规则裁剪，不导入全量 340 万条。"
|
|
},
|
|
{
|
|
"name": "CMUdict",
|
|
"role": "ipa-phonetic",
|
|
"source": "https://raw.githubusercontent.com/cmusphinx/cmudict/master/cmudict.dict",
|
|
"mirrors": [
|
|
"https://cdn.jsdelivr.net/gh/cmusphinx/cmudict@master/cmudict.dict"
|
|
],
|
|
"sha256": "81917843c7f44ce2b094ac63873c2c7a4cf802040792c455ba3ca406891c3d22",
|
|
"bytes": 3618488,
|
|
"license": "CMUDICT-LICENSE.txt",
|
|
"licenseUrl": "https://github.com/cmusphinx/cmudict/blob/master/LICENSE",
|
|
"note": "BSD-2。ARPAbet 音素按固定映射表转写为 IPA；仅用于需要音标的词条。"
|
|
},
|
|
{
|
|
"name": "Princeton WordNet",
|
|
"role": "subset-wordlist",
|
|
"source": "https://raw.githubusercontent.com/nltk/nltk_data/96f9b3252457a2b97e52aec64c3dfceeb5c312d5/packages/corpora/wordnet.zip",
|
|
"mirrors": [
|
|
"https://cdn.jsdelivr.net/gh/nltk/nltk_data@96f9b3252457a2b97e52aec64c3dfceeb5c312d5/packages/corpora/wordnet.zip"
|
|
],
|
|
"sha256": "cbda5ea6eef7f36a97a43d4a75f85e07fccbb4f23657d27b4ccbc93e2646ab59",
|
|
"license": "WORDNET-LICENSE.txt",
|
|
"note": "与 server/wordnet-resource.json 同一 pin，仅用于取 lemma 词表决定子集范围，不复制其释义。"
|
|
}
|
|
],
|
|
"subset": {
|
|
"rule": "lowcase single alphabetic headword, unique, present in the WordNet lemma list or carrying an ECDICT exam tag (研/四/六/托/专/雅/高/中/初)",
|
|
"expectedEntries": 82721,
|
|
"toleranceEntries": 2000,
|
|
"measuredAt": "2026-09-16",
|
|
"measured": {
|
|
"entries": 82721,
|
|
"withCmudictIpa": 35799,
|
|
"withEcdictPhoneticOnly": 30580,
|
|
"withoutPhonetic": 16342,
|
|
"droppedEmpty": 45,
|
|
"rawPayloadMiB": 8.1
|
|
}
|
|
},
|
|
"phoneticRules": "prefer CMUdict ARPAbet->IPA mapping; otherwise keep the ECDICT notation with character-level normalisation only and mark the source; no phonetic rather than a guess",
|
|
"runtime": "Go only; the prepared ZIP is explicitly imported and persisted in MySQL; no runtime download and no Python process",
|
|
"preparedArtifact": {
|
|
"bytes": 3003208,
|
|
"sha256": "a4f4ab972c6e233e39909a861015c99ad7943e8db24353afc9cc9d8160dded24",
|
|
"note": "由 scripts/dict_prepare.py 在本机生成，不入库；每次重建都会打印自己的 sha256"
|
|
}
|
|
}
|