rime-ice/rime_ice.schema.yaml

382 lines
15 KiB
YAML
Raw Normal View History

2022-10-30 16:47:40 +01:00
# Rime schema
# encoding: utf-8
# 方案说明
schema:
schema_id: rime_ice
name: 雾凇拼音
version: "2024-01-30"
2022-10-30 16:47:40 +01:00
author:
- Dvel
description: |
2022-11-16 03:50:59 +01:00
雾凇拼音
2022-10-30 16:47:40 +01:00
https://github.com/iDvel/rime-ice
dependencies:
- melt_eng # 英文输入,作为次翻译器挂载到拼音方案
- radical_pinyin # 部件拆字,反查及辅码
2022-10-30 16:47:40 +01:00
2022-12-02 07:14:59 +01:00
# 开关
# states: 方案选单显示的名称。可以注释掉,仍可以通过快捷键切换。
# reset: 默认状态。注释掉后,切换窗口时不会重置到默认状态。
2022-10-30 16:47:40 +01:00
switches:
- name: ascii_mode
states: [ 中, ]
- name: ascii_punct # 中英标点
states: [ ¥, $ ]
2022-10-30 16:47:40 +01:00
- name: traditionalization
states: [ 简, 繁 ]
2022-10-30 16:47:40 +01:00
- name: emoji
states: [ 💀, 😄 ]
2023-03-01 05:19:27 +01:00
reset: 1
- name: full_shape
states: [ 半角, 全角 ]
- name: single_char # search.lua 的功能开关,辅码查词时是否单字优先
abbrev: [词, 单]
states: [正常, 单字]
2022-10-30 16:47:40 +01:00
2023-02-08 19:06:20 +01:00
# 输入引擎
2022-10-30 16:47:40 +01:00
engine:
processors:
- lua_processor@select_character # 以词定字
- lua_processor@auto_commit_single_punct # 让写在 alphabet 中的某标点自动上屏
2022-10-30 16:47:40 +01:00
- ascii_composer
- recognizer
- key_binder
- speller
- punctuator
- selector
- navigator
- express_editor
segmentors:
- ascii_segmentor
- matcher
- abc_segmentor
- affix_segmentor@radical_lookup # 部件拆字自定义 tag
2022-10-30 16:47:40 +01:00
- punct_segmentor
- fallback_segmentor
translators:
- punct_translator
- script_translator
- lua_translator@date_translator # 时间、日期、星期
- lua_translator@lunar # 农历
- table_translator@custom_phrase # 自定义短语 custom_phrase.txt
- table_translator@melt_eng # 英文输入
- table_translator@radical_lookup # 部件拆字反查
- lua_translator@unicode # Unicode
- lua_translator@number_translator # 数字、金额大写
2022-10-30 16:47:40 +01:00
filters:
- lua_filter@corrector # 错音错字提示
- simplifier@emoji # Emoji
- simplifier@traditionalize # 简繁切换
- reverse_lookup_filter@radical_reverse_lookup # 部件拆字滤镜
- lua_filter@search@radical_pinyin # 部件拆字辅码
- lua_filter@v_filter # v 模式 symbols 优先(否则是英文优先)
- lua_filter@autocap_filter # 英文自动大写
- lua_filter@reduce_english_filter # 降低部分英语单词在候选项的位置
- uniquifier # 去重
2022-10-30 16:47:40 +01:00
# Lua 配置: 日期、时间、星期、ISO 8601、时间戳的触发关键字
date_translator:
date: rq # 日期: 2022-11-29
time: sj # 时间: 18:13
week: xq # 星期: 星期二
datetime: dt # ISO 8601 2022-11-29T18:13:11+08:00
timestamp: ts # 时间戳: 1669716794
# Lua 配置:农历的触发关键字
lunar: nl # 农历: 二〇二三年冬月二十 癸卯年(兔)冬月二十
# Lua 配置:直接上屏单个字符
# lua_filter@search@radical_pinyin 辅码的引导符在 key_binder/search 定义,目前为 `(反引号)。
# ` 只在中间时起辅码查询作用,但是输入单个 ` 时仍然需要按空格选择,这个 Lua 让单个的 ` 直接上屏。
auto_commit_single_punct: "`"
# Lua 配置:为 corrector 格式化 comment占位符为 {comment}
# 默认 "{comment}" 输入 hun dun 时会在「馄饨」旁边生成 hún tun 的 comment
# 例如左右加个括号 "({comment})" 就会变成 (hún tun)
corrector: "{comment}"
2024-01-02 15:25:13 +01:00
# Lua 配置: 降低部分英语单词在候选项的位置。
# 详细介绍 https://dvel.me/posts/make-rime-en-better/#短单词置顶的问题
# 正常情况: 输入 rug 得到 「1.rug 2.如果 …… 」
# 降低之后: 输入 rug 得到 「1.如果 2.rug …… 」
# 几种模式:
# all 降低脚本内置的单词(所有 3~4 位长度、前 2~3 位是完整拼音、最后一位是声母words 作为自定义的额外补充
# custom 完全自定义,只降低 words 里的
# none 不降低任何单词,相当于没有启用这个 Lua
# (匹配的是编码,不是单词)
reduce_english_filter:
mode: custom # all | custom | none
idx: 2 # 降低到第 idx 个位置
# 自定义的单词列表,示例列表没有降低部分常用单词,如 and cat mail Mac but bad shit ……
words: [aid, ann,
bail, bait, bam, band, bans, bat, bay, bend, bent, benz, bib, bid, bien, biz, boc, bop, bos, bud, buf,
cab, cad, cain, cam, cans, cap, cas, cef, chad, chan, chap, chef, cher, chew, chic, chin, chip, chit, coup, cum, cunt, cur,
dab, dag, dal, dam, dent, dew, dial, diet, dim, din, dip, dis, dit, doug, dub, dug, dunn,
fab, fax, fob, fog, foul, fur,
gag, gail, gain, gal, gam, gaol, ged, gel, ger, guam, gus, gut,
hail, ham, hank, hans, hat, hay, heil, heir, hem, hep, hud, hum, hung, hunk, hut,
jim, jug,
kat,
lab, lad, lag, laid, lam, laos, lap, lat, lax, lay, led, leg, lex, liam, lib, lid, lied, lien, lies, linn, lip, lit, liz, lob, lug, lund, lung, lux,
2024-01-02 13:41:11 +01:00
mag, maid, mann, mar, mat, med, mel, mend, mens, ment, mil, mins, mint, mob, moc, mop, mos, mot, mud, mug, mum, nail,
nap, nat, nay, neil, nib, nip, noun, nous, nun, nut,
pac, paid, pail, pain, pair, pak, pal, pam, pans, pant, pap, par, pat, paw, pax, pens, pic, pier, pies, pins, pint, pit, pix, pod, pop, pos, pot, pour, pow, pub,
2024-01-05 12:32:31 +01:00
rand, rant, rent, rep, res, ret, rex, rib, rid, rig, rim, rub, rug, rum, runc, runs,
sac, sail, sal, sam, sans, sap, saw, sax, sew, sham, shaw, shin, sig, sin, sip, sis, suit, sung, suns, sup, sur, sus,
tad, tail, taj, tar, tax, tec, ted, tel, ter, tex, tic, tied, tier, ties, tim, tin, tit, tour, tout, tum,
2023-11-24 13:08:27 +01:00
wag, wand, womens, wap, wax, weir, won,
yan, yen]
2022-12-02 07:14:59 +01:00
# 主翻译器,拼音
2022-10-30 16:47:40 +01:00
translator:
dictionary: rime_ice # 挂载词库 rime_ice.dict.yaml
spelling_hints: 8 # corrector.lua :为了让错音错字提示的 Lua 同时适配全拼双拼,将拼音显示在 comment 中
always_show_comments: true # corrector.lua Rime 默认在 preedit 等于 comment 时取消显示 comment这里强制一直显示供 corrector.lua 做判断用。
initial_quality: 1.2 # 拼音的权重应该比英文大
preedit_format: # preedit_format 影响到输入框的显示和“Shift+回车”上屏的字符
- xform/([jqxy])v/$1u/ # 显示为 ju qu xu yu
# - xform/([nl])v/$1ü/ # 显示为 nü lü
# - xform/([nl])ue/$1üe/ # 显示为 nüe lüe
- xform/([nl])v/$1v/ # 显示为 nv lv
- xform/([nl])ue/$1ve/ # 显示为 nve lve
2022-10-30 16:47:40 +01:00
2022-12-02 07:14:59 +01:00
# 次翻译器,英文
2022-10-30 16:47:40 +01:00
melt_eng:
2022-12-02 07:14:59 +01:00
dictionary: melt_eng # 挂载词库 melt_eng.dict.yaml
enable_sentence: false # 禁止造句
enable_user_dict: false # 禁用用户词典
initial_quality: 1.1 # 初始权重
2023-02-27 10:18:06 +01:00
comment_format: # 自定义提示码
- xform/.*// # 清空提示码
2022-10-30 16:47:40 +01:00
# 自定义短语custom_phrase.txt
custom_phrase:
dictionary: ""
user_dict: custom_phrase # 可以修改这里,改成自己的 txt 文件
2024-01-25 20:41:38 +01:00
db_class: stabledb # 只读数据库,无法动态调频;设为 tabledb 可以动态调频
enable_completion: false # 补全提示
enable_sentence: false # 禁止造句
initial_quality: 99 # custom_phrase 的权重应该比 pinyin 和 melt_eng 大
# Emoji
emoji:
option_name: emoji
opencc_config: emoji.json
inherit_comment: false # 在 corrector.lua 及反查中emoji 返回空注释
# 简繁切换
traditionalize:
option_name: traditionalization
opencc_config: s2t.json # s2t.json | s2hk.json | s2tw.json | s2twp.json
tips: none # 转换提示: all 都显示 | char 仅单字显示 | none 不显示。
tags: [ abc, number, gregorian_to_lunar ] # 限制在对应 tag不对其他如反查的内容做简繁转换
# 标点符号
# punctuator 下面有三个子项:
# full_shape 全角标点映射
# half_shape 半角标点映射
# symbols Rime 的预设配置是以 '/' 前缀开头输出一系列字符,自定义的 symbols_v.yaml 修改成了 'v' 开头。
2022-10-30 16:47:40 +01:00
punctuator:
full_shape:
__include: default:/punctuator/full_shape # 从 default.yaml 导入配置
half_shape:
__include: default:/punctuator/half_shape # 从 default.yaml 导入配置
symbols:
__include: symbols_v:/symbols # 从 symbols_v.yaml 导入配置
2022-10-30 16:47:40 +01:00
# 部件拆字反查
radical_lookup:
tag: radical_lookup
dictionary: radical_pinyin
enable_user_dict: false
prefix: "uU" # 反查前缀(反查时前缀会消失影响打英文所以设定为两个字母,或可改成一个非字母符号),与 recognizer/patterns/radical_lookup 匹配
tips: " 〔拆字〕"
comment_format:
- erase/^.*$//
# 部件拆字滤镜
radical_reverse_lookup:
tags: [ radical_lookup ]
dictionary: rime_ice # 拼音标注来源,可将 other/single_pinyin.reverse.bin 拷贝至 build/ 目录,此处改为 single_pinyin 来为所有字显示拼音
# comment_format: # 自定义 comment例如在左右加上括号
# - xform/^/(/
# - xform/$/)/
2022-10-30 16:47:40 +01:00
# 处理符合特定规则的输入码,如网址、反查
recognizer:
import_preset: default # 从 default.yaml 继承通用的
patterns: # 再增加方案专有的:
2024-01-14 17:14:26 +01:00
punct: "^v([0-9]|10|[A-Za-z]+)$" # 响应 symbols_v.yaml 的 symbols用 'v' 替换 '/'
radical_lookup: "^uU[a-z]+$" # 响应部件拆字的反查,与 radical_lookup/prefix 匹配
unicode: "^U[a-f0-9]+" # 脚本将自动获取第 2 个字符 U 作为触发前缀,响应 lua_translator@unicode输出 Unicode 字符
number: "^R[0-9]+[.]?[0-9]*" # 脚本将自动获取第 2 个字符 R 作为触发前缀,响应 lua_translator@number_translator数字金额大写
gregorian_to_lunar: "^N[0-9]{1,8}" # 脚本将自动获取第 2 个字符 N 作为触发前缀,响应 lua_translator@lunar公历转农历输入 N20240115 得到「二〇二三年腊月初五」
2022-10-30 16:47:40 +01:00
# 从 default 继承快捷键
key_binder:
import_preset: default # 从 default.yaml 继承通用的
search: "`" # 辅码引导符,要添加到 speller/alphabet
# bindings: # 也可以再增加方案专有的快捷键
2022-10-30 16:47:40 +01:00
# 拼写设定
speller:
# 如果不想让什么标点直接上屏,可以加在 alphabet或者编辑标点符号为两个及以上的映射
alphabet: zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA`
2022-12-02 07:14:59 +01:00
delimiter: " '" # 第一位<空格>是拼音之间的分隔符;第二位<'>表示可以手动输入单引号来分割拼音。
2022-10-30 16:47:40 +01:00
algebra:
### 模糊音
# 声母
# - derive/^([zcs])h/$1/ # zh* ch* sh* 派生出 z* c* s*
# - derive/^([zcs])([^h])/$1h$2/ # z* c* s* 派生出 zh* ch* sh*
# - derive/^l/n/ # 解释:为 l 开头的拼写派生出 n 开头,即 nai 也可以输出 lai来、莱、赖……
# - derive/^n/l/ # lai 可输出 nai奶、乃、奈……。 可以单向或成对儿启用模糊音
2023-08-15 16:22:22 +02:00
# - derive/^f/h/
# - derive/^h/f/
2022-10-30 16:47:40 +01:00
# - derive/^l/r/
# - derive/^r/l/
# - derive/^g/k/
# - derive/^k/g/
# 韵母
# - derive/ang$/an/
# - derive/an$/ang/
# - derive/eng$/en/
# - derive/en$/eng/
# - derive/in$/ing/
# - derive/ing$/in/
# - derive/ian$/iang/
# - derive/iang$/ian/
# - derive/uan$/uang/
# - derive/uang$/uan/
# - derive/ai$/an/
# - derive/an$/ai/
# - derive/ong$/un/
# - derive/un$/ong/
# - derive/ong$/on/
# - derive/iong$/un/
# - derive/un$/iong/
# - derive/ong$/eng/
# - derive/eng$/ong/
2022-10-30 16:47:40 +01:00
# 拼音音节
# - derive/^fei$/hui/
# - derive/^hui$/fei/
# - derive/^hu$/fu/
# - derive/^fu$/hu/
# - derive/^wang$/huang/
# - derive/^huang$/wang/
### 旧时的拼写规则
# - derive/un$/uen/
# - derive/ui$/uei/
# - derive/iu$/iou/
### 超级简拼
- erase/^hm$/ # 响应超级简拼,取消「噷 hm」的独占
- erase/^m$/ # 响应超级简拼,取消「呣 m」的独占
- erase/^n$/ # 响应超级简拼,取消「嗯 n」的独占
- erase/^ng$/ # 响应超级简拼,取消「嗯 ng」的独占
- abbrev/^([a-z]).+$/$1/ # 超级简拼
- abbrev/^([zcs]h).+$/$1/ # 超级简拼中zh ch sh 视为整体ch'sh → 城市而不是像这样分开c'h's'h → 吃好睡好)。
### v u 转换
# 雾凇的词库严格按照正确的 u v(ü) 注音的,如 qu 和 nve支持使用 qv 和 nue 来输入
# 再增加对其他词库的 qv nue 这种注音的支持
2022-10-30 16:47:40 +01:00
- derive/^([nl])ue$/$1ve/
- derive/^([nl])ve$/$1ue/
- derive/^([jqxy])u/$1v/
- derive/^([jqxy])v/$1u/
### 自动纠错
# 有些规则对全拼简拼混输有副作用如「x'ai 喜爱」被纠错为「xia 下」
# zh、ch、sh
- derive/([zcs])h(a|e|i|u|ai|ei|an|en|ou|uo|ua|un|ui|uan|uai|uang|ang|eng|ong)$/h$1$2/ # hzi → zhi
- derive/([zcs])h([aeiu])$/$1$2h/ # zih → zhi
# ai
2023-02-17 19:50:10 +01:00
- derive/^([wghk])ai$/$1ia/ # wia → wai
2022-10-30 16:47:40 +01:00
# ia
- derive/([qjx])ia$/$1ai/ # qai → qia
# ei
- derive/([wfghkz])ei$/$1ie/
2022-10-30 16:47:40 +01:00
# ie
- derive/([jqx])ie$/$1ei/
# ao
- derive/([rtypsdghklzcbnm])ao$/$1oa/
# ou
- derive/([ypfm])ou$/$1uo/
# uo
# an
- derive/([wrtypsdfghklzcbnm])an$/$1na/
2022-10-30 16:47:40 +01:00
# en
- derive/([wrpsdfghklzcbnm])en$/$1ne/
# ang
- derive/([wrtypsdfghklzcbnm])ang$/$1nag/
- derive/([wrtypsdfghklzcbnm])ang$/$1agn/
# eng
- derive/([wrtpsdfghklzcbnm])eng$/$1neg/
- derive/([wrtpsdfghklzcbnm])eng$/$1egn/
# ing
- derive/([qtypdjlxbnm])ing$/$1nig/
- derive/([qtypdjlxbnm])ing$/$1ign/
# ong
- derive/([rtysdghklzcn])ong$/$1nog/
- derive/([rtysdghklzcn])ong$/$1ogn/
# iao
- derive/([qtpdjlxbnm])iao$/$1ioa/
- derive/([qtpdjlxbnm])iao$/$1oia/
# ui
- derive/([rtsghkzc])ui$/$1iu/
# iu
- derive/([qjlxnm])iu$/$1ui/
# ian
- derive/([qtpdjlxbnm])ian$/$1ain/
# - derive/([qtpdjlxbnm])ian$/$1ina/ # 和「李娜、蒂娜、缉拿」等常用词有冲突
# in
- derive/([qypjlxbnm])in$/$1ni/
# iang
- derive/([qjlxn])iang$/$1aing/
- derive/([qjlxn])iang$/$1inag/
# ua
- derive/([g|k|h|zh|sh])ua$/$1au/
# uai
- derive/([g|h|k|zh|ch|sh])uai$/$1aui/
# - derive/([g|h|k|zh|ch|sh])uai$/$1uia/ # 和「会啊、追啊」等常用词有冲突
2022-10-30 16:47:40 +01:00
# uan
- derive/([qrtysdghjklzxcn])uan$/$1aun/
# - derive/([qrtysdghjklzxcn])uan$/$1una/ # 和「去哪、露娜」等常用词有冲突
# un
- derive/([qrtysdghjklzxc])un$/$1nu/
# ue
- derive/([nlyjqx])ue$/$1eu/
# uang
- derive/([g|h|k|zh|ch|sh])uang$/$1aung/
- derive/([g|h|k|zh|ch|sh])uang$/$1uagn/
- derive/([g|h|k|zh|ch|sh])uang$/$1unag/
- derive/([g|h|k|zh|ch|sh])uang$/$1augn/
# iong
- derive/([jqx])iong$/$1inog/
- derive/([jqx])iong$/$1oing/
- derive/([jqx])iong$/$1iogn/
- derive/([jqx])iong$/$1oign/
# 其他
- derive/([rtsdghkzc])o(u|ng)$/$1o/ # do → dou|dong
- derive/ong$/on/ # lon → long
- derive/([tl])eng$/$1en/ # ten → teng
- derive/([qwrtypsdfghjklzxcbnm])([aeio])ng$/$1ng/ # lng → lang、leng、ling、long