码表 #
下载 #
为了最大限度的自定义，我们肯定要找到纯文本码表，官方给的 Rime 挂接都只有二进制文件，但是我找了找竟然发现搜狗五笔是纯文本码表：下载链接
码表去重以及转换为 Rime 格式 #
#include <iostream>
#include <fstream>
#include <map>
#include <vector>
using namespace std;
int main() {
ifstream in{"sougou.txt"};
string code, text;
multimap<string, pair<string, int>> mp;
int order = 0;
auto is_substr = [&](const string& a, const string& b) {
return a.substr(0, size(b)) == b || b.substr(0, size(a)) == a;
};
while (in >> code >> text) {
if (!mp.count(text)) {
mp.emplace(text, pair{code, order});
} else if (size(text) == 3) { // 单字保留多音字
auto it = mp.find(text);
bool dup = 0;
while (it!=end(mp) && it->first == text) {
if (is_substr(it->second.first, code)) {
dup = true;
if (size(code) < size(it->second.first))
it->second = {code, order};
}
++it;
}
if (!dup)
mp.emplace(text, pair{code, order});
} else if (size(text) == 6) { //两字词两码和四码优先
auto it = mp.find(text);
if (size(code) % 2 == 0 &&
(size(it->second.first) % 2 || it->second.first.size() > code.size()))
it->second = {code, order};
} else { //其他词短码优先
auto it = mp.find(text);
if (it->second.first.size() > code.size())
it->second = {code, order};
}
order++;
}
vector<tuple<string, int, string>> v;
for (const auto& [text, code] : mp)
v.emplace_back(code.first, code.second, text);
sort(begin(v), end(v));
ofstream out{"flypy.dict.yaml"};
out << R"(---
name: flypy
version: "0.1"
sort: original
use_preset_vocabulary: false
...
)";
for (const auto& [code, odder, text] : v) {
out << text << '\t' << code << '\n';
}
}Rime #
常用的文件有：
- 输入方案头文件
<方案标识>.schema.yaml- 定义一个输入方案
- 输入方案码表
<方案标识>.dict.yaml- 输入方案的码表（如果用到码表的话）
- 自定义全局设定
default.custom.yaml- 比如开启了哪些输入法
- 针对不同发行版的设定
<squirrel|weasel>.custom.yaml- 比如输入法皮肤
flypy.schema.yaml示例
# Rime schema settings
# encoding: utf-8
schema:
schema_id: flypy
name: 小鹤音形
version: "10.8.4"
author:
- 方案设计：何海峰 <flypy@qq.com>
description: |
小鹤音形输入法
punctuator:
import_preset: default
switches:
- name: ascii_mode
reset: 0
states: [ 中文，英文 ]
- name: full_shape
states: [ 半角，全角 ]
- name: ascii_punct
# states: [ 。，, ．， ]
reset: 0
engine:
processors:
- ascii_composer
- recognizer
- key_binder
- speller
- punctuator
- selector
- navigator
- express_editor
segmentors:
- ascii_segmentor
- matcher
- abc_segmentor
- punct_segmentor
- fallback_segmentor
translators:
- punct_translator
- table_translator
- table_translator@user_dict
- history_translator@history
filters:
- simplifier
- reverse_lookup_filter
- uniquifier
speller:
alphabet: "abcdefghijklmnopqrstuvwxyz;'"
initials: ';abcdefghijklmnopqrstuvwxyz'
finals: "'"
#delimiter: " '"
max_code_length: 4
auto_select: true #顶字上屏
auto_select_pattern: ^;.$|^\w{4}$
auto_clear: max_length #manual|auto|max_length 空码按下一键确认清屏 | 空码自动清 | 达到最长码时后码顶上清屏
translator:
dictionary: flypy
enable_charset_filter: false
enable_sentence: false
enable_completion: false # 编码提示开关
enable_user_dict: false
history:
input: ;f
size: 1 #重复前几次上屏
initial_quality: 1 #首选
user_dict:
dictionary: ""
user_dict: user_dict
db_class: tabledb
enable_sentence: false
enable_completion: false
initial_quality: -1 #优先级
key_binder:
import_preset: default #方案切换相关
bindings:
- {accept: bracketleft, send: Page_Up, when: paging} # [上翻页
- {accept: bracketright, send: Page_Down, when: has_menu} # ] 下翻页
- {accept: comma, send: comma, when: paging} #注销逗号翻页
- {accept: period, send: period, when: has_menu} #注销句号翻页
- {accept: semicolon, send: 2, when: has_menu} #分号次选
# - {accept: Release+semicolon, send: semicolon, when: has_menu} #如启用此行，则分号引导符号功能无效
- {accept: Release+period, send: period, when: composing} #句号顶屏
- {accept: Release+comma, send: comma, when: composing} #逗号顶屏
- {accept: "Tab", send: Escape, when: composing}
- {accept: "Shift_R", send: Escape, when: composing}
# - {accept: "Shift+space", toggle: full_shape, when: always} #切换全半角
- {accept: "Control+period", toggle: ascii_punct, when: always} #切换中英标点
# - {accept: "Control+j", toggle: simplification, when: always} #切换简繁
recognizer:
import_preset: default
patterns:
# reverse_lookup: "[a-z`]*`+[a-z`]*"反查应该不需要
menu:
page_size: 5 #候选项数
style:
horizontal: true #竖排为 false