ok

Author: ipcjs

.gitignore

@@ -1 +1,5 @@
 *.zip
+/out/
+/.vscode/
+__pycache__/
+

README.md

@@ -1 +1,8 @@
-# wechat-emoji-dict
+# wechat-emoji-dict
+
+将所有微信表情导出成如下字典文件:
+
+- GBoard(全拼)
+- GBoard(小鹤双拼)
+
+![preview](resources/preview.png)

flypy.py

@@ -0,0 +1,94 @@
+# copy from https://github.com/bcaso/pinyin_to_double_pinyin/blob/main/dp.py
+
+first = {'ch': 'i',
+         'sh': 'u',
+         'zh': 'v'}
+
+second = {
+    'ua': 'x',
+    'ei': 'w',
+    'e': 'e',
+    'ou': 'z',
+    'iu': 'q',
+    've': 't',
+    'ue': 't',
+    'u': 'u',
+    'i': 'i',
+    'o': 'o',
+    'uo': 'o',
+    'ie': 'p',
+    'a': 'a',
+    'ong': 's',
+    'iong': 's',
+    'ai': 'd',
+    'ing': 'k',
+    'uai': 'k',
+    'ang': 'h',
+    'uan': 'r',
+    'an': 'j',
+    'en': 'f',
+    'ia': 'x',
+    'iang': 'l',
+    'uang': 'l',
+    'eng': 'g',
+    'in': 'b',
+    'ao': 'c',
+    'v': 'v',
+    'ui': 'v',
+    'un': 'y',
+    'iao': 'n',
+    'ian': 'm'
+}
+
+# 特殊，只有䪨母，且总长不过 3
+# 零声母，单双三䪨母
+special = {
+    'a': 'aa',
+    'ai': 'ai',
+    'an': 'an',
+    'ang': 'ah',
+    'ao': 'ao',
+    'e': 'ee',
+    'ei': 'ei',
+    'en': 'en',
+    'er': 'er',
+    'o': 'oo',
+    'ou': 'ou'
+}
+
+
+def todouble(s: str) -> str:
+    """
+    传入单汉字的全拼编码，反回其小鹤双拼编码
+
+    :param s: 全拼编码
+    :return: 双拼编码
+    """
+    new_s = ''
+    # 特列情况: 无声母，a, an, ang
+    if len(s) <= 3 and s[0] in ['a', 'e', 'o']:
+        if s in special.keys():
+            return special[s]
+        else:
+            print('未知情况1', s)
+
+    # 一般: 声母 + 䪨母
+
+    # 最长的情况：first+second，例如 chuang = ch + uang
+    # 2 位声母 + 最多 4 位韵母
+    if s[:2] in first.keys():
+        new_s += first[s[:2]]
+        # 最多 4 位䪨母
+        if s[2:] in second.keys():
+            new_s += second[s[2:]]
+    # 较短的情况：second+second，例如 h uang, x iang
+    # 1 位声母 + 最多 4 位䪨母
+    else:
+        new_s += s[0]  # 1 位声母
+        # 最多 4 位䪨母
+        if s[1:] in second.keys():
+            new_s += second[s[1:]]
+        else:
+            new_s += s[1:]
+
+    return new_s

main.py

@@ -0,0 +1,54 @@
+from pypinyin import pinyin, lazy_pinyin, Style
+from zipfile import ZipFile
+import flypy
+import re
+
+_RE_ENGLISH = re.compile(r'\w+', re.ASCII)
+
+
+def kanji_to_pinyin(kanji: str, py_type='pinyin') -> str:
+    if _RE_ENGLISH.fullmatch(kanji):
+        # 纯英文字母, 转小写就行
+        return kanji.lower()
+    py = lazy_pinyin(kanji, style=Style.NORMAL)
+    if py_type == 'flypy':
+        py = list(map(lambda s: flypy.todouble(s), py))
+    return ''.join(py)
+
+
+def save_to_gboard(list, py_type='pinyin', lang='zh-CN'):
+    base_path = f'out/wechat-emoji-gboard-{py_type}'
+    dict_file = base_path+'.txt'
+    zip_file = base_path + '.zip'
+
+    with open(dict_file, 'w', encoding='utf-8') as f:
+        f.write('# Gboard Dictionary version:1\n')
+        for item in list:
+            f.write('\t'.join((item[py_type], item['word'], lang)) + '\n')
+    # archive
+    with ZipFile(zip_file, 'w') as f:
+        f.write(dict_file, 'dictionary.txt')
+
+
+def main():
+    list = []
+    with open('source.txt', 'r', encoding='utf-8') as f:
+        for line in f:
+            segments = line.rstrip('\n').split('\t')
+            word = segments[0]
+            kanji_list = [word.strip('[]')]
+            if len(segments) > 1:
+                kanji_list.extend(segments[1].split(','))
+            for kanji in kanji_list:
+                list.append({
+                    'word': word,
+                    'pinyin': kanji_to_pinyin(kanji),
+                    'flypy': kanji_to_pinyin(kanji, py_type='flypy'),
+                })
+    save_to_gboard(list)
+    save_to_gboard(list, py_type='flypy')
+
+
+if __name__ == '__main__':
+    main()
+    print('completed.')

resources/preview.png

@@ -23,10 +23,10 @@
 [傲慢]
 [困]
 [惊恐]
-[憨笑]
+[憨笑]	笑
 [悠闲]
 [咒骂]
-[疑问]
+[疑问]	问
 [嘘]
 [晕]
 [衰]
@@ -47,12 +47,12 @@
 [笑脸]
 [生病]
 [脸红]
-[破涕为笑]
+[破涕为笑]	笑哭
 [恐惧]
 [失望]
 [无语]
 [嘿哈]
-[捂脸]
+[捂脸]	笑,无语
 [奸笑]
 [机智]
 [皱眉]
@@ -63,8 +63,8 @@
 [天啊]
 [Emm]
 [社会社会]
-[旺柴]
-[好的]
+[旺柴]	狗头
+[好的]	ok
 [打脸]
 [哇]
 [翻白眼]