def main():
"""
预测韦氏所有声韵母组合.
IN: more/韦氏拼音表_{声母,韵母,零声母}.txt
OT: 非标准韦氏拼音表.txt
注意: 生成的文件中含有大量的不可靠韦氏拼音, 有些是形而似的, 有些甚至
毫无逻辑, 可能对英文单词形成干扰. 请谨慎使用.
请在脚本生成此文件后, 对此文件进行人工清理.
之后不建议再通过本脚本生成. 以后的维护工作全部转到 "非标准韦氏拼音表
.txt" 本身由人工维护.
"""
initials = read_and_write.read_lines('more/韦氏拼音表_声母.txt')
finals = read_and_write.read_lines('more/韦氏拼音表_韵母.txt')
aoe = read_and_write.read_lines('more/韦氏拼音表_零声母.txt')
combo = []
for i in initials:
for j in finals:
combo.append(i + j)
combo.extend(aoe)
combo = list(set(combo))
# extra (customs)
combo.extend(['xean'])
combo.sort()
read_and_write.write_file(combo, '非标准韦氏拼音表.txt')
def generate_trie_file(*ifiles, ofile, check_duplicate=False):
if len(ifiles) == 1:
words = read_and_write.read_lines(ifiles[0])
trie.add_list(words)
else:
words_sum = []
for f in ifiles:
words = read_and_write.read_lines(f)
for w in words:
if w in words_sum:
if check_duplicate:
lk.logt('[W5317]', w, f, h='parent')
else:
words_sum.append(w)
words_sum.sort()
trie.add_list(words_sum)
trie.save(ofile)
trie.clear()
lk.loga('file saved', ofile)
def generate_postrie_file(*ifiles, ofile, check_duplicate=False):
if len(ifiles) == 1:
words = read_and_write.read_lines(ifiles[0])
nwords = [x[::-1] for x in words] # 反转字符串. 'ABC' -> 'CBA'
trie.add_list(nwords)
else:
words_sum = []
for f in ifiles:
words = read_and_write.read_lines(f)
for w in words:
w = w[::-1]
if w in words_sum:
if check_duplicate:
lk.logt('[W5318]', w, f, h='parent')
else:
words_sum.append(w)
words_sum.sort()
trie.add_list(words_sum)
trie.save(ofile)
trie.clear()
lk.loga('file saved', ofile)
def generate_plain_file(*ifiles, ofile, check_duplicate=False):
if len(ifiles) == 1:
copyfile(ifiles[0], ofile)
lk.loga('file saved', ofile)
return
data_sum = []
for f in ifiles:
data = read_and_write.read_lines(f)
for d in data:
if d in data_sum:
if check_duplicate:
lk.logt('[W5316]', d, f, h='parent')
else:
data_sum.append(d)
data_sum.sort()
read_and_write.write_file(data_sum, ofile)
lk.loga('file saved', ofile)
def main():
"""
通过大量的样本来预测可能的 "模棱两可" 的拼音的特征.
"""
i_file = 'train_in.txt'
o_file = 'train_out.txt'
aoe = ("a", "ai", "an", "ang", "ao", "o", "ou", "e", "ei", "en", "eng",
"er")
r = read_and_write.read_lines(i_file)
w = []
for i in r:
for j in aoe:
if i.endswith(j) and i.rsplit(j, 1)[0] in r:
w.append(f'{i}\t{i[:-1 * len(j)]}\t{j}')
break
read_and_write.write_file(w, o_file)