fix bugs

chpronounce/chp.py

@@ -24,6 +24,7 @@ def _append_phrase(self, ph, res):
     def _append_word(self, word, pos, res):
         if word in self.dic[1]:
             for dy, poses in self.dic[1][word]:
+
                 if pos in poses:
                     res.append(dy)
                     break
@@ -55,7 +56,7 @@ def get_duyin(self, sentence):
             if len(word) == 1:
                 self._append_word(word, pos, res)
             else:
-                if word in self.dic[len(word)]:
+                if len(word) in self.dic and word in self.dic[len(word)]:
                     self._append_phrase(word, res)
                 else:
                     sub = ""
@@ -64,13 +65,14 @@ def get_duyin(self, sentence):
                             self._break_down(sub, pos, res)
                             sub = ""
                         sub += c
-                        if len(sub) > 1 and sub in self.dic[len(sub)]:
+                        if len(word) in self.dic and len(sub) > 1 and sub in self.dic[len(sub)]:
                             self._append_phrase(sub, res)
                             sub = ""
                     self._break_down(sub, pos, res)
+
         if any(c in sentence for c in "不一"):
             for i in range(len(sentence)):
-                if any(sentence[i] == c for c in "不一") and i + 1 != len(sentence):
+                if sentence[i] in "不一" and i + 1 != len(sentence):
                     if res[i + 1][2] == 4:
                         py, zy, tone = res[i]
                         tone = 2

chpronounce/postag

@@ -2,7 +2,7 @@ n 名词 名
 t 时间词 副
 s 处所词 副
 f 方位词 副
-m 数词 X
+m 数词 名
 q 量词 量
 b 区别词 动
 r 代词 名
@@ -22,12 +22,12 @@ j 简称 缀
 h 前接成分 缀
 k 后接成分 缀
 g 语素 名
-x 非语素字 X
-w 标点符号 X
+x 非语素字 名
+w 标点符号 名
 nr 人名 名
 ns 地名 名
 nt 机构名称 名
-nx 外文字符 X
+nx 外文字符 名
 nz 其它专名 名
 vd 副动词 动
 vn 名动词 动

verify_dict.py

@@ -36,7 +36,7 @@ def main():
                     ignore_flag = True
                     break
             if ignore_flag:
-                continue
+                to_pop.append((word_len, word, 0))
 
             if not isinstance(lis, list):
                 lis = [lis]
@@ -80,7 +80,7 @@ def main():
                                 print("t |", "\t ".join(map(str, tones)))
                                 print("+++++++++++++++-----------------------------")
 
-                if len(cys) != word_len or len(tones) != word_len:
+                if len(cys) != word_len or len(tones) != word_len or len(pys) != word_len:
                     to_pop.append((word_len, word, iol))
                     logging.warning(f"{word}[{iol}] is invalid.")
                     continue
@@ -90,17 +90,16 @@ def main():
                     x = [y for y in x if y in CHEWINGS]
                     x = "".join(x)
                     if xx != x:
-
                         logging.warning(f"[{word}] {xx} ---> {x}")
                     cys[i] = x
 
                 ori = dic[word_len][word]
+                # if word_len == 1:
+                #     pys, cys, tones = pys[0], cys[0], tones[0]
                 if word_len == 1:
-                    pys, cys, tones = pys[0], cys[0], tones[0]
-                if isinstance(dic[word_len][word], list):
-                    dic[word_len][word][iol] = ((pys, cys, tones), pos) if pos != '' else (pys, cys, tones)
+                    dic[word_len][word][iol] = ((pys, cys, tones), pos)  # if pos != '' else (pys, cys, tones)
                 else:
-                    dic[word_len][word] = ((pys, cys, tones), pos) if pos != '' else (pys, cys, tones)
+                    dic[word_len][word] = (pys, cys, tones)  # if pos != '' else (pys, cys, tones)
 
                 if ori != dic[word_len][word]:
                     n_diff += 1
@@ -110,7 +109,8 @@ def main():
                 # input()
 
     for word_len, word, iol in to_pop:
-        dic[word_len].pop(word)
+        if word in dic[word_len]:
+            dic[word_len].pop(word)
 
     # summary
     total = 0