feat: 使用 opencc 将歌曲名转化为简体 (#192)

pyproject.toml

@@ -16,6 +16,7 @@ dependencies = [
     "aiofiles>=24.1.0",
     "ga4mp>=2.0.4",
     "apscheduler>=3.10.4",
+    "opencc-python-reimplemented==0.1.7",
 ]
 requires-python = ">=3.10,<3.12"
 readme = "README.md"

xiaomusic/utils.py

@@ -31,6 +31,10 @@ from mutagen.wave import WAVE
 from requests.utils import cookiejar_from_dict
 
 from xiaomusic.const import SUPPORT_MUSIC_TYPE
+from opencc import OpenCC
+
+
+cc = OpenCC('t2s')  # convert from Traditional Chinese to Simplified Chinese
 
 
 ### HELP FUNCTION ###
@@ -90,6 +94,10 @@ def fuzzyfinder(user_input, collection, extra_search_index=None):
                            extra_search_index=extra_search_index)
 
 
+def traditional_to_simple(to_convert: str):
+    return cc.convert(to_convert)
+
+
 # 关键词检测
 def keyword_detection(user_input, str_list, n):
     # 过滤包含关键字的字符串
@@ -119,15 +127,17 @@ def real_search(prompt, candidates, cutoff, n):
 
 
 def find_best_match(user_input, collection, cutoff=0.6, n=1, extra_search_index=None):
-    lower_collection = {item.lower(): item for item in collection}
-    user_input = user_input.lower()
+    lower_collection = {traditional_to_simple(item.lower()): item for item in collection}
+    user_input = traditional_to_simple(user_input.lower())
     matches = real_search(user_input, lower_collection.keys(), cutoff, n)
     cur_matched_collection = [lower_collection[match] for match in matches]
     if len(matches) >= n or extra_search_index is None:
         return cur_matched_collection[:n]
 
     # 如果数量不满足，继续搜索
-    lower_extra_search_index = {k.lower():v for k, v in extra_search_index.items()}
+    lower_extra_search_index = {
+        traditional_to_simple(k.lower()): v for k, v in extra_search_index.items()
+    }
     matches = real_search(user_input, lower_extra_search_index.keys(), cutoff, n)
     cur_matched_collection += [lower_extra_search_index[match] for match in matches]
     return cur_matched_collection[:n]