Add dir_rule.normalize_zh option (default 'zh-cn') and implement configurable zh conversion; add JmcomicText.to_zh

Author: hect0x7

src/jmcomic/jm_config.py

@@ -411,7 +411,7 @@ def new_postman(cls, session=False, **kwargs):
 
     DEFAULT_OPTION_DICT: dict = {
         'log': None,
-        'dir_rule': {'rule': 'Bd_Pname', 'base_dir': None},
+    'dir_rule': {'rule': 'Bd_Pname', 'base_dir': None, 'normalize_zh': 'zh-cn'},
         'download': {
             'cache': True,
             'image': {'decode': True, 'suffix': None},

src/jmcomic/jm_option.py

@@ -61,10 +61,16 @@ def enable_client_cache_on_condition(cls,
 class DirRule:
     RULE_BASE_DIR = 'Bd'
 
-    def __init__(self, rule: str, base_dir=None):
+    def __init__(self, rule: str, base_dir=None, normalize_zh='zh-cn'):
+        """
+        :param rule: DSL rule
+        :param base_dir: base directory
+        :param normalize_zh: 'zh-cn'|'zh-tw'|'none' or None. 控制是否以及如何进行繁简体归一化，默认 'zh-cn'
+        """
         base_dir = JmcomicText.parse_to_abspath(base_dir)
         self.base_dir = base_dir
         self.rule_dsl = rule
+        self.normalize_zh = normalize_zh
         self.parser_list: List[Tuple[str, Callable]] = self.get_rule_parser_list(rule)
 
     def decide_image_save_dir(self,
@@ -89,14 +95,18 @@ def apply_rule_to_path(self, album, photo, only_album_rules=False) -> str:
                 jm_log('dir_rule', f'路径规则"{rule}"的解析出错: {e}, album={album}, photo={photo}')
                 raise e
             if parser != self.parse_bd_rule:
-                # 统一将路径段转换为简体，避免繁体/简体导致的重复下载目录
+                # 根据配置 normalize_zh 进行繁简体统一或跳过
                 try:
-                    path = JmcomicText.to_zh_cn(str(path))
+                    target = getattr(self, 'normalize_zh', None)
+                    if target is None:
+                        # 默认为不转换
+                        conv_path = str(path)
+                    else:
+                        conv_path = JmcomicText.to_zh(str(path), target)
                 except Exception:
-                    # 如果转换不可用（例如缺少zhconv），退回原字符串
-                    path = str(path)
+                    conv_path = str(path)
 
-                path = fix_windir_name(path).strip()
+                path = fix_windir_name(conv_path).strip()
 
             path_ls.append(path)
 
@@ -209,6 +219,7 @@ def copy_option(self):
             dir_rule={
                 'rule': self.dir_rule.rule_dsl,
                 'base_dir': self.dir_rule.base_dir,
+                'normalize_zh': getattr(self.dir_rule, 'normalize_zh', None),
             },
             download=self.download.src_dict,
             client=self.client.src_dict,
@@ -334,6 +345,7 @@ def deconstruct(self) -> Dict:
             'dir_rule': {
                 'rule': self.dir_rule.rule_dsl,
                 'base_dir': self.dir_rule.base_dir,
+                'normalize_zh': getattr(self.dir_rule, 'normalize_zh', None),
             },
             'download': self.download.src_dict,
             'client': self.client.src_dict,

src/jmcomic/jm_toolkit.py

@@ -329,8 +329,34 @@ def find_right_pair(left_pair, i):
 
     @classmethod
     def to_zh_cn(cls, s):
-        import zhconv
-        return zhconv.convert(s, 'zh-cn')
+        # 兼容旧接口，默认转换为简体
+        return cls.to_zh(s, 'zh-cn')
+
+    @classmethod
+    def to_zh(cls, s, target='zh-cn'):
+        """
+        通用的繁简体转换接口。
+
+        :param s: 待转换字符串
+        :param target: 目标编码: 'zh-cn'（简体）, 'zh-tw'（繁体），或 None/'none' 表示不转换
+        :return: 转换后的字符串（若转换失败或未安装 zhconv，返回原始字符串）
+        """
+        if s is None:
+            return s
+
+        if target is None:
+            return s
+
+        t = str(target).strip().lower()
+        if t in ('none', ''):
+            return s
+
+        try:
+            import zhconv
+            return zhconv.convert(s, t)
+        except Exception:
+            # 如果 zhconv 不可用或转换失败，则回退原字符串
+            return s
 
     @classmethod
     def try_mkdir(cls, save_dir: str):