feat 支持自定义词表

app/core/meta/metaanime.py

@@ -2,7 +2,7 @@ import re
 import zhconv
 import anitopy
 from app.core.meta.metabase import MetaBase
-from app.core.meta.release_groups import ReleaseGroupsMatcher
+from app.core.meta.releasegroup import ReleaseGroupsMatcher
 from app.utils.string import StringUtils
 from app.schemas.types import MediaType
 

app/core/meta/metabase.py

@@ -15,7 +15,9 @@ class MetaBase(object):
     """
     # 是否处理的文件
     isfile: bool = False
-    # 原字符串
+    # 原标题字符串
+    title: str = ""
+    # 识别用字符串
     org_string: Optional[str] = None
     # 副标题
     subtitle: Optional[str] = None
@@ -53,6 +55,8 @@ class MetaBase(object):
     video_encode: Optional[str] = None
     # 音频编码
     audio_encode: Optional[str] = None
+    # 应用的识别词信息
+    apply_words: Optional[List[str]] = None
 
     # 副标题解析
     _subtitle_flag = False

app/core/meta/metavideo.py

@@ -3,7 +3,7 @@ from pathlib import Path
 
 from app.core.config import settings
 from app.core.meta.metabase import MetaBase
-from app.core.meta.release_groups import ReleaseGroupsMatcher
+from app.core.meta.releasegroup import ReleaseGroupsMatcher
 from app.utils.string import StringUtils
 from app.utils.tokens import Tokens
 from app.schemas.types import MediaType

app/core/meta/releasegroup.py

@@ -1,5 +1,7 @@
 import regex as re
 
+from app.db.systemconfig_oper import SystemConfigOper
+from app.schemas.types import SystemConfigKey
 from app.utils.singleton import Singleton
 
 
@@ -8,8 +10,7 @@ class ReleaseGroupsMatcher(metaclass=Singleton):
     识别制作组、字幕组
     """
     __release_groups: str = None
-    custom_release_groups: str = None
-    custom_separator: str = None
+    # 内置组
     RELEASE_GROUPS: dict = {
         "0ff": ['FF(?:(?:A|WE)B|CD|E(?:DU|B)|TV)'],
         "1pt": [],
@@ -74,6 +75,7 @@ class ReleaseGroupsMatcher(metaclass=Singleton):
     }
 
     def __init__(self):
+        self.systemconfig = SystemConfigOper()
         release_groups = []
         for site_groups in self.RELEASE_GROUPS.values():
             for release_group in site_groups:
@@ -89,8 +91,10 @@ class ReleaseGroupsMatcher(metaclass=Singleton):
         if not title:
             return ""
         if not groups:
-            if self.custom_release_groups:
-                groups = f"{self.__release_groups}|{self.custom_release_groups}"
+            # 自定义组
+            custom_release_groups = self.systemconfig.get(SystemConfigKey.CustomReleaseGroups)
+            if custom_release_groups:
+                groups = f"{self.__release_groups}|{custom_release_groups}"
             else:
                 groups = self.__release_groups
         title = f"{title} "
@@ -100,12 +104,4 @@ class ReleaseGroupsMatcher(metaclass=Singleton):
         for item in re.findall(groups_re, title):
             if item not in unique_groups:
                 unique_groups.append(item)
-        separator = self.custom_separator or "@"
-        return separator.join(unique_groups)
-
-    def update_custom(self, release_groups: str = None, separator: str = None):
-        """
-        更新自定义制作组/字幕组，自定义分隔符
-        """
-        self.custom_release_groups = release_groups
-        self.custom_separator = separator
+        return "@".join(unique_groups)

app/core/meta/words.py

@@ -0,0 +1,118 @@
+from typing import List, Tuple
+
+import cn2an
+import regex as re
+
+from app.db.systemconfig_oper import SystemConfigOper
+from app.log import logger
+from app.schemas.types import SystemConfigKey
+from app.utils.singleton import Singleton
+
+
+class WordsMatcher(metaclass=Singleton):
+
+    def __init__(self):
+        self.systemconfig = SystemConfigOper()
+
+    def prepare(self, title: str) -> Tuple[str, List[str]]:
+        """
+        预处理标题，支持三种格式
+        1：屏蔽词
+        2：被替换词 => 替换词
+        3：前定位词 <> 后定位词 >> 偏移量（EP）
+        """
+        appley_words = []
+        # 读取自定义识别词
+        words: List[str] = self.systemconfig.get(SystemConfigKey.CustomIdentifiers) or []
+        for word in words:
+            if not word:
+                continue
+            try:
+                if word.count(" => "):
+                    # 替换词
+                    strings = word.split(" => ")
+                    title, message, state = self.__replace_regex(title, strings[0], strings[1])
+                elif word.count(" >> ") and word.count(" <> "):
+                    # 集偏移
+                    strings = word.split(" <> ")
+                    offsets = strings[1].split(" >> ")
+                    title, message, state = self.__episode_offset(title, strings[0], strings[1],
+                                                                  offsets[1])
+                else:
+                    # 屏蔽词
+                    title, message, state = self.__replace_regex(title, word, "")
+
+                if state:
+                    appley_words.append(word)
+                else:
+                    logger.error(f"自定义识别词替换失败：{message}")
+            except Exception as err:
+                print(str(err))
+
+        return title, appley_words
+
+    @staticmethod
+    def __replace_regex(title: str, replaced: str, replace: str) -> Tuple[str, str, bool]:
+        """
+        正则替换
+        """
+        try:
+            if not re.findall(r'%s' % replaced, title):
+                return title, "", False
+            else:
+                return re.sub(r'%s' % replaced, r'%s' % replace, title), "", True
+        except Exception as err:
+            print(str(err))
+            return title, str(err), False
+
+    @staticmethod
+    def __episode_offset(title: str, front: str, back: str, offset: str) -> Tuple[str, str, bool]:
+        """
+        集数偏移
+        """
+        try:
+            if back and not re.findall(r'%s' % back, title):
+                return title, "", False
+            if front and not re.findall(r'%s' % front, title):
+                return title, "", False
+            offset_word_info_re = re.compile(r'(?<=%s.*?)[0-9一二三四五六七八九十]+(?=.*?%s)' % (front, back))
+            episode_nums_str = re.findall(offset_word_info_re, title)
+            if not episode_nums_str:
+                return title, "", False
+            episode_nums_offset_str = []
+            offset_order_flag = False
+            for episode_num_str in episode_nums_str:
+                episode_num_int = int(cn2an.cn2an(episode_num_str, "smart"))
+                offset_caculate = offset.replace("EP", str(episode_num_int))
+                episode_num_offset_int = int(eval(offset_caculate))
+                # 向前偏移
+                if episode_num_int > episode_num_offset_int:
+                    offset_order_flag = True
+                # 向后偏移
+                elif episode_num_int < episode_num_offset_int:
+                    offset_order_flag = False
+                # 原值是中文数字，转换回中文数字，阿拉伯数字则还原0的填充
+                if not episode_num_str.isdigit():
+                    episode_num_offset_str = cn2an.an2cn(episode_num_offset_int, "low")
+                else:
+                    count_0 = re.findall(r"^0+", episode_num_str)
+                    if count_0:
+                        episode_num_offset_str = f"{count_0[0]}{episode_num_offset_int}"
+                    else:
+                        episode_num_offset_str = str(episode_num_offset_int)
+                episode_nums_offset_str.append(episode_num_offset_str)
+            episode_nums_dict = dict(zip(episode_nums_str, episode_nums_offset_str))
+            # 集数向前偏移，集数按升序处理
+            if offset_order_flag:
+                episode_nums_list = sorted(episode_nums_dict.items(), key=lambda x: x[1])
+            # 集数向后偏移，集数按降序处理
+            else:
+                episode_nums_list = sorted(episode_nums_dict.items(), key=lambda x: x[1], reverse=True)
+            for episode_num in episode_nums_list:
+                episode_offset_re = re.compile(
+                    r'(?<=%s.*?)%s(?=.*?%s)' % (front, episode_num[0], back))
+                title = re.sub(episode_offset_re, r'%s' % episode_num[1], title)
+            return title, "", True
+        except Exception as err:
+            print(str(err))
+            return title, str(err), False