v2.3.2: 优化正则表达式，异常处理，取消master的GitHub Actions限制 (#141)

Author: hect0x7

.github/workflows/download.yml

@@ -3,8 +3,6 @@ name: 下载JM本子
 on:
   workflow_dispatch:
   push:
-    branches-ignore:
-      - 'master' # master专门用于发布pip和repo介绍
     paths:
       - '.github/workflows/download.yml' # 工作流定义
       - 'usage/workflow_download.py' # 下载脚本

src/jmcomic/__init__.py

@@ -2,7 +2,7 @@
 # 被依赖方 <--- 使用方
 # config <--- entity <--- toolkit <--- client <--- option <--- downloader
 
-__version__ = '2.3.1'
+__version__ = '2.3.2'
 
 from .api import *
 from .jm_plugin import *

src/jmcomic/jm_config.py

@@ -17,7 +17,7 @@ def default_postman_constructor(session, **kwargs):
     return Postmans.new_postman(**kwargs)
 
 
-def default_raise_regex_error(msg, *_args, **_kwargs):
+def default_raise_exception_executor(msg, **_kwargs):
     raise JmModuleConfig.exception(msg)
 
 
@@ -76,7 +76,7 @@ class JmModuleConfig:
     # postman构造函数
     postman_constructor = default_postman_constructor
     # 网页正则表达式解析失败时，执行抛出异常的函数，可以替换掉用于debug
-    raise_regex_error_executor = default_raise_regex_error
+    raise_exception_executor = default_raise_exception_executor
 
     # debug开关标记
     enable_jm_debug = True
@@ -135,11 +135,25 @@ def client_impl_class(cls, client_key: str):
 
     @classmethod
     def exception(cls, msg: str):
+        """
+        获取jmcomic模块的异常类
+        """
         if cls.CLASS_EXCEPTION is not None:
             return cls.CLASS_EXCEPTION(msg)
 
         return JmcomicException(msg)
 
+    @classmethod
+    def raises(cls, msg: str, **kwargs):
+        """
+        抛出异常，支持把一些上下文参数传递为kwargs
+        真正抛出异常的是函数 cls.raise_exception_executor，用户可自定义此字段
+
+        如果只想抛异常，不想支持一些扩展处理，使用 raise cls.exception(msg)
+        如果想支持一些扩展处理，使用 cls.raises(msg, context=context)
+        """
+        cls.raise_exception_executor(msg, **kwargs)
+
     @classmethod
     @field_cache("DOMAIN")
     def domain(cls, postman=None):

src/jmcomic/jm_toolkit.py

@@ -13,7 +13,7 @@ class JmcomicText:
     pattern_html_photo_name = compile('<title>([\s\S]*?)\|.*</title>')
     # pattern_html_photo_data_original_list = compile('data-original="(.*?)" id="album_photo_.+?"')
     pattern_html_photo_data_original_domain = compile('src="https://(.*?)/media/albums/blank')
-    pattern_html_photo_data_original_0 = compile('data-original="(.*?)"[ \n]*?id="album_photo')
+    pattern_html_photo_data_original_0 = compile('data-original="(.*?)"[^>]*?id="album_photo[^>]*?data-page="0"')
     pattern_html_photo_keywords = compile('<meta name="keywords"[\s\S]*?content="(.*?)"')
     pattern_html_photo_series_id = compile('var series_id = (\d+);')
     pattern_html_photo_sort = compile('var sort = (\d+);')
@@ -31,29 +31,29 @@ class JmcomicText:
     # 作品
     pattern_html_album_works = [
         compile('<span itemprop="author" data-type="works">([\s\S]*?)</span>'),
-        compile('<a[\s\S]*?>(.*?)</a>')
+        compile('<a[^>]*?>(.*?)</a>')
     ]
     # 登場人物
     pattern_html_album_actors = [
         compile('<span itemprop="author" data-type="actor">([\s\S]*?)</span>'),
-        compile('<a[\s\S]*?>(.*?)</a>')
+        compile('<a[^>]*?>(.*?)</a>')
     ]
     # 标签
     pattern_html_album_tags = [
         compile('<span itemprop="genre" data-type="tags">([\s\S]*?)</span>'),
-        compile('<a[\s\S]*?>(.*?)</a>')
+        compile('<a[^>]*?>(.*?)</a>')
     ]
     # 作者
     pattern_html_album_authors = [
         compile('作者： *<span itemprop="author" data-type="author">([\s\S]*?)</span>'),
-        compile("<a[\s\S]*?>(.*?)</a>"),
+        compile("<a[^>]*?>(.*?)</a>"),
     ]
     # 點擊喜歡
     pattern_html_album_likes = compile('<span id="albim_likes_\d+">(.*?)</span>')
     # 觀看
     pattern_html_album_views = compile('<span>(.*?)</span> (次觀看|观看次数)')
     # 評論(div)
-    pattern_html_album_comment_count = compile('<div class="badge"\n? *id="total_video_comments">(\d+)</div>'), 0
+    pattern_html_album_comment_count = compile('<div class="badge"[^>]*?id="total_video_comments">(\d+)</div>'), 0
 
     @classmethod
     def parse_to_jm_domain(cls, text: str):
@@ -169,11 +169,11 @@ def match_field(field_key: str, pattern: Union[Pattern, List[Pattern]], text):
 
             if field_value is None:
                 if default is None:
-                    JmModuleConfig.raise_regex_error_executor(
+                    JmModuleConfig.raises(
                         f"文本没有匹配上字段：字段名为'{field_name}'，pattern: [{pattern}]",
-                        html,
-                        field_name,
-                        pattern
+                        re_match_html=html,
+                        re_match_field_name=field_name,
+                        re_match_pattern=pattern,
                     )
                 else:
                     field_value = default
@@ -260,12 +260,12 @@ def parse_html_to_page(cls, html: str) -> JmSearchPage:
         match = cls.pattern_html_search_error.search(html)
         if match is not None:
             topic, reason = match[1], match[2]
-            JmModuleConfig.raise_regex_error_executor(f'{topic}: {reason}', html)
+            JmModuleConfig.raises(f'{topic}: {reason}', re_search_html=html)
 
         # 缩小文本范围
         match = cls.pattern_html_search_shorten_for.search(html)
         if match is None:
-            JmModuleConfig.raise_regex_error_executor('未匹配到搜索结果', html)
+            JmModuleConfig.raises('未匹配到搜索结果', re_shorten_html=html)
         html = match[0]
 
         # 提取结果