v2.1.3: 实现对禁漫图片URL中的v参数的模拟 (#78)

Author: hect0x7

.github/workflows/test.yml

@@ -3,10 +3,12 @@ name: 跑测试
 on:
   push:
     branches: [ "dev" ]
-    paths:
-      - 'assets/config/*.yml' # option配置文件
-      - 'src/**/*.py' # 源码
-      - 'tests/**/*.py' # 测试代码
+    paths-ignore:
+      - 'src/**/__init__.py'
+      - '.github/workflows/*'
+      - 'usage/workflow_download.py'
+      - 'assets/config/option_workflow_download.yml'
+      - 'assets/docs/*'
 
 jobs:
   test: # This code is based on https://github.com/gaogaotiantian/viztracer/blob/master/.github/workflows/python-package.yml

src/jmcomic/__init__.py

@@ -2,6 +2,6 @@
 # 被依赖方 <--- 使用方
 # config <--- entity <--- toolkit <--- client <--- option
 
-__version__ = '2.1.2'
+__version__ = '2.1.3'
 
 from .api import *

src/jmcomic/jm_client_impl.py

@@ -246,7 +246,7 @@ def get_if_fail_raise(url):
             """
             使用此方法包装 self.get
             """
-            resp = JmImageResp(self.get(url, params={'v': format_ts()}))
+            resp = JmImageResp(self.get(url))
 
             if resp.is_success:
                 return resp

src/jmcomic/jm_client_interface.py

@@ -237,7 +237,7 @@ def download_by_image_detail(self,
                                  decode_image=True,
                                  ):
         self.download_image(
-            img_detail.img_url,
+            img_detail.download_url,
             img_save_path,
             img_detail.scramble_id,
             decode_image=decode_image,

src/jmcomic/jm_entity.py

@@ -48,6 +48,7 @@ def __init__(self,
                  img_file_name,
                  img_file_suffix,
                  from_photo=None,
+                 query_params=None,
                  ) -> None:
         self.aid: str = aid
         self.scramble_id: str = scramble_id
@@ -56,17 +57,31 @@ def __init__(self,
         self.img_file_suffix: str = img_file_suffix
 
         self.from_photo: Optional[JmPhotoDetail] = from_photo
+        self.query_params: StrNone = query_params
 
     @property
-    def filename(self):
+    def filename(self) -> str:
         return self.img_file_name + self.img_file_suffix
 
+    @property
+    def download_url(self) -> str:
+        """
+        图片的下载路径
+        与 self.img_url 的唯一不同是，在最后会带上 ?{self.query_params}
+        @return:
+        """
+        if self.query_params is None:
+            return self.img_url
+
+        return f'{self.img_url}?{self.query_params}'
+
     @classmethod
     def of(cls,
            photo_id: str,
            scramble_id: str,
            data_original: str,
            from_photo=None,
+           query_params=None,
            ) -> 'JmImageDetail':
         """
         该方法用于创建 JmImageDetail 对象
@@ -85,6 +100,7 @@ def of(cls,
             img_file_name=data_original[x + 1:y],
             img_file_suffix=data_original[y:],
             from_photo=from_photo,
+            query_params=query_params,
         )
 
 
@@ -99,6 +115,7 @@ def __init__(self,
                  sort,
                  page_arr=None,
                  data_original_domain=None,
+                 data_original_0=None,
                  author=None,
                  from_album=None,
                  ):
@@ -111,17 +128,28 @@ def __init__(self,
 
         self._author: StrNone = author
         self.from_album: Optional[JmAlbumDetail] = from_album
+        self.index = self.album_index
 
         # 下面的属性和图片url有关
         if isinstance(page_arr, str):
             import json
             page_arr = json.loads(page_arr)
 
-        # 该photo的所有图片名 img_name
+        # page_arr存放了该photo的所有图片文件名 img_name
         self.page_arr: List[str] = page_arr
-        # 图片域名
+        # 图片的cdn域名
         self.data_original_domain: StrNone = data_original_domain
-        self.index = self.album_index
+        # 第一张图的URL
+        self.data_original_0 = data_original_0
+
+        # 2023-07-14
+        # 禁漫的图片url加上了一个参数v，如果没有带上这个参数v，图片会返回空数据
+        # 参数v的特点：
+        # 1. 值似乎是该photo的更新时间的时间戳，因此所有图片都使用同一个值
+        # 2. 值目前在网页端只在photo页面的图片标签的data-original属性出现
+        # 这里的模拟思路是，获取到第一个图片标签的data-original，
+        # 取出其query参数 → self.data_original_query_params, 该值未来会传递给 JmImageDetail
+        self.data_original_query_params = self.get_data_original_query_params(data_original_0)
 
     @property
     def is_single_album(self) -> bool:
@@ -179,7 +207,8 @@ def create_image_detail(self, index) -> JmImageDetail:
             self.photo_id,
             self.scramble_id,
             data_original,
-            from_photo=self
+            from_photo=self,
+            query_params=self.data_original_query_params,
         )
 
     def get_img_data_original(self, img_name: str) -> str:
@@ -194,6 +223,17 @@ def get_img_data_original(self, img_name: str) -> str:
 
         return f'https://{data_original_domain}/media/photos/{self.photo_id}/{img_name}'
 
+    # noinspection PyMethodMayBeStatic
+    def get_data_original_query_params(self, data_original_0: StrNone) -> str:
+        if data_original_0 is None:
+            return f'v={time_stamp()}'
+
+        index = data_original_0.rfind('?')
+        if index == -1:
+            return f'v={time_stamp()}'
+
+        return data_original_0[index + 1:]
+
     def __getitem__(self, item) -> JmImageDetail:
         return self.create_image_detail(item)
 

src/jmcomic/jm_toolkit.py

@@ -11,6 +11,7 @@ class JmcomicText:
     pattern_html_photo_title = compile('<title>(.*?)\|.*</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="(.*?)" id="album_photo')
     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+);')

tests/test_jmcomic/test_jm_api.py

@@ -102,7 +102,31 @@ def run(aid):
             )
 
     def test_get_jmcomic_url(self):
-        print(self.client.get_jmcomic_url())
-        print(self.client.get_jmcomic_domain_all())
-        print(JmModuleConfig.get_jmcomic_url())
-        print(JmModuleConfig.get_jmcomic_domain_all())
+        func_list = {
+            self.client.get_jmcomic_url,
+            self.client.get_jmcomic_domain_all,
+            JmModuleConfig.get_jmcomic_url,
+            JmModuleConfig.get_jmcomic_domain_all,
+        }
+
+        exception_list = []
+
+        def run_func_async(func):
+            try:
+                func()
+            except BaseException as e:
+                exception_list.append(e)
+                traceback_print_exec()
+
+        multi_thread_launcher(
+            iter_objs=func_list,
+            apply_each_obj_func=run_func_async,
+        )
+
+        if len(exception_list) == 0:
+            return
+
+        for e in exception_list:
+            print(e)
+
+        raise AssertionError(exception_list)