++

manual/zhuimh/main.py

@@ -27,17 +27,15 @@ class Zhuimh:
         # 模拟鼠标滚轮操作，滚动到页面底部
         page.mouse.wheel(0, scroll_distance)
 
-    def get_chapter_img(self, chapter_name_list, chapter_url_list):
+    def get_chapter_img(self, chapter_name_list, chapter_url_list, path):
         for chapter_name, chapter_url in zip(chapter_name_list, chapter_url_list):
             print(f'章节名: {chapter_name}, 章节url: {chapter_url}')
             with sync_playwright() as playwright:
-                browser = playwright.chromium.launch(headless=True)  # headless=False 可以开启浏览器界面，便于调试
+                browser = playwright.chromium.launch(headless=False)  # headless=False 可以开启浏览器界面，便于调试
                 page = browser.new_page()
 
                 page.goto(chapter_url)
 
-                time.sleep(1)
-
                 self.window_scroll(page)
 
                 # for _ in range(5):
@@ -47,21 +45,54 @@ class Zhuimh:
                 #
                 # page.wait_for_timeout(1000)
 
-                time.sleep(1)
+                page.wait_for_timeout(3)
 
                 element = page.query_selector('body > div.chpater-images')
 
                 links = element.query_selector_all('img')
 
-                chpater_img_links = []
+                time.sleep(5)
+
+                chapter_file = os.path.join(path, chapter_name)
+                if not os.path.exists(chapter_file):
+                    os.mkdir(chapter_file)
+
+                #图片后缀都是webp
+                img_suffix = '.webp'
+
+                img_count = 1
 
                 for link in links:
                     # 获取每个 <a> 标签的 href 属性
                     img_src = link.get_attribute('src')
                     if 'blob:' in img_src:
-                        chpater_img_links.append(img_src)
-
-                print(chpater_img_links)
+                        # 这里开始保存图片
+                        # 检测一下图片是否有下载过， 如果有就跳过
+                        img_name = str(img_count).zfill(4)+img_suffix
+                        img_path = os.path.join(chapter_file, img_name)
+                        if not os.path.exists(img_path):
+                            # 使用fetch API获取blob数据
+                            self.save_blob_as_file(page, img_src, img_path)
+                            img_count += 1
+                        else:
+                            img_count += 1
+
+    def save_blob_as_file(self, page, blob_url, file_path):
+        # 使用 playwright 的 evaluate 方法来获取 blob 数据
+        buffer = page.evaluate(f"""
+            () => {
+                const response = fetch('{blob_url}');
+                const blob = response.blob();
+                const reader = new FileReader();
+                reader.readAsArrayBuffer(blob);
+                return new Promise((resolve) => {
+                    reader.onloadend = () => resolve(reader.result);
+                });
+            }
+        """)
+        # 将 ArrayBuffer 转换为 Node.js 的 Buffer 对象
+        with open(file_path, 'wb') as file:
+            file.write(buffer)
 
     def get_chapter(self):
         with sync_playwright() as playwright:
@@ -102,11 +133,11 @@ class Zhuimh:
                 print('元素未找到')
                 exit(0)
 
-            return chapter_name_list, chapter_url_list
+            return chapter_name_list, chapter_url_list, path
 
     def main(self):
-        chapter_name_list, chapter_url_list = self.get_chapter()
-        self.get_chapter_img(chapter_name_list, chapter_url_list)
+        chapter_name_list, chapter_url_list, path = self.get_chapter()
+        self.get_chapter_img(chapter_name_list, chapter_url_list, path)
 
 
 if __name__ == '__main__':