From 5ae8489f4c688165a3f61626227d9acec4e34db4 Mon Sep 17 00:00:00 2001 From: ucmao <294323976@qq.com> Date: Tue, 8 Sep 2026 00:10:01 +0800 Subject: [PATCH] =?UTF-8?q?=E6=94=AF=E6=8C=81=E5=BE=AE=E4=BF=A1=E5=85=AC?= =?UTF-8?q?=E4=BC=97=E5=8F=B7=E5=A4=9A=E8=A7=86=E9=A2=91=E6=8F=90=E5=8F=96?= =?UTF-8?q?=E4=B8=8E=E6=B8=85=E6=99=B0=E5=BA=A6=E6=8B=A9=E4=BC=98?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- README.md | 104 ++++++++++++++++---------------- docs/parsers/wechat-mp.md | 13 ++-- src/parsers/wechat_mp_parser.py | 64 ++++++++++++++------ tests/live_parser_samples.json | 17 ++++++ tests/test_wechat_mp_parser.py | 77 +++++++++++++++++++++++ 5 files changed, 200 insertions(+), 75 deletions(-) diff --git a/README.md b/README.md index a14de40..8a04a16 100644 --- a/README.md +++ b/README.md @@ -37,58 +37,58 @@ Media-Parser是一款专为短视频创作者与开发者打造的**100%原生 ## 💾 支持的平台矩阵 -| 平台名称 | 作者 | 标题 | 封面 | 视频 | 图集 | 音频 | 字幕 | 实况 | -| --- | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | -| **抖音** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | -| **小红书** | ✓ | ✓ | ✓ | ✓ | ✓ | | | ✓ | -| **视频号** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | | -| **微信公众号** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | | -| **快手** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | ✓ | -| **哔哩哔哩** | ✓ | ✓ | ✓ | ✓ | | ✓ | | | -| **豆包** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **即梦AI** | ✓ | ✓ | ✓ | ✓ | | | | | -| **小云雀AI** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **可灵AI** | ✓ | ✓ | ✓ | ✓ | | | | | -| **海螺AI** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **夸克AI** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **通义千问** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **腾讯元宝** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **闲鱼** | | ✓ | ✓ | | ✓ | | | | -| **拼多多** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **Soul** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **汽水音乐** | ✓ | ✓ | ✓ | ✓ | | ✓ | ✓ | | -| **QQ音乐** | ✓ | ✓ | ✓ | ✓ | | | | | -| **网易云音乐** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | -| **酷狗音乐** | ✓ | ✓ | ✓ | ✓ | | ✓ | | | -| **配音秀** | ✓ | ✓ | ✓ | ✓ | | | | | -| **松果时刻** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | | -| **腾讯频道** | ✓ | ✓ | ✓ | ✓ | | | | | -| **剪映 / CapCut** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | | -| **快影** | ✓ | ✓ | ✓ | ✓ | | ✓ | | | -| **皮皮搞笑** | ✓ | ✓ | ✓ | ✓ | | | | | -| **微视** | ✓ | ✓ | ✓ | ✓ | | | | | -| **AcFun** | ✓ | ✓ | ✓ | ✓ | | | | | -| **西瓜视频** | ✓ | ✓ | ✓ | ✓ | | | | | -| **今日头条** | ✓ | ✓ | ✓ | ✓ | | | | | -| **绿洲** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **皮皮虾** | ✓ | ✓ | ✓ | ✓ | | | | | -| **全民K歌** | | ✓ | ✓ | ✓ | | | | | -| **新片场** | ✓ | ✓ | ✓ | ✓ | | | | | -| **好看视频** | ✓ | ✓ | ✓ | ✓ | | | | | -| **梨视频** | ✓ | ✓ | ✓ | ✓ | | | | | -| **微博** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **知乎** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **虎牙** | | ✓ | ✓ | ✓ | | | | | -| **美拍** | ✓ | ✓ | ✓ | ✓ | | | | | -| **最右** | ✓ | ✓ | | ✓ | | | | | -| **番茄小说** | ✓ | ✓ | ✓ | ✓ | | | | | -| **红果短剧** | ✓ | ✓ | ✓ | ✓ | | | | | -| **红果漫剧** | ✓ | ✓ | ✓ | ✓ | | | | | -| **得物** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **网易LOFTER** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **星绘AI** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | -| **央视** | ✓ | ✓ | ✓ | ✓ | | | | | -| **央视频** | ✓ | ✓ | ✓ | | ✓ | | | | +| 平台名称 | 作者 | 标题 | 封面 | 视频 | 图集 | 多视频 | 实况 | 音频 | 字幕 | +| --- | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | +| **抖音** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | +| **小红书** | ✓ | ✓ | ✓ | ✓ | ✓ | | ✓ | | | +| **视频号** | ✓ | ✓ | ✓ | ✓ | ✓ | | | ✓ | | +| **微信公众号** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | ✓ | | +| **快手** | ✓ | ✓ | ✓ | ✓ | ✓ | | ✓ | ✓ | | +| **哔哩哔哩** | ✓ | ✓ | ✓ | ✓ | | ✓ | | ✓ | | +| **豆包** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | | | +| **即梦AI** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **小云雀AI** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **可灵AI** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **海螺AI** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **夸克AI** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **通义千问** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | | | +| **腾讯元宝** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | | | +| **闲鱼** | | ✓ | ✓ | | ✓ | | | | | +| **拼多多** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **Soul** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **汽水音乐** | ✓ | ✓ | ✓ | ✓ | | | | ✓ | ✓ | +| **QQ音乐** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **网易云音乐** | ✓ | ✓ | ✓ | ✓ | ✓ | | ✓ | ✓ | ✓ | +| **酷狗音乐** | ✓ | ✓ | ✓ | ✓ | | | | ✓ | | +| **配音秀** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **松果时刻** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | ✓ | | +| **腾讯频道** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **剪映 / CapCut** | ✓ | ✓ | ✓ | ✓ | ✓ | | | ✓ | | +| **快影** | ✓ | ✓ | ✓ | ✓ | | | | ✓ | | +| **皮皮搞笑** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **微视** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **AcFun** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **西瓜视频** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **今日头条** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **绿洲** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **皮皮虾** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **全民K歌** | | ✓ | ✓ | ✓ | | | | | | +| **新片场** | ✓ | ✓ | ✓ | ✓ | | ✓ | | | | +| **好看视频** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **梨视频** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **微博** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **知乎** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **虎牙** | | ✓ | ✓ | ✓ | | | | | | +| **美拍** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **最右** | ✓ | ✓ | | ✓ | | | | | | +| **番茄小说** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **红果短剧** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **红果漫剧** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **得物** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **网易LOFTER** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **星绘AI** | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | +| **央视** | ✓ | ✓ | ✓ | ✓ | | | | | | +| **央视频** | ✓ | ✓ | ✓ | | ✓ | | | | | *注:腾讯元宝等极少数平台提取的是官方存储桶原画质直链,素材保留官方原生水印。 diff --git a/docs/parsers/wechat-mp.md b/docs/parsers/wechat-mp.md index af2244d..9e9e3c3 100644 --- a/docs/parsers/wechat-mp.md +++ b/docs/parsers/wechat-mp.md @@ -10,13 +10,14 @@ * **支持媒体类型**: * **超高清正文插图图集**:自动升级为无损原始画质 (`/0?`),过滤内置 emoji 图标 * **文章封面图** (`cover_url`) - * **1080P 超清视频直链** (`video_url` / `video_list`):支持公众号视频动态与正文内嵌视频 + * **1080P 超清视频直链 / 多视频列表** (`video_url` / `video_list`):支持公众号视频动态与正文内嵌多视频(自动按视频 ID 分组并为每个视频提取最高画质流) * **内嵌语音播报 / 音频直链** (`audio_url`) * **文章标题与发布者信息**:包含文章/视频标题、公众号作者昵称、`gh_` 标识及官方头像 * **支持域名**: * `mp.weixin.qq.com`(公众号文章详情页、视频动态页与分享短链) * **常见链接形态**: * **文章长链/短链**:`https://mp.weixin.qq.com/s/LqZUts7aWOz7D7W_jIEXsA` + * **多视频文章长链**:`https://mp.weixin.qq.com/s/aNmLs3qJVhFGcgtGxAVlig` * **视频消息动态**:`https://mp.weixin.qq.com/s/GQiMB7CC6r7wil54I0TMwQ` * **历史参数长链**:`https://mp.weixin.qq.com/s?__biz=MzA5...&mid=...&idx=1&sn=...` * **Cookie 依赖**:🟢 免配置(开箱即用)。 @@ -43,10 +44,12 @@ full_res_url = re.sub(r'/(?:640|300|0)\?', '/0?', img_src) ``` -### 4. 视频流提取与清晰度择优算法 -公众号视频动态或正文内嵌视频在页面 JS 对象 `mp_video_trans_info` 中包含了多档转码格式(如 1080P 超清、720P 高清、480P 流畅等): -1. 扫描页面中的转码规格列表,提取所有 `mpvideo.qpic.cn` 视频流; -2. 依据 `video_quality_level`、分辨率(`width * height`)以及 `filesize` 进行多维度打分,**自动择优选择最高分辨率(如 1080×1920 超清)的 MP4 视频直链**。 +### 4. 多视频流提取与清晰度择优算法 +公众号视频动态或正文内嵌的多个视频在页面 JS 变量 `video_page_info` 或 `videoPageInfos` / `video_page_infos` 的 `mp_video_trans_info` 中包含了多档转码格式(如 1080P 超清、720P 高清、480P 流畅等): +1. 扫描页面中所有的转码规格列表,提取所有 `mpvideo.qpic.cn` 视频流; +2. **多视频按唯一文件指纹分组**:基于视频文件的唯一标识(`file_id`)将不同清晰度的分发流归类到各自的内嵌视频; +3. **独立打分择优**:对每个视频的各个清晰度依据 `video_quality_level`、分辨率(`width * height`)以及 `filesize` 进行多维度打分,**自动为每个视频挑选最高分辨率(如 1080×1920 超清)的 MP4 直链**; +4. **全量聚合返回**:将文章中包含的全部视频直链按正文顺序聚合至 `video_list`,同时 `video_url` 指向首个视频。 ### 5. 语音/音频资源提取 * 从正文 `` 标签提取语音文件 ID; diff --git a/src/parsers/wechat_mp_parser.py b/src/parsers/wechat_mp_parser.py index 4a80137..b127928 100644 --- a/src/parsers/wechat_mp_parser.py +++ b/src/parsers/wechat_mp_parser.py @@ -101,10 +101,10 @@ class WechatMpParser(BaseParser): if voice_tag and voice_tag.get("voice_encode_fileid"): audio_url = f"https://res.wx.qq.com/voice/getvoice?mediaid={voice_tag.get('voice_encode_fileid')}" - # 6. 视频提取 (mpvideo / 视频消息 / 嵌入式视频) + # 6. 视频提取 (mpvideo / 视频消息 / 正文内嵌多视频) video_url = None video_list = [] - video_items = [] + video_groups = {} for block in re.finditer( r"\{[^{}]*?url\s*:\s*['\"`](https?://mpvideo\.qpic\.cn/[^'\"`]+)['\"`][^{}]*?\}", html, @@ -122,31 +122,59 @@ class WechatMpParser(BaseParser): filesize = int(filesize_m.group(1)) if filesize_m else 0 quality = int(quality_m.group(1)) if quality_m else 0 - video_items.append({ + item = { "url": raw_vurl, "width": width, "height": height, "filesize": filesize, "quality": quality, - }) + } - if video_items: - best_video = max( - video_items, - key=lambda v: (v["quality"], v["width"] * v["height"], v["filesize"]), - ) - video_url = best_video["url"] - video_list = [best_video["url"]] + # 依据视频唯一标识分组(如 mpvideo.qpic.cn/{file_id}.f10002.mp4 中的 file_id) + m_f = re.search(r"mpvideo\.qpic\.cn/([a-zA-Z0-9_-]+)\.f\d+", raw_vurl) + if m_f: + group_key = m_f.group(1) + else: + m_prefix = re.search(r"mpvideo\.qpic\.cn/([a-zA-Z0-9_-]+)", raw_vurl) + group_key = m_prefix.group(1) if m_prefix else raw_vurl.split("?")[0] + + if group_key not in video_groups: + video_groups[group_key] = [] + video_groups[group_key].append(item) + + # 单视频文章(如 video_page_info)场景保护 + if "video_page_info" in html and "video_page_infos" not in html and "videoPageInfos" not in html: + all_items = [item for items in video_groups.values() for item in items] + if all_items: + best_video = max( + all_items, + key=lambda v: (v["quality"], v["width"] * v["height"], v["filesize"]), + ) + video_list = [best_video["url"]] + elif video_groups: + for items in video_groups.values(): + best_video = max( + items, + key=lambda v: (v["quality"], v["width"] * v["height"], v["filesize"]), + ) + if best_video["url"] not in video_list: + video_list.append(best_video["url"]) else: # 备选匹配任意 mpvideo.qpic.cn 链接 direct_matches = re.findall( r"https?://mpvideo\.qpic\.cn/[^\s'\"`]+", html, ) - if direct_matches: - clean_vurl = direct_matches[0].replace(r"\x26amp;", "&").replace("&", "&") - video_url = clean_vurl - video_list = [clean_vurl] + seen_keys = set() + for dm in direct_matches: + clean_vurl = dm.replace(r"\x26amp;", "&").replace("&", "&") + m_prefix = re.search(r"mpvideo\.qpic\.cn/([a-zA-Z0-9_-]+)", clean_vurl) + key = m_prefix.group(1) if m_prefix else clean_vurl.split("?")[0] + if key not in seen_keys: + seen_keys.add(key) + video_list.append(clean_vurl) + + video_url = video_list[0] if video_list else None return { "title": title, @@ -166,7 +194,7 @@ class WechatMpParser(BaseParser): return {} def get_real_video_url(self): - """提取视频直链(如文章内嵌视频源)。""" + """提取首个视频直链(如文章内嵌视频源)。""" return self.article_data.get("video_url") def get_title_content(self): @@ -195,5 +223,5 @@ class WechatMpParser(BaseParser): return self.article_data.get("image_list") or [] def get_video_list(self): - video_url = self.get_real_video_url() - return [video_url] if video_url else [] + """提取文章内嵌的所有视频直链列表。""" + return self.article_data.get("video_list") or [] diff --git a/tests/live_parser_samples.json b/tests/live_parser_samples.json index 6f8bb34..6eee3c0 100644 --- a/tests/live_parser_samples.json +++ b/tests/live_parser_samples.json @@ -4203,6 +4203,23 @@ ], "note": "2026-08 真实反馈验证: 《销量低的新能源,敢不敢买?》公众号视频" }, + { + "platform": "微信公众号", + "url": "https://mp.weixin.qq.com/s/aNmLs3qJVhFGcgtGxAVlig", + "pattern": "mp.weixin.qq.com 正文多视频嵌入 (s/aNmLs3qJVhFGcgtGxAVlig)", + "media_types": [ + "video", + "images" + ], + "expected_fields": [ + "video", + "images", + "cover", + "title", + "author" + ], + "note": "2026-09 真实反馈验证: 《这,才是下一个5年的工作操作系统》多视频内嵌文章" + }, { "platform": "红果短剧", "url": "https://novelquickapp.com/s/YlMu9l4vgE4/", diff --git a/tests/test_wechat_mp_parser.py b/tests/test_wechat_mp_parser.py index a9f7015..58b31ba 100644 --- a/tests/test_wechat_mp_parser.py +++ b/tests/test_wechat_mp_parser.py @@ -102,6 +102,81 @@ class WechatMpParserTest(unittest.TestCase): self.assertEqual(parser.get_real_video_url(), "http://mpvideo.qpic.cn/video_hd.mp4?k=2&p=1") self.assertEqual(parser.get_video_list(), ["http://mpvideo.qpic.cn/video_hd.mp4?k=2&p=1"]) + def test_maps_multiple_videos_response(self): + html_content = """ + + + + + + + + + + """ + response = Mock() + response.status_code = 200 + response.raise_for_status.return_value = None + response.text = html_content + + with patch("requests.Session.get", return_value=response): + parser = WechatMpParser(self.URL) + + self.assertEqual(parser.get_title_content(), "多视频文章测试") + self.assertEqual(parser.get_real_video_url(), "http://mpvideo.qpic.cn/0bc3vid1aaaa.f10002.mp4?k=2") + self.assertEqual( + parser.get_video_list(), + [ + "http://mpvideo.qpic.cn/0bc3vid1aaaa.f10002.mp4?k=2", + "http://mpvideo.qpic.cn/0bc3vid2bbbb.f10002.mp4?k=4", + ], + ) + def test_handles_request_failure_gracefully(self): response = Mock() response.raise_for_status.side_effect = Exception("HTTP 404") @@ -112,8 +187,10 @@ class WechatMpParserTest(unittest.TestCase): self.assertEqual(parser.get_title_content(), "微信公众号文章") self.assertEqual(parser.get_image_list(), []) self.assertIsNone(parser.get_real_video_url()) + self.assertEqual(parser.get_video_list(), []) if __name__ == "__main__": unittest.main() +