From a77a914f0de51456f63f14348fef24f5f307db93 Mon Sep 17 00:00:00 2001 From: ucmao <294323976@qq.com> Date: Tue, 8 Sep 2026 09:50:44 +0800 Subject: [PATCH] =?UTF-8?q?=E6=89=A9=E5=B1=95QQ=E9=9F=B3=E4=B9=90=E6=AD=8C?= =?UTF-8?q?=E6=9B=B2=E9=9F=B3=E9=A2=91=E4=B8=8E=E6=AD=8C=E8=AF=8D=E8=A7=A3?= =?UTF-8?q?=E6=9E=90=E6=94=AF=E6=8C=81?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- README.md | 2 +- configs/business_config.json | 1 + docs/index.md | 2 +- docs/parsers/qqmusic.md | 44 ++++--- src/parsers/qqmusic_parser.py | 224 +++++++++++++++++++++++++++++---- templates/admin/dashboard.html | 10 +- tests/test_qqmusic_parser.py | 55 ++++++++ utils/web_fetcher.py | 10 +- 8 files changed, 287 insertions(+), 61 deletions(-) diff --git a/README.md b/README.md index ab2fcfd..3fd6477 100644 --- a/README.md +++ b/README.md @@ -57,7 +57,7 @@ Media-Parser是一款专为短视频创作者与开发者打造的**100%原生 | **拼多多** | ✓ | ✓ | | ✓ | ✓ | ✓ | | | | | | **Soul** | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | | | | | **汽水音乐** | ✓ | ✓ | | ✓ | ✓ | | | | ✓ | ✓ | -| **QQ音乐** | ✓ | ✓ | | ✓ | ✓ | | | | | | +| **QQ音乐** | ✓ | ✓ | | ✓ | ✓ | | ✓ | | ✓ | ✓ | | **网易云音乐** | ✓ | ✓ | | ✓ | ✓ | ✓ | | ✓ | ✓ | ✓ | | **酷狗音乐** | ✓ | ✓ | | ✓ | ✓ | | | | ✓ | | | **配音秀** | ✓ | ✓ | | ✓ | ✓ | | | | | | diff --git a/configs/business_config.json b/configs/business_config.json index a25821e..9fc7f8d 100755 --- a/configs/business_config.json +++ b/configs/business_config.json @@ -86,6 +86,7 @@ "qishui.douyin.com": "汽水音乐", "music.douyin.com": "汽水音乐", "c6.y.qq.com": "QQ音乐", + "i.y.qq.com": "QQ音乐", "i2.y.qq.com": "QQ音乐", "y.qq.com": "QQ音乐", "www.y.qq.com": "QQ音乐", diff --git a/docs/index.md b/docs/index.md index dc41b9e..aa95141 100644 --- a/docs/index.md +++ b/docs/index.md @@ -69,7 +69,7 @@ | 38 | **番茄小说** | 短剧视频 | ✅ 支持 | 🟢 免配置 | H5 推广页 HTML 提取 | [查看指南](parsers/fanqie.md) | | 39 | **红果短剧** | 短剧视频 | ✅ 支持 | 🟢 免配置 | H5 推广页 HTML 提取 | [查看指南](parsers/fanqie.md) | | 40 | **红果漫剧** | 漫剧/短剧视频 | ✅ 支持 | 🟢 免配置 | H5 推广页 HTML 提取 | [查看指南](parsers/fanqie.md) | -| 41 | **QQ音乐** | MV / 分享视频 / 多档 MP4 | ✅ 支持 | 🟢 免配置 | SSR 元数据 + MusicU 播放接口 | [查看指南](parsers/qqmusic.md) | +| 41 | **QQ音乐** | 歌曲音频 / LRC 歌词 / MV / 多档 MP4 | ✅ 支持 | 🟢 免配置 | PC Web 协议 + MusicU 播放与歌词接口 | [查看指南](parsers/qqmusic.md) | | 42 | **网易云音乐** | MV / Mlog / 歌曲 / 动态图集与 LivePhoto | ✅ 支持 | 🟢 公开内容免配置 | SSR / Event 数据 + 公开播放接口 | [查看指南](parsers/netease-music.md) | | 43 | **酷狗音乐** | MV / 免费歌曲 / 多档 MP4 | ✅ 支持 | 🟢 公开内容免配置 | H5 公开接口签名 + 分享页数据 | [查看指南](parsers/kugou-music.md) | | 44 | **配音秀** | 配音作品视频 (MP4) | ✅ 支持 | 🟢 公开内容免配置 | 作品页初始化数据提取 | [查看指南](parsers/peiyinxiu.md) | diff --git a/docs/parsers/qqmusic.md b/docs/parsers/qqmusic.md index 4c91362..88e7644 100644 --- a/docs/parsers/qqmusic.md +++ b/docs/parsers/qqmusic.md @@ -3,37 +3,35 @@ ## 支持范围 - `c6.y.qq.com/base/fcgi-bin/u` 分享短链 +- `i.y.qq.com/v8/playsong.html?songid=...` / `songmid=...` 移动端歌曲播放页 - `i2.y.qq.com/n3/other/pages/details/mv.html?vid=...` MV / 分享视频页 - `y.qq.com/n/ryqq/mv/{vid}` 桌面端 MV 地址 -- 标题、封面、上传者或歌手信息 -- 多档公开 MP4 地址,默认把最高可用清晰度作为 `video_url` - -当前接入针对反馈库中实际出现的 QQ 音乐 MV / 分享视频。普通歌曲的完整音频受会员、版权区域和登录态控制,暂不把试听或会员流伪装成可下载音频。 +- `y.qq.com/n/ryqq/songDetail/{songmid}` 桌面端歌曲详情页 +- 歌曲音频直链(`audio_url`)与带时间戳的 LRC 歌词(`subtitles`) +- 标题、800x800 高清封面、歌手头像及昵称信息 +- MV 多档公开 MP4 地址,默认把最高可用清晰度作为 `video_url` ## 解析流程 -1. 跟随 `c6.y.qq.com` 的 302 跳转,保留落地页的 `vid`。 -2. 读取分享页 `window.__ssrFirstPageData__`,补充标题、封面与上传者。 -3. 调用 QQ 音乐 `musicu.fcg` 的 MV 信息和播放地址模块。 -4. 过滤接口中不可播放的档位,按 `filetype` 从高到低排序并去重。 -5. API 返回时统一将 `http://` 媒体地址升级为 HTTPS。 +1. 识别并提取分享链接中的媒体标识(`songid`、`songmid` 或 MV `vid`)。 +2. 若为短链则跟随重定向并读取页面 `window.__ssrFirstPageData__` 补充元数据。 +3. 调用 QQ 音乐 `musicu.fcg`(PC Web 协议): + - 歌曲:请求 `songinfo`、`playUrl`(`CgiGetVkey`)和 `lyric`(`PlayLyricInfo`),直接解析音频流、高清封面与 LRC 歌词。 + - MV:请求 `video.VideoDataServer` 与 `gosrf.Stream.MvUrlProxy` 获取多档 MP4。 +4. API 返回时统一将 `http://` 媒体地址升级为 HTTPS。 ## 返回字段 -| 字段 | 来源 | -| --- | --- | -| `video_id` | 分享页查询参数 `vid` 或桌面端 MV 路径 | -| `title` | SSR / MV 信息接口中的 `name` | -| `cover_url` | `cover_pic` 或 `first_frame_pic` | -| `author` | 分享上传者,缺失时回退到歌手 | -| `video_url` | 最高可用 MP4 档位 | -| `video_list` | 两档及以上时返回全部可用清晰度 | - -## 限制 - -- 私密、下架、付费或地区受限 MV 可能不返回播放地址。 -- 播放 URL 带有效期签名,应在解析后及时使用,不建议长期缓存。 -- 普通歌曲音频不在当前支持范围内。 +| 字段 | 适用类型 | 来源 | +| --- | --- | --- | +| `video_id` | 歌曲 / MV | 查询参数 `songid`/`songmid`/`vid` 或页面路径 | +| `title` | 歌曲 / MV | SSR 或接口中的 `name`/`title` | +| `cover_url` | 歌曲 / MV | 800x800 专辑封面图或 MV `cover_pic` | +| `author` | 歌曲 / MV | 歌手名与头像,或 MV 上传者 | +| `audio_url` | 歌曲 | `CgiGetVkey` 返回的音频流地址(`sip + purl`) | +| `subtitles` | 歌曲 | `PlayLyricInfo` 解码并解析出的带时间戳 LRC 歌词列表 | +| `video_url` | MV | 最高可用 MP4 档位 | +| `video_list` | MV | 两档及以上时返回全部可用清晰度 | ## 验证 diff --git a/src/parsers/qqmusic_parser.py b/src/parsers/qqmusic_parser.py index 57e48a2..3328f7c 100644 --- a/src/parsers/qqmusic_parser.py +++ b/src/parsers/qqmusic_parser.py @@ -1,5 +1,6 @@ -"""QQ 音乐 MV 与分享视频解析器。""" +"""QQ 音乐歌曲与 MV 分享解析器。""" +import base64 import json import re from urllib.parse import parse_qs, urlparse @@ -14,7 +15,7 @@ logger = get_logger(__name__) @register_parser("QQ音乐") class QQMusicParser(BaseParser): - """解析 QQ 音乐公开 MV 页面,并返回所有可用 MP4 清晰度。""" + """解析 QQ 音乐公开歌曲与 MV 页面,返回音频流、歌词或多清晰度视频。""" API_URL = "https://u.y.qq.com/cgi-bin/musicu.fcg" MOBILE_UA = ( @@ -22,6 +23,10 @@ class QQMusicParser(BaseParser): "AppleWebKit/605.1.15 (KHTML, like Gecko) Version/18.0 " "Mobile/15E148 Safari/604.1" ) + DESKTOP_UA = ( + "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) " + "AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36" + ) def __init__(self, real_url): super().__init__(real_url) @@ -29,26 +34,48 @@ class QQMusicParser(BaseParser): "User-Agent": self.MOBILE_UA, "Referer": "https://y.qq.com/", } - self.vid = self._extract_vid(real_url) self.title = "" self.cover_url = None self.author = {"nickname": "", "author_id": "", "avatar": ""} self.video_list = [] + self.audio_url = None + self.subtitles = None + + self.media_type, self.vid, self.songmid, self.songid = self._detect_media(real_url) self._parse_page() - self._fetch_mv_data() + if self.media_type == "song" or self.songmid or self.songid: + self._fetch_song_data() + elif self.media_type == "mv" or self.vid: + self._fetch_mv_data() @staticmethod - def _extract_vid(url): + def _detect_media(url): parsed = urlparse(url or "") - if vid := (parse_qs(parsed.query).get("vid") or [None])[0]: - return vid - match = re.search(r"/(?:mv|mvDetail)/([A-Za-z0-9]+)(?:/|$)", parsed.path) - return match.group(1) if match else None + params = parse_qs(parsed.query) + path = parsed.path + + # 检查 MV + if vid := (params.get("vid") or [None])[0]: + return "mv", vid, None, None + if match := re.search(r"/(?:mv|mvDetail)/([A-Za-z0-9]+)(?:/|$)", path): + return "mv", match.group(1), None, None + + # 检查歌曲 + songmid = (params.get("songmid") or params.get("songMid") or [None])[0] + songid = (params.get("songid") or params.get("songId") or params.get("id") or [None])[0] + if songmid or songid: + return "song", None, songmid, songid + + if match := re.search(r"/(?:songDetail|song)/([A-Za-z0-9]+)(?:/|$)", path): + return "song", None, match.group(1), None + + if "playsong" in path: + return "song", None, songmid, songid + + return None, None, None, None def _parse_page(self): - """从 SSR 数据补充上传者信息;接口仍是媒体地址的权威来源。""" - if not self.vid: - return + """抓取页面 HTML 并尝试解析 SSR 数据,用于补充元数据或识别短链。""" try: response = self.session.get( self.real_url, @@ -58,11 +85,15 @@ class QQMusicParser(BaseParser): ) response.raise_for_status() self.html_content = response.text - self.vid = self.vid or self._extract_vid(response.url) + final_url = response.url or self.real_url except Exception as exc: logger.warning("Failed to fetch QQMusic page: %s", exc) return + # 如果初始未检测到媒体类型,从重定向后的 URL 再次检测 + if not self.media_type and not self.vid and not self.songmid and not self.songid: + self.media_type, self.vid, self.songmid, self.songid = self._detect_media(final_url) + match = re.search( r"window\.__ssrFirstPageData__\s*=\s*(\"(?:\\.|[^\"\\])*\")", self.html_content or "", @@ -70,24 +101,158 @@ class QQMusicParser(BaseParser): ) if not match: return + try: encoded_payload = json.loads(match.group(1)) payload = json.loads(encoded_payload) - data = payload.get("data") or {} - video = data.get("video") or {} - self._apply_metadata(video) + data = payload.get("data") or payload - creators = data.get("creator") or [] - if creators and isinstance(creators[0], dict): - creator = creators[0] - self.author = { - "nickname": creator.get("nick") or creator.get("name") or "", - "author_id": str(creator.get("mid") or creator.get("uin") or ""), - "avatar": creator.get("avatar") or creator.get("headurl") or "", - } + # MV 页面 SSR 结构 + video = data.get("video") + if video and isinstance(video, dict): + self.media_type = self.media_type or "mv" + self._apply_mv_metadata(video) + creators = data.get("creator") or [] + if creators and isinstance(creators[0], dict): + creator = creators[0] + self.author = { + "nickname": creator.get("nick") or creator.get("name") or "", + "author_id": str(creator.get("mid") or creator.get("uin") or ""), + "avatar": creator.get("avatar") or creator.get("headurl") or "", + } + + # 歌曲页面 SSR 结构 + song = data.get("song") + if song and isinstance(song, dict): + self.media_type = self.media_type or "song" + self.songmid = self.songmid or song.get("mid") + self.songid = self.songid or (str(song.get("id")) if song.get("id") else None) + self._apply_song_metadata(song) except (json.JSONDecodeError, TypeError) as exc: logger.warning("Failed to parse QQMusic SSR payload: %s", exc) + def _fetch_song_data(self): + """调用 QQ 音乐 PC Web 协议接口获取歌曲信息、音频播放地址及歌词。""" + numeric_id = 0 + if self.songid: + try: + numeric_id = int(self.songid) + except (ValueError, TypeError): + numeric_id = 0 + + payload = { + "comm": { + "cv": 4747474, + "ct": 24, + "format": "json", + "inCharset": "utf-8", + "outCharset": "utf-8", + "notice": 0, + "platform": "yqq.json", + "needNewCode": 1, + "uin": 0, + }, + "songinfo": { + "module": "music.pf_song_detail_svr", + "method": "get_song_detail_yqq", + "param": { + "song_mid": self.songmid or "", + "song_id": numeric_id, + }, + }, + "playUrl": { + "module": "vkey.GetVkeyServer", + "method": "CgiGetVkey", + "param": { + "guid": "2333333333", + "songmid": [self.songmid] if self.songmid else [], + "songtype": [0], + "uin": "0", + "loginflag": 1, + "platform": "20", + }, + }, + "lyric": { + "module": "music.musichallSong.PlayLyricInfo", + "method": "GetPlayLyricInfo", + "param": { + "songMID": self.songmid or "", + "songID": numeric_id, + }, + }, + } + + try: + response = self.session.post( + self.API_URL, + json=payload, + headers={ + "User-Agent": self.DESKTOP_UA, + "Referer": "https://y.qq.com/", + "Content-Type": "application/json", + }, + timeout=10, + ) + response.raise_for_status() + result = response.json() + except Exception as exc: + logger.warning("Failed to fetch QQMusic song data: %s", exc) + return + + # 1. 歌曲元数据 + track_info = ((result.get("songinfo") or {}).get("data") or {}).get("track_info") or {} + if track_info: + self._apply_song_metadata(track_info) + + # 2. 音频直链 + play_data = (result.get("playUrl") or {}).get("data") or {} + sips = play_data.get("sip") or [] + midurlinfo = play_data.get("midurlinfo") or [] + if midurlinfo and isinstance(midurlinfo[0], dict): + purl = midurlinfo[0].get("purl") + if purl and sips: + sip = next((s for s in sips if s.startswith("http")), sips[0]) + self.audio_url = f"{sip}{purl}" + + # 3. 歌词解析 + lyric_b64 = ((result.get("lyric") or {}).get("data") or {}).get("lyric") + if lyric_b64 and isinstance(lyric_b64, str): + try: + decoded = base64.b64decode(lyric_b64).decode("utf-8", errors="ignore") + self.subtitles = self._parse_lrc(decoded) + except Exception as exc: + logger.debug("Failed to decode QQMusic lyrics: %s", exc) + + def _apply_song_metadata(self, track): + if not isinstance(track, dict): + return + self.title = self.title or track.get("name") or track.get("title") or "" + album = track.get("album") or {} + album_mid = album.get("mid") or (album.get("pmid", "").rstrip("_1") if isinstance(album.get("pmid"), str) else None) + if album_mid and not self.cover_url: + self.cover_url = f"https://y.qq.com/music/photo_new/T002R800x800M000{album_mid}.jpg" + + singers = track.get("singer") or track.get("singers") or [] + if singers and isinstance(singers[0], dict) and not self.author["nickname"]: + singer = singers[0] + singer_mid = singer.get("mid") or (singer.get("pmid", "").rstrip("_1") if isinstance(singer.get("pmid"), str) else "") + avatar = f"https://y.qq.com/music/photo_new/T001R300x300M000{singer_mid}.jpg" if singer_mid else "" + self.author = { + "nickname": singer.get("name") or singer.get("title") or "", + "author_id": str(singer.get("mid") or singer.get("id") or ""), + "avatar": avatar, + } + + @staticmethod + def _parse_lrc(lrc_text): + subtitles = [] + for line in lrc_text.splitlines(): + match = re.match(r"\[(\d+):(\d+(?:\.\d+)?)\](.*)", line.strip()) + if match and match.group(3).strip(): + start = int(match.group(1)) * 60 + float(match.group(2)) + subtitles.append({"start": round(start, 3), "text": match.group(3).strip()}) + return subtitles or None + def _fetch_mv_data(self): if not self.vid: return @@ -138,11 +303,11 @@ class QQMusicParser(BaseParser): return metadata = (((result.get("mvInfo") or {}).get("data") or {}).get(self.vid) or {}) - self._apply_metadata(metadata) + self._apply_mv_metadata(metadata) stream_data = (((result.get("mvUrl") or {}).get("data") or {}).get(self.vid) or {}) self.video_list = self._extract_streams(stream_data) - def _apply_metadata(self, video): + def _apply_mv_metadata(self, video): if not isinstance(video, dict): return self.title = self.title or video.get("name") or video.get("raw_name") or "" @@ -193,6 +358,12 @@ class QQMusicParser(BaseParser): def get_video_list(self): return self.video_list + def get_audio_url(self): + return self.audio_url + + def get_subtitles(self): + return self.subtitles + def get_title_content(self): return self.title @@ -201,3 +372,4 @@ class QQMusicParser(BaseParser): def get_author_info(self): return self.author + diff --git a/templates/admin/dashboard.html b/templates/admin/dashboard.html index ec62048..1996b60 100644 --- a/templates/admin/dashboard.html +++ b/templates/admin/dashboard.html @@ -689,15 +689,7 @@