mirror of
https://github.com/NanmiCoder/MediaCrawler.git
synced 2026-09-25 17:48:12 +08:00
fix(xhs): 适配上游 EF* 分档,修复视频只下到封面
小红书把 DASH 分档名从按编码命名(h264/h265/av1)改成了按内部档位命名 (EF4/EF5/EF6/EF7),同时 video.consumer 里的 origin_video_key 也没了。 extract_video_urls 写死 stream["h264"],两条路都取不到地址,于是静默返回 空列表:笔记照常入库,但 video_url 为空、媒体只下到封面,全程不报错。 - media.py: 遍历 stream 下所有列表型分档,不再写死分档名;同一分档内含多个 分辨率(720P→4K),按 (height, avg_bitrate) 降序,最高档作主地址,其余作为 备用地址交给下载器回退;origin_video_key 分支保留 - 测试 fixture 换成真实响应结构,旧的 h264 结构另存一份确保兼容性不被改坏 验证:真实响应下旧实现取到 0 条、新实现 5 条;真实下载得到 4K 视频 (54079413 字节 / hevc 3840x2160 / 281.87s),旧结构与图文笔记行为不变。
This commit is contained in:
@@ -27,6 +27,8 @@ from media_platform.xhs import media as xhs_media
|
||||
|
||||
# --------------------------------------------------------------------------- xhs fixture
|
||||
|
||||
# 旧结构:分档按编码命名(h264),consumer 带无水印源片 key。
|
||||
# 保留这份 fixture 是为了确保兼容性不被改坏。
|
||||
XHS_VIDEO_NOTE = {
|
||||
"note_id": "video-note-1",
|
||||
"type": "video",
|
||||
@@ -39,6 +41,47 @@ XHS_VIDEO_NOTE = {
|
||||
},
|
||||
}
|
||||
|
||||
# 上游当前结构(2026-09 真实响应裁剪):分档改按内部档位命名(EF4/EF5/EF6/EF7),
|
||||
# 同一分档内含多个分辨率,consumer 里不再有 origin_video_key。
|
||||
# 回归背景:写死 stream["h264"] 时这里的视频地址会被静默取空,只下到封面。
|
||||
XHS_VIDEO_NOTE_NEW_SHAPE = {
|
||||
"note_id": "video-note-2",
|
||||
"type": "video",
|
||||
"image_list": [{"url_default": "https://sns-webpic.xhscdn.com/img-cover-2"}],
|
||||
"video": {
|
||||
"consumer": {"chapters": []},
|
||||
"media": {
|
||||
"stream": {
|
||||
"EF4": [
|
||||
{
|
||||
"master_url": "https://sns-video-v4.xhscdn.com/720p-ef4",
|
||||
"backup_urls": ["https://sns-bak-v1.xhscdn.com/720p-ef4"],
|
||||
"width": 1280,
|
||||
"height": 720,
|
||||
"avg_bitrate": 414724,
|
||||
}
|
||||
],
|
||||
"EF5": [
|
||||
{
|
||||
"master_url": "https://sns-video-v4.xhscdn.com/1080p-ef5",
|
||||
"width": 1920,
|
||||
"height": 1080,
|
||||
"avg_bitrate": 635797,
|
||||
},
|
||||
{
|
||||
"master_url": "https://sns-video-v4.xhscdn.com/4k-ef5",
|
||||
"width": 3840,
|
||||
"height": 2160,
|
||||
"avg_bitrate": 1534891,
|
||||
},
|
||||
],
|
||||
"EF6": [],
|
||||
"EF7": [],
|
||||
}
|
||||
},
|
||||
},
|
||||
}
|
||||
|
||||
XHS_IMAGE_NOTE = {
|
||||
"note_id": "image-note-1",
|
||||
"type": "normal",
|
||||
@@ -116,6 +159,44 @@ def test_xhs_video_note_without_origin_key_uses_master_url():
|
||||
assert video.backup_urls == ()
|
||||
|
||||
|
||||
def test_xhs_video_note_new_stream_buckets():
|
||||
"""上游把分档名从 h264 换成 EF4/EF5/... 之后仍要能取到地址(本次回归核心)"""
|
||||
urls = xhs_media.extract_video_urls(XHS_VIDEO_NOTE_NEW_SHAPE)
|
||||
|
||||
# 分辨率高的排前面作主地址,其余按清晰度降序作为备用
|
||||
assert urls == [
|
||||
"https://sns-video-v4.xhscdn.com/4k-ef5",
|
||||
"https://sns-video-v4.xhscdn.com/1080p-ef5",
|
||||
"https://sns-video-v4.xhscdn.com/720p-ef4",
|
||||
]
|
||||
|
||||
|
||||
def test_xhs_video_note_new_shape_builds_video_item():
|
||||
"""新结构下必须产出 video 任务,而不是只剩封面"""
|
||||
items = xhs_media.build_media_items(XHS_VIDEO_NOTE_NEW_SHAPE)
|
||||
|
||||
assert [item.stem for item in items] == ["cover", "video"]
|
||||
video = items[1]
|
||||
assert video.media_type == MediaType.VIDEO
|
||||
assert video.url == "https://sns-video-v4.xhscdn.com/4k-ef5"
|
||||
assert video.backup_urls == (
|
||||
"https://sns-video-v4.xhscdn.com/1080p-ef5",
|
||||
"https://sns-video-v4.xhscdn.com/720p-ef4",
|
||||
)
|
||||
|
||||
|
||||
def test_xhs_video_note_missing_dimensions_keeps_all_urls():
|
||||
"""缺 height/avg_bitrate 时不能抛异常,也不能丢地址"""
|
||||
note = {
|
||||
"type": "video",
|
||||
"video": {
|
||||
"media": {"stream": {"EF5": [{"master_url": "https://a"}, {"master_url": "https://b"}]}}
|
||||
},
|
||||
}
|
||||
|
||||
assert sorted(xhs_media.extract_video_urls(note)) == ["https://a", "https://b"]
|
||||
|
||||
|
||||
def test_xhs_international_does_not_build_xhscdn_url(monkeypatch):
|
||||
monkeypatch.setattr(config, "XHS_INTERNATIONAL", True)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user