From 995fc27931c2e2cd417af69d3e4321fd6c8e0d0e Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Thu, 10 Oct 2024 20:01:20 +1300
Subject: [PATCH 01/11] support storyboards
---
yt_dlp/extractor/bilibili.py | 70 ++++++++++++++++++++++++++++++++----
1 file changed, 64 insertions(+), 6 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index 62f68fbc6d..80752ea2fb 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -33,6 +33,7 @@ from ..utils import (
parse_qs,
parse_resolution,
qualities,
+ sanitize_url,
smuggle_url,
srt_subtitles_timecode,
str_or_none,
@@ -66,7 +67,61 @@ class BilibiliBaseIE(InfoExtractor):
f'Format(s) {missing_formats} are missing; you have to login or '
f'become a premium member to download them. {self._login_hint()}')
- def extract_formats(self, play_info):
+ def _extract_storyboard(self, duration, aid=None, bvid=None, cid=None):
+ video_id = aid or bvid
+ query = filter_dict({
+ 'aid': aid,
+ 'bvid': bvid,
+ 'cid': cid,
+ })
+ if not aid and not bvid:
+ return {}
+ idx = 1
+ while storyboard_info := traverse_obj(self._download_json(
+ 'https://api.bilibili.com/x/player/videoshot', video_id,
+ 'Downloading storyboard info', query={
+ 'index': idx,
+ **query,
+ }), 'data'):
+ if storyboard_info.get('image') and storyboard_info.get('index'):
+ rows, cols = storyboard_info.get('img_x_len'), storyboard_info.get('img_y_len')
+ fragments = []
+ last_duration = 0.0
+ for i, url in enumerate(storyboard_info['image'], start=1):
+ duration_index = i * rows * cols - 1
+ if duration_index < len(storyboard_info['index']) - 1:
+ current_duration = traverse_obj(storyboard_info, ('index', duration_index))
+ else:
+ current_duration = duration
+ if current_duration > last_duration and current_duration <= duration:
+ fragments.append({
+ 'url': sanitize_url(url),
+ 'duration': current_duration - last_duration,
+ })
+ last_duration = current_duration
+ else:
+ break
+ if fragments:
+ yield {
+ 'format_id': f'sb{idx}',
+ 'format_note': 'storyboard',
+ 'ext': 'mhtml',
+ 'protocol': 'mhtml',
+ 'acodec': 'none',
+ 'vcodec': 'none',
+ 'url': 'about:invalid',
+ 'width': storyboard_info.get('img_x_size'),
+ 'height': storyboard_info.get('img_y_size'),
+ 'fps': len(storyboard_info['image']) * rows * cols / duration,
+ 'rows': rows,
+ 'columns': cols,
+ 'fragments': fragments,
+ }
+ else:
+ return
+ idx += 1
+
+ def extract_formats(self, play_info, aid=None, bvid=None, cid=None):
format_names = {
r['quality']: traverse_obj(r, 'new_description', 'display_desc')
for r in traverse_obj(play_info, ('support_formats', lambda _, v: v['quality']))
@@ -128,6 +183,8 @@ class BilibiliBaseIE(InfoExtractor):
}),
**parse_resolution(format_names.get(play_info.get('quality'))),
})
+ formats.extend(self._extract_storyboard(
+ float_or_none(play_info.get('timelength'), scale=1000), aid=aid, bvid=bvid, cid=cid))
return formats
def _get_wbi_key(self, video_id):
@@ -291,7 +348,7 @@ class BilibiliBaseIE(InfoExtractor):
**metainfo,
'id': f'{video_id}_{cid}',
'title': f'{metainfo.get("title")} - {next(iter(edges.values())).get("title")}',
- 'formats': self.extract_formats(play_info),
+ 'formats': self.extract_formats(play_info, bvid=video_id, cid=cid),
'description': f'{json.dumps(edges, ensure_ascii=False)}\n{metainfo.get("description", "")}',
'duration': float_or_none(play_info.get('timelength'), scale=1000),
'subtitles': self.extract_subtitles(video_id, cid),
@@ -728,14 +785,14 @@ class BiliBiliIE(BilibiliBaseIE):
duration=traverse_obj(initial_state, ('videoData', 'duration', {int_or_none})),
__post_extractor=self.extract_comments(aid))
else:
- formats = self.extract_formats(play_info)
+ formats = self.extract_formats(play_info, bvid=video_id, cid=cid)
if not traverse_obj(play_info, ('dash')):
# we only have legacy formats and need additional work
has_qn = lambda x: x in traverse_obj(formats, (..., 'quality'))
for qn in traverse_obj(play_info, ('accept_quality', lambda _, v: not has_qn(v), {int})):
formats.extend(traverse_obj(
- self.extract_formats(self._download_playinfo(video_id, cid, headers=headers, qn=qn)),
+ self.extract_formats(self._download_playinfo(video_id, cid, headers=headers, qn=qn), bvid=video_id, cid=cid),
lambda _, v: not has_qn(v['quality'])))
self._check_missing_formats(play_info, formats)
flv_formats = traverse_obj(formats, lambda _, v: v['fragments'])
@@ -865,9 +922,10 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'Extracting episode', query={'fnval': '4048', 'ep_id': episode_id},
headers=headers)
premium_only = play_info.get('code') == -10403
+ bvid, cid = traverse_obj(play_info, ('result', 'play_view_business_info', 'episode_info', ('bvid', 'cid')))
play_info = traverse_obj(play_info, ('result', 'video_info', {dict})) or {}
- formats = self.extract_formats(play_info)
+ formats = self.extract_formats(play_info, bvid=bvid, cid=cid)
if not formats and (premium_only or '成为大会员抢先看' in webpage or '开通大会员观看' in webpage):
self.raise_login_required('This video is for premium members only')
@@ -1040,7 +1098,7 @@ class BilibiliCheeseBaseIE(BilibiliBaseIE):
return {
'id': str_or_none(ep_id),
'episode_id': str_or_none(ep_id),
- 'formats': self.extract_formats(play_info),
+ 'formats': self.extract_formats(play_info, aid=aid, cid=cid),
'extractor_key': BilibiliCheeseIE.ie_key(),
'extractor': BilibiliCheeseIE.IE_NAME,
'webpage_url': f'https://www.bilibili.com/cheese/play/ep{ep_id}',
From 47254db76dc76ba8afca1c5042d258cb022f14ba Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Thu, 10 Oct 2024 21:14:16 +1300
Subject: [PATCH 02/11] misc
---
yt_dlp/extractor/bilibili.py | 93 ++++++++++++++++--------------------
1 file changed, 42 insertions(+), 51 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index 80752ea2fb..bd3d87d931 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -68,58 +68,49 @@ class BilibiliBaseIE(InfoExtractor):
f'become a premium member to download them. {self._login_hint()}')
def _extract_storyboard(self, duration, aid=None, bvid=None, cid=None):
- video_id = aid or bvid
- query = filter_dict({
- 'aid': aid,
- 'bvid': bvid,
- 'cid': cid,
- })
- if not aid and not bvid:
+ if not (video_id := aid or bvid):
return {}
- idx = 1
- while storyboard_info := traverse_obj(self._download_json(
+ if storyboard_info := traverse_obj(self._download_json(
'https://api.bilibili.com/x/player/videoshot', video_id,
- 'Downloading storyboard info', query={
- 'index': idx,
- **query,
- }), 'data'):
- if storyboard_info.get('image') and storyboard_info.get('index'):
- rows, cols = storyboard_info.get('img_x_len'), storyboard_info.get('img_y_len')
- fragments = []
- last_duration = 0.0
- for i, url in enumerate(storyboard_info['image'], start=1):
- duration_index = i * rows * cols - 1
- if duration_index < len(storyboard_info['index']) - 1:
- current_duration = traverse_obj(storyboard_info, ('index', duration_index))
- else:
- current_duration = duration
- if current_duration > last_duration and current_duration <= duration:
- fragments.append({
- 'url': sanitize_url(url),
- 'duration': current_duration - last_duration,
- })
- last_duration = current_duration
- else:
- break
- if fragments:
- yield {
- 'format_id': f'sb{idx}',
- 'format_note': 'storyboard',
- 'ext': 'mhtml',
- 'protocol': 'mhtml',
- 'acodec': 'none',
- 'vcodec': 'none',
- 'url': 'about:invalid',
- 'width': storyboard_info.get('img_x_size'),
- 'height': storyboard_info.get('img_y_size'),
- 'fps': len(storyboard_info['image']) * rows * cols / duration,
- 'rows': rows,
- 'columns': cols,
- 'fragments': fragments,
- }
- else:
- return
- idx += 1
+ note='Downloading storyboard info', errnote='Failed to download storyboard info',
+ query=filter_dict({
+ 'index': 1,
+ 'aid': aid,
+ 'bvid': bvid,
+ 'cid': cid,
+ })), ('data', {lambda v: v if v['image'] and v['index'] else None})):
+ rows, cols = traverse_obj(storyboard_info, (('img_x_len', 'img_y_len'),))
+ fragments = []
+ last_duration = 0.0
+ for i, url in enumerate(storyboard_info['image'], start=1):
+ duration_index = i * rows * cols - 1
+ if duration_index < len(storyboard_info['index']) - 1:
+ current_duration = traverse_obj(storyboard_info, ('index', duration_index))
+ else:
+ current_duration = duration
+ if not current_duration or current_duration <= last_duration or current_duration > duration:
+ break
+ fragments.append({
+ 'url': sanitize_url(url),
+ 'duration': current_duration - last_duration,
+ })
+ if fragments:
+ return {
+ 'format_id': 'sb',
+ 'format_note': 'storyboard',
+ 'ext': 'mhtml',
+ 'protocol': 'mhtml',
+ 'acodec': 'none',
+ 'vcodec': 'none',
+ 'url': 'about:invalid',
+ 'width': storyboard_info.get('img_x_size'),
+ 'height': storyboard_info.get('img_y_size'),
+ 'fps': len(storyboard_info['image']) * rows * cols / duration if rows and cols else None,
+ 'rows': rows,
+ 'columns': cols,
+ 'fragments': fragments,
+ }
+ return {}
def extract_formats(self, play_info, aid=None, bvid=None, cid=None):
format_names = {
@@ -183,7 +174,7 @@ class BilibiliBaseIE(InfoExtractor):
}),
**parse_resolution(format_names.get(play_info.get('quality'))),
})
- formats.extend(self._extract_storyboard(
+ formats.append(self._extract_storyboard(
float_or_none(play_info.get('timelength'), scale=1000), aid=aid, bvid=bvid, cid=cid))
return formats
From 22e5e37c82f32e8c99596aad43ce27718c26e6b1 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Fri, 11 Oct 2024 01:30:07 +1300
Subject: [PATCH 03/11] extract episode_info in advance
---
yt_dlp/extractor/bilibili.py | 13 +++++++------
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index bd3d87d931..e7cfac55a0 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -68,7 +68,7 @@ class BilibiliBaseIE(InfoExtractor):
f'become a premium member to download them. {self._login_hint()}')
def _extract_storyboard(self, duration, aid=None, bvid=None, cid=None):
- if not (video_id := aid or bvid):
+ if not (video_id := aid or bvid) or not duration:
return {}
if storyboard_info := traverse_obj(self._download_json(
'https://api.bilibili.com/x/player/videoshot', video_id,
@@ -913,10 +913,11 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'Extracting episode', query={'fnval': '4048', 'ep_id': episode_id},
headers=headers)
premium_only = play_info.get('code') == -10403
- bvid, cid = traverse_obj(play_info, ('result', 'play_view_business_info', 'episode_info', ('bvid', 'cid')))
+ episode_info = traverse_obj(play_info, ('result', 'play_view_business_info', 'episode_info'))
+ aid, cid = episode_info.get('aid'), episode_info.get('cid')
play_info = traverse_obj(play_info, ('result', 'video_info', {dict})) or {}
- formats = self.extract_formats(play_info, bvid=bvid, cid=cid)
+ formats = self.extract_formats(play_info, aid=aid, cid=cid)
if not formats and (premium_only or '成为大会员抢先看' in webpage or '开通大会员观看' in webpage):
self.raise_login_required('This video is for premium members only')
@@ -927,7 +928,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
episode_number, episode_info = next((
(idx, ep) for idx, ep in enumerate(traverse_obj(
bangumi_info, (('episodes', ('section', ..., 'episodes')), ..., {dict})), 1)
- if str_or_none(ep.get('id')) == episode_id), (1, {}))
+ if str_or_none(ep.get('id')) == episode_id), (1, episode_info))
season_id = bangumi_info.get('season_id')
season_number, season_title = season_id and next((
@@ -936,7 +937,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
if e.get('season_id') == season_id
), (None, None))
- aid = episode_info.get('aid')
+ aid, cid = episode_info.get('aid', aid), episode_info.get('cid', cid)
return {
'id': episode_id,
@@ -957,7 +958,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'season_id': str_or_none(season_id),
'season_number': season_number,
'duration': float_or_none(play_info.get('timelength'), scale=1000),
- 'subtitles': self.extract_subtitles(episode_id, episode_info.get('cid'), aid=aid),
+ 'subtitles': self.extract_subtitles(episode_id, cid, aid=aid),
'__post_extractor': self.extract_comments(aid),
'http_headers': {'Referer': url},
}
From 7679b5241e87a2675ad238c99dad10954fe39d4f Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Fri, 11 Oct 2024 05:57:41 +1300
Subject: [PATCH 04/11] update tests
---
yt_dlp/extractor/bilibili.py | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index e7cfac55a0..b14fe19843 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -69,7 +69,7 @@ class BilibiliBaseIE(InfoExtractor):
def _extract_storyboard(self, duration, aid=None, bvid=None, cid=None):
if not (video_id := aid or bvid) or not duration:
- return {}
+ return
if storyboard_info := traverse_obj(self._download_json(
'https://api.bilibili.com/x/player/videoshot', video_id,
note='Downloading storyboard info', errnote='Failed to download storyboard info',
@@ -110,7 +110,6 @@ class BilibiliBaseIE(InfoExtractor):
'columns': cols,
'fragments': fragments,
}
- return {}
def extract_formats(self, play_info, aid=None, bvid=None, cid=None):
format_names = {
@@ -174,8 +173,9 @@ class BilibiliBaseIE(InfoExtractor):
}),
**parse_resolution(format_names.get(play_info.get('quality'))),
})
- formats.append(self._extract_storyboard(
- float_or_none(play_info.get('timelength'), scale=1000), aid=aid, bvid=bvid, cid=cid))
+ if storyboard_format := self._extract_storyboard(
+ float_or_none(play_info.get('timelength'), scale=1000), aid=aid, bvid=bvid, cid=cid):
+ formats.append(storyboard_format)
return formats
def _get_wbi_key(self, video_id):
@@ -1784,7 +1784,7 @@ class BilibiliAudioIE(BilibiliAudioBaseIE):
'id': '1003142',
'ext': 'm4a',
'title': '【tsukimi】YELLOW / 神山羊',
- 'artist': 'tsukimi',
+ 'artists': ['tsukimi'],
'comment_count': int,
'description': 'YELLOW的mp3版！',
'duration': 183,
@@ -1796,7 +1796,7 @@ class BilibiliAudioIE(BilibiliAudioBaseIE):
'thumbnail': r're:^https?://.+\.jpg',
'timestamp': 1564836614,
'upload_date': '20190803',
- 'uploader': 'tsukimi-つきみぐー',
+ 'uploader': '十六夜tsukimiつきみぐ',
'view_count': int,
},
}
@@ -1851,10 +1851,10 @@ class BilibiliAudioAlbumIE(BilibiliAudioBaseIE):
'url': 'https://www.bilibili.com/audio/am10624',
'info_dict': {
'id': '10624',
- 'title': '每日新曲推荐（每日11:00更新）',
+ 'title': '新曲推荐',
'description': '每天11:00更新，为你推送最新音乐',
},
- 'playlist_count': 19,
+ 'playlist_mincount': 10,
}
def _real_extract(self, url):
From 65a28bd5148f866ea385c7338ddc8c06f74efc6f Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Fri, 11 Oct 2024 06:19:04 +1300
Subject: [PATCH 05/11] add test for storyboard
---
yt_dlp/extractor/bilibili.py | 21 +++++++++++++++++++++
1 file changed, 21 insertions(+)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index b14fe19843..9b31e6fc90 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -604,6 +604,27 @@ class BiliBiliIE(BilibiliBaseIE):
'_old_archive_ids': ['bilibili 292734508_part1'],
},
}],
+ }, {
+ 'note': 'storyboard',
+ 'url': 'https://www.bilibili.com/video/av170001/',
+ 'info_dict': {
+ 'id': 'BV17x411w7KC_p1',
+ 'title': '【MV】保加利亚妖王AZIS视频合辑 p01 Хоп',
+ 'ext': 'mhtml',
+ 'upload_date': '20111109',
+ 'uploader_id': '122541',
+ 'view_count': int,
+ '_old_archive_ids': ['bilibili 170001_part1'],
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
+ 'uploader': '冰封.虾子',
+ 'timestamp': 1320850533,
+ 'comment_count': int,
+ 'tags': ['Hop', '保加利亚妖王', '保加利亚', 'Азис', 'azis', 'mv'],
+ 'description': 'md5:acfd7360b96547f031f7ebead9e66d9e',
+ 'like_count': int,
+ 'duration': 199.4,
+ },
+ 'params': {'format': 'sb', 'playlist_items': '1'},
}, {
'note': '301 redirect to bangumi link',
'url': 'https://www.bilibili.com/video/BV1TE411f7f1',
From a032d2b0d5fe57519505c6460e0b5c68e2efd628 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Wed, 23 Oct 2024 20:26:32 +1300
Subject: [PATCH 06/11] add heatmap, update tests
---
yt_dlp/extractor/bilibili.py | 105 +++++++++++++++++++++++++++--------
1 file changed, 83 insertions(+), 22 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index 9b31e6fc90..5fa4c08894 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -67,6 +67,41 @@ class BilibiliBaseIE(InfoExtractor):
f'Format(s) {missing_formats} are missing; you have to login or '
f'become a premium member to download them. {self._login_hint()}')
+ def _extract_heatmap(self, cid):
+ heatmap_json = self._download_json(
+ 'https://bvc.bilivideo.com/pbp/data', cid,
+ note='Downloading heatmap', errnote='Failed to download heatmap', fatal=False,
+ query={'cid': cid})
+ if not isinstance(heatmap_json, dict):
+ return
+ try:
+ duration = self._parse_json(heatmap_json['debug'])['max_time']
+ except Exception:
+ duration = None
+ step_sec = heatmap_json.get('step_sec', {int})
+ heatmap_data = traverse_obj(heatmap_json, ('events', 'default', {list}))
+ if not step_sec or not heatmap_data:
+ return
+ peak = max(heatmap_data)
+ if not peak:
+ return
+
+ for idx, heatmap_entry in enumerate(heatmap_data):
+ start_time = idx * step_sec
+ end_time = start_time + step_sec
+ if duration and end_time >= duration:
+ yield {
+ 'start_time': start_time,
+ 'end_time': duration,
+ 'value': heatmap_entry / peak,
+ }
+ break
+ yield {
+ 'start_time': start_time,
+ 'end_time': end_time,
+ 'value': heatmap_entry / peak,
+ }
+
def _extract_storyboard(self, duration, aid=None, bvid=None, cid=None):
if not (video_id := aid or bvid) or not duration:
return
@@ -343,6 +378,7 @@ class BilibiliBaseIE(InfoExtractor):
'description': f'{json.dumps(edges, ensure_ascii=False)}\n{metainfo.get("description", "")}',
'duration': float_or_none(play_info.get('timelength'), scale=1000),
'subtitles': self.extract_subtitles(video_id, cid),
+ 'heatmap': list(self._extract_heatmap(cid)),
}
@@ -358,7 +394,7 @@ class BiliBiliIE(BilibiliBaseIE):
'description': '滴妹今天唱Closer給你聽! 有史以来，被推最多次也是最久的歌曲，其实歌词跟我原本想像差蛮多的，不过还是好听！ 微博@阿滴英文',
'uploader_id': '65880958',
'uploader': '阿滴英文',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'duration': 554.117,
'tags': list,
'comment_count': int,
@@ -367,6 +403,7 @@ class BiliBiliIE(BilibiliBaseIE):
'like_count': int,
'view_count': int,
'_old_archive_ids': ['bilibili 8903802_part1'],
+ 'heatmap': [],
},
}, {
'note': 'old av URL version',
@@ -385,8 +422,9 @@ class BiliBiliIE(BilibiliBaseIE):
'comment_count': int,
'view_count': int,
'tags': list,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 1074402_part1'],
+ 'heatmap': [],
},
'params': {'skip_download': True},
}, {
@@ -404,7 +442,7 @@ class BiliBiliIE(BilibiliBaseIE):
'title': '物语中的人物是如何吐槽自己的OP的 p01 Staple Stable/战场原+羽川',
'tags': 'count:10',
'timestamp': 1589601697,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'uploader': '打牌还是打桩',
'uploader_id': '150259984',
'like_count': int,
@@ -414,6 +452,7 @@ class BiliBiliIE(BilibiliBaseIE):
'description': 'md5:e3c401cf7bc363118d1783dd74068a68',
'duration': 90.314,
'_old_archive_ids': ['bilibili 498159642_part1'],
+ 'heatmap': 'count:90',
},
}],
}, {
@@ -425,7 +464,7 @@ class BiliBiliIE(BilibiliBaseIE):
'title': '物语中的人物是如何吐槽自己的OP的 p01 Staple Stable/战场原+羽川',
'tags': 'count:10',
'timestamp': 1589601697,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'uploader': '打牌还是打桩',
'uploader_id': '150259984',
'like_count': int,
@@ -435,6 +474,7 @@ class BiliBiliIE(BilibiliBaseIE):
'description': 'md5:e3c401cf7bc363118d1783dd74068a68',
'duration': 90.314,
'_old_archive_ids': ['bilibili 498159642_part1'],
+ 'heatmap': 'count:90',
},
}, {
'url': 'https://www.bilibili.com/video/av8903802/',
@@ -447,13 +487,14 @@ class BiliBiliIE(BilibiliBaseIE):
'timestamp': 1488353834,
'uploader_id': '65880958',
'uploader': '阿滴英文',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'duration': 554.117,
'tags': list,
'comment_count': int,
'view_count': int,
'like_count': int,
'_old_archive_ids': ['bilibili 8903802_part1'],
+ 'heatmap': [],
},
'params': {
'skip_download': True,
@@ -476,8 +517,9 @@ class BiliBiliIE(BilibiliBaseIE):
'comment_count': int,
'view_count': int,
'like_count': int,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 463665680_part1'],
+ 'heatmap': 'count:96',
},
'params': {'skip_download': True},
}, {
@@ -495,8 +537,9 @@ class BiliBiliIE(BilibiliBaseIE):
'uploader_id': '528182630',
'view_count': int,
'like_count': int,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 893839363_part1'],
+ 'heatmap': [],
},
}, {
'note': 'newer festival video',
@@ -513,8 +556,9 @@ class BiliBiliIE(BilibiliBaseIE):
'uploader_id': '8469526',
'view_count': int,
'like_count': int,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 778246196_part1'],
+ 'heatmap': 'count:93',
},
}, {
'note': 'legacy flv/mp4 video',
@@ -532,8 +576,9 @@ class BiliBiliIE(BilibiliBaseIE):
'comment_count': int,
'like_count': int,
'tags': list,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 4120229_part4'],
+ 'heatmap': [],
},
'params': {'extractor_args': {'bilibili': {'prefer_multi_flv': ['32']}}},
'playlist_count': 19,
@@ -562,8 +607,9 @@ class BiliBiliIE(BilibiliBaseIE):
'view_count': int,
'like_count': int,
'tags': list,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 15700301_part1'],
+ 'heatmap': [],
},
}, {
'note': 'interactive/split-path video',
@@ -581,7 +627,7 @@ class BiliBiliIE(BilibiliBaseIE):
'comment_count': int,
'view_count': int,
'like_count': int,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 292734508_part1'],
},
'playlist_count': 33,
@@ -600,8 +646,9 @@ class BiliBiliIE(BilibiliBaseIE):
'comment_count': int,
'view_count': int,
'like_count': int,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 292734508_part1'],
+ 'heatmap': [],
},
}],
}, {
@@ -623,6 +670,7 @@ class BiliBiliIE(BilibiliBaseIE):
'description': 'md5:acfd7360b96547f031f7ebead9e66d9e',
'like_count': int,
'duration': 199.4,
+ 'heatmap': 'count:68',
},
'params': {'format': 'sb', 'playlist_items': '1'},
}, {
@@ -643,7 +691,8 @@ class BiliBiliIE(BilibiliBaseIE):
'duration': 1183.957,
'timestamp': 1571648124,
'upload_date': '20191021',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
+ 'heatmap': [],
},
}, {
'note': 'video has subtitles, which requires login',
@@ -662,7 +711,7 @@ class BiliBiliIE(BilibiliBaseIE):
'comment_count': int,
'view_count': int,
'like_count': int,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'subtitles': 'count:2', # login required for CC subtitle
'_old_archive_ids': ['bilibili 898179753_part1'],
},
@@ -842,6 +891,7 @@ class BiliBiliIE(BilibiliBaseIE):
'__post_extractor': self.extract_comments(aid) if idx == 0 else None,
} for idx, fragment in enumerate(formats[0]['fragments'])],
'duration': float_or_none(play_info.get('timelength'), scale=1000),
+ 'heatmap': list(self._extract_heatmap(cid)),
}
else:
return {
@@ -851,6 +901,7 @@ class BiliBiliIE(BilibiliBaseIE):
'chapters': self._get_chapters(aid, cid),
'subtitles': self.extract_subtitles(video_id, cid),
'__post_extractor': self.extract_comments(aid),
+ 'heatmap': list(self._extract_heatmap(cid)),
}
@@ -874,7 +925,8 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'duration': 1420.791,
'timestamp': 1320412200,
'upload_date': '20111104',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
+ 'heatmap': 'count:96',
},
}, {
'url': 'https://www.bilibili.com/bangumi/play/ep267851',
@@ -893,7 +945,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'duration': 1425.256,
'timestamp': 1554566400,
'upload_date': '20190406',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
},
'skip': 'Geo-restricted',
}, {
@@ -914,7 +966,8 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'duration': 1922.129,
'timestamp': 1602853860,
'upload_date': '20201016',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
+ 'heatmap': 'count:97',
},
}]
@@ -982,6 +1035,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'subtitles': self.extract_subtitles(episode_id, cid, aid=aid),
'__post_extractor': self.extract_comments(aid),
'http_headers': {'Referer': url},
+ 'heatmap': list(self._extract_heatmap(cid)),
}
@@ -1019,7 +1073,8 @@ class BiliBiliBangumiMediaIE(BilibiliBaseIE):
'duration': 1525.777,
'timestamp': 1425074413,
'upload_date': '20150227',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
+ 'heatmap': 'count:96',
},
}],
}]
@@ -1074,7 +1129,8 @@ class BiliBiliBangumiSeasonIE(BilibiliBaseIE):
'duration': 1436.992,
'timestamp': 1343185080,
'upload_date': '20120725',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
+ 'heatmap': 'count:96',
},
}],
}]
@@ -1132,6 +1188,7 @@ class BilibiliCheeseBaseIE(BilibiliBaseIE):
'subtitles': self.extract_subtitles(ep_id, cid, aid=aid),
'__post_extractor': self.extract_comments(aid),
'http_headers': self._HEADERS,
+ 'heatmap': list(self._extract_heatmap(cid)),
}
def _download_season_info(self, query_key, video_id):
@@ -1157,8 +1214,9 @@ class BilibiliCheeseIE(BilibiliCheeseBaseIE):
'duration': 221,
'timestamp': 1695549606,
'upload_date': '20230924',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'view_count': int,
+ 'heatmap': 'count:74',
},
}]
@@ -1190,8 +1248,9 @@ class BilibiliCheeseSeasonIE(BilibiliCheeseBaseIE):
'duration': 221,
'timestamp': 1695549606,
'upload_date': '20230924',
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'view_count': int,
+ 'heatmap': 'count:74',
},
}],
'params': {'playlist_items': '1'},
@@ -1563,6 +1622,7 @@ class BilibiliPlaylistIE(BilibiliSpaceListBaseIE):
'view_count': int,
'like_count': int,
'_old_archive_ids': ['bilibili 687146339_part1'],
+ 'heatmap': [],
},
'params': {'noplaylist': True},
}, {
@@ -1757,8 +1817,9 @@ class BiliBiliSearchIE(SearchInfoExtractor):
'comment_count': int,
'view_count': int,
'like_count': int,
- 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+ 'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 988222410_part1'],
+ 'heatmap': [],
},
}],
}]
From 43c6c434aefa7da2fb75ba870e201df2be4aa11a Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Wed, 23 Oct 2024 20:40:14 +1300
Subject: [PATCH 07/11] make storyboard extraction non-fatal
---
yt_dlp/extractor/bilibili.py | 12 +++++++-----
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index 5fa4c08894..90505029df 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -113,13 +113,15 @@ class BilibiliBaseIE(InfoExtractor):
'aid': aid,
'bvid': bvid,
'cid': cid,
- })), ('data', {lambda v: v if v['image'] and v['index'] else None})):
- rows, cols = traverse_obj(storyboard_info, (('img_x_len', 'img_y_len'),))
+ })), ('data', {lambda v: v if v.get('image') and v.get('index') else None})):
+ rows, cols = storyboard_info.get('img_x_len'), storyboard_info.get('img_y_len')
fragments = []
last_duration = 0.0
for i, url in enumerate(storyboard_info['image'], start=1):
- duration_index = i * rows * cols - 1
- if duration_index < len(storyboard_info['index']) - 1:
+ if not rows or not cols:
+ fragments.append({'url': sanitize_url(url)})
+ continue
+ elif (duration_index := i * rows * cols - 1) < len(storyboard_info['index']) - 1:
current_duration = traverse_obj(storyboard_info, ('index', duration_index))
else:
current_duration = duration
@@ -127,7 +129,7 @@ class BilibiliBaseIE(InfoExtractor):
break
fragments.append({
'url': sanitize_url(url),
- 'duration': current_duration - last_duration,
+ 'duration': current_duration - last_duration if current_duration is not None else None,
})
if fragments:
return {
From 6f5a908dffa7a0ed51998355da7126fce69718ec Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Wed, 23 Oct 2024 20:45:03 +1300
Subject: [PATCH 08/11] fix heatmap extraction
---
yt_dlp/extractor/bilibili.py | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index 90505029df..3ac9c7cd00 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -74,11 +74,8 @@ class BilibiliBaseIE(InfoExtractor):
query={'cid': cid})
if not isinstance(heatmap_json, dict):
return
- try:
- duration = self._parse_json(heatmap_json['debug'])['max_time']
- except Exception:
- duration = None
- step_sec = heatmap_json.get('step_sec', {int})
+ duration = self._parse_json(heatmap_json['debug']).get('max_time')
+ step_sec = traverse_obj(heatmap_json, ('step_sec', {int}))
heatmap_data = traverse_obj(heatmap_json, ('events', 'default', {list}))
if not step_sec or not heatmap_data:
return
From eb03632cc7fd5f27c48cff0ee76fedf7d0d598e8 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Wed, 23 Oct 2024 21:10:05 +1300
Subject: [PATCH 09/11] do not extract storyboard in extract_formats
---
yt_dlp/extractor/bilibili.py | 39 ++++++++++++++++++++++++------------
1 file changed, 26 insertions(+), 13 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index 3ac9c7cd00..45183a8c7a 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -42,6 +42,7 @@ from ..utils import (
unsmuggle_url,
url_or_none,
urlencode_postdata,
+ value,
variadic,
)
@@ -145,7 +146,7 @@ class BilibiliBaseIE(InfoExtractor):
'fragments': fragments,
}
- def extract_formats(self, play_info, aid=None, bvid=None, cid=None):
+ def extract_formats(self, play_info):
format_names = {
r['quality']: traverse_obj(r, 'new_description', 'display_desc')
for r in traverse_obj(play_info, ('support_formats', lambda _, v: v['quality']))
@@ -207,9 +208,6 @@ class BilibiliBaseIE(InfoExtractor):
}),
**parse_resolution(format_names.get(play_info.get('quality'))),
})
- if storyboard_format := self._extract_storyboard(
- float_or_none(play_info.get('timelength'), scale=1000), aid=aid, bvid=bvid, cid=cid):
- formats.append(storyboard_format)
return formats
def _get_wbi_key(self, video_id):
@@ -369,13 +367,19 @@ class BilibiliBaseIE(InfoExtractor):
cid_edges = self._get_divisions(video_id, graph_version, {1: {'cid': cid}}, 1)
for cid, edges in cid_edges.items():
play_info = self._download_playinfo(video_id, cid, headers=headers)
+ formats = self.extract_formats(play_info)
+ duration = float_or_none(play_info.get('timelength'), scale=1000)
+ if storyboard_format := self._extract_storyboard(
+ duration=duration,
+ bvid=video_id, cid=cid):
+ formats.append(storyboard_format)
yield {
**metainfo,
'id': f'{video_id}_{cid}',
'title': f'{metainfo.get("title")} - {next(iter(edges.values())).get("title")}',
- 'formats': self.extract_formats(play_info, bvid=video_id, cid=cid),
+ 'formats': formats,
'description': f'{json.dumps(edges, ensure_ascii=False)}\n{metainfo.get("description", "")}',
- 'duration': float_or_none(play_info.get('timelength'), scale=1000),
+ 'duration': duration,
'subtitles': self.extract_subtitles(video_id, cid),
'heatmap': list(self._extract_heatmap(cid)),
}
@@ -845,14 +849,17 @@ class BiliBiliIE(BilibiliBaseIE):
duration=traverse_obj(initial_state, ('videoData', 'duration', {int_or_none})),
__post_extractor=self.extract_comments(aid))
else:
- formats = self.extract_formats(play_info, bvid=video_id, cid=cid)
+ formats = self.extract_formats(play_info)
+ formats.append(self._extract_storyboard(
+ duration=float_or_none(play_info.get('timelength'), scale=1000),
+ bvid=video_id, cid=cid))
if not traverse_obj(play_info, ('dash')):
# we only have legacy formats and need additional work
has_qn = lambda x: x in traverse_obj(formats, (..., 'quality'))
for qn in traverse_obj(play_info, ('accept_quality', lambda _, v: not has_qn(v), {int})):
formats.extend(traverse_obj(
- self.extract_formats(self._download_playinfo(video_id, cid, headers=headers, qn=qn), bvid=video_id, cid=cid),
+ self.extract_formats(self._download_playinfo(video_id, cid, headers=headers, qn=qn)),
lambda _, v: not has_qn(v['quality'])))
self._check_missing_formats(play_info, formats)
flv_formats = traverse_obj(formats, lambda _, v: v['fragments'])
@@ -990,7 +997,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
aid, cid = episode_info.get('aid'), episode_info.get('cid')
play_info = traverse_obj(play_info, ('result', 'video_info', {dict})) or {}
- formats = self.extract_formats(play_info, aid=aid, cid=cid)
+ formats = self.extract_formats(play_info)
if not formats and (premium_only or '成为大会员抢先看' in webpage or '开通大会员观看' in webpage):
self.raise_login_required('This video is for premium members only')
@@ -1011,7 +1018,9 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
), (None, None))
aid, cid = episode_info.get('aid', aid), episode_info.get('cid', cid)
-
+ duration = float_or_none(play_info.get('timelength'), scale=1000)
+ if storyboard_format := self._extract_storyboard(duration=duration, aid=aid, cid=cid):
+ formats.append(storyboard_format)
return {
'id': episode_id,
'formats': formats,
@@ -1030,7 +1039,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'season': str_or_none(season_title),
'season_id': str_or_none(season_id),
'season_number': season_number,
- 'duration': float_or_none(play_info.get('timelength'), scale=1000),
+ 'duration': duration,
'subtitles': self.extract_subtitles(episode_id, cid, aid=aid),
'__post_extractor': self.extract_comments(aid),
'http_headers': {'Referer': url},
@@ -1163,10 +1172,14 @@ class BilibiliCheeseBaseIE(BilibiliBaseIE):
query={'avid': aid, 'cid': cid, 'ep_id': ep_id, 'fnval': 16, 'fourk': 1},
headers=self._HEADERS, note='Downloading playinfo')['data']
+ formats = self.extract_formats(play_info)
+ duration = traverse_obj(episode_info, ('duration', {int_or_none}))
+ if storyboard_format := self._extract_storyboard(duration=duration, aid=aid, cid=cid):
+ formats.append(storyboard_format)
return {
'id': str_or_none(ep_id),
'episode_id': str_or_none(ep_id),
- 'formats': self.extract_formats(play_info, aid=aid, cid=cid),
+ 'formats': formats,
'extractor_key': BilibiliCheeseIE.ie_key(),
'extractor': BilibiliCheeseIE.IE_NAME,
'webpage_url': f'https://www.bilibili.com/cheese/play/ep{ep_id}',
@@ -1174,7 +1187,7 @@ class BilibiliCheeseBaseIE(BilibiliBaseIE):
'episode': ('title', {str}),
'title': {lambda v: v and join_nonempty('index', 'title', delim=' - ', from_dict=v)},
'alt_title': ('subtitle', {str}),
- 'duration': ('duration', {int_or_none}),
+ 'duration': {value(duration)},
'episode_number': ('index', {int_or_none}),
'thumbnail': ('cover', {url_or_none}),
'timestamp': ('release_date', {int_or_none}),
From 8631ff86d93f6fe8fb43580c71d772a3d1b30c6b Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Wed, 23 Oct 2024 22:50:34 +1300
Subject: [PATCH 10/11] fix: _parse_json needs `video_id`
---
yt_dlp/extractor/bilibili.py | 24 ++++++++++++------------
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index 45183a8c7a..41684751ab 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -75,7 +75,7 @@ class BilibiliBaseIE(InfoExtractor):
query={'cid': cid})
if not isinstance(heatmap_json, dict):
return
- duration = self._parse_json(heatmap_json['debug']).get('max_time')
+ duration = self._parse_json(heatmap_json['debug'], cid).get('max_time')
step_sec = traverse_obj(heatmap_json, ('step_sec', {int}))
heatmap_data = traverse_obj(heatmap_json, ('events', 'default', {list}))
if not step_sec or not heatmap_data:
@@ -455,7 +455,7 @@ class BiliBiliIE(BilibiliBaseIE):
'description': 'md5:e3c401cf7bc363118d1783dd74068a68',
'duration': 90.314,
'_old_archive_ids': ['bilibili 498159642_part1'],
- 'heatmap': 'count:90',
+ 'heatmap': list,
},
}],
}, {
@@ -477,7 +477,7 @@ class BiliBiliIE(BilibiliBaseIE):
'description': 'md5:e3c401cf7bc363118d1783dd74068a68',
'duration': 90.314,
'_old_archive_ids': ['bilibili 498159642_part1'],
- 'heatmap': 'count:90',
+ 'heatmap': list,
},
}, {
'url': 'https://www.bilibili.com/video/av8903802/',
@@ -522,7 +522,7 @@ class BiliBiliIE(BilibiliBaseIE):
'like_count': int,
'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 463665680_part1'],
- 'heatmap': 'count:96',
+ 'heatmap': list,
},
'params': {'skip_download': True},
}, {
@@ -561,7 +561,7 @@ class BiliBiliIE(BilibiliBaseIE):
'like_count': int,
'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'_old_archive_ids': ['bilibili 778246196_part1'],
- 'heatmap': 'count:93',
+ 'heatmap': list,
},
}, {
'note': 'legacy flv/mp4 video',
@@ -673,7 +673,7 @@ class BiliBiliIE(BilibiliBaseIE):
'description': 'md5:acfd7360b96547f031f7ebead9e66d9e',
'like_count': int,
'duration': 199.4,
- 'heatmap': 'count:68',
+ 'heatmap': list,
},
'params': {'format': 'sb', 'playlist_items': '1'},
}, {
@@ -932,7 +932,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'timestamp': 1320412200,
'upload_date': '20111104',
'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
- 'heatmap': 'count:96',
+ 'heatmap': list,
},
}, {
'url': 'https://www.bilibili.com/bangumi/play/ep267851',
@@ -973,7 +973,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
'timestamp': 1602853860,
'upload_date': '20201016',
'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
- 'heatmap': 'count:97',
+ 'heatmap': list,
},
}]
@@ -1082,7 +1082,7 @@ class BiliBiliBangumiMediaIE(BilibiliBaseIE):
'timestamp': 1425074413,
'upload_date': '20150227',
'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
- 'heatmap': 'count:96',
+ 'heatmap': list,
},
}],
}]
@@ -1138,7 +1138,7 @@ class BiliBiliBangumiSeasonIE(BilibiliBaseIE):
'timestamp': 1343185080,
'upload_date': '20120725',
'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
- 'heatmap': 'count:96',
+ 'heatmap': list,
},
}],
}]
@@ -1228,7 +1228,7 @@ class BilibiliCheeseIE(BilibiliCheeseBaseIE):
'upload_date': '20230924',
'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'view_count': int,
- 'heatmap': 'count:74',
+ 'heatmap': list,
},
}]
@@ -1262,7 +1262,7 @@ class BilibiliCheeseSeasonIE(BilibiliCheeseBaseIE):
'upload_date': '20230924',
'thumbnail': r're:https?://.*\.(?:jpg|jpeg|png)$',
'view_count': int,
- 'heatmap': 'count:74',
+ 'heatmap': list,
},
}],
'params': {'playlist_items': '1'},
From fe0541845dd455e4e0ee15d91c6f663d17ece439 Mon Sep 17 00:00:00 2001
From: grqx_wsl <173253225+grqx@users.noreply.github.com>
Date: Tue, 3 Dec 2024 17:36:08 +1300
Subject: [PATCH 11/11] extract storyboard formats after the flv check
---
yt_dlp/extractor/bilibili.py | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py
index 0c4b888acc..242c332b5c 100644
--- a/yt_dlp/extractor/bilibili.py
+++ b/yt_dlp/extractor/bilibili.py
@@ -857,10 +857,6 @@ class BiliBiliIE(BilibiliBaseIE):
f'This is a supporter-only video, only the preview will be extracted: {msg}',
video_id=video_id)
- formats.append(self._extract_storyboard(
- duration=float_or_none(play_info.get('timelength'), scale=1000),
- bvid=video_id, cid=cid))
-
if not traverse_obj(play_info, 'dash'):
# we only have legacy formats and need additional work
has_qn = lambda x: x in traverse_obj(formats, (..., 'quality'))
@@ -906,6 +902,10 @@ class BiliBiliIE(BilibiliBaseIE):
'duration': float_or_none(play_info.get('timelength'), scale=1000),
}
+ formats.append(self._extract_storyboard(
+ duration=float_or_none(play_info.get('timelength'), scale=1000),
+ bvid=video_id, cid=cid))
+
return {
**metainfo,
'formats': formats,