mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-14 04:13:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

BBC Issue 9701: NEXT_DATA field video extraction for bbc

Browse source 
Some bbc articles with embedded video have the data
for them within a json structure tagged with NEXT_DATA.
Add a parser for this case.

Links tested:
https://www.bbc.com/news/uk-68546268
https://www.bbc.com/news/world-middle-east-68778149
https://www.bbc.com/reel/video/p07c6sb6/how-positive-thinking-is-harming-your-happiness
This commit is contained in:
Kyle Gonsalves 2024-04-16 16:35:51 -07:00
parent 315b354429
commit 401e28b318
1 changed files with 51 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

52
yt_dlp/extractor/bbc.py
View file

@ -602,7 +602,7 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'url': 'http://www.bbc.com/news/world-europe-32668511',
'info_dict': {
'id': 'world-europe-32668511',
'title': 'Russia stages massive WW2 parade',
'title': 'Russia stages massive WW2 parade despite Western boycott',
'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
},
'playlist_count': 2,
@ -791,6 +791,17 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'timestamp': 1638230731,
'upload_date': '20211130',
},
}, {
# video with script id __NEXT_DATA__ and value as JSON string
'url': 'https://www.bbc.com/news/uk-68546268',
'info_dict': {
'id': 'p0hj0lq7',
'ext': 'mp4',
'title': 'Nasser Hospital doctor describes his treatment by IDF',
'description': 'Doctor Abu Sabha said he was detained by Israeli forces after the raid on Nasser Hospital and feared for his life.\n\nThe IDF said "during the activity, about 200 terrorists and suspects of terrorist activity were detained, including some who posed as medical teams, many weapons were found, as well as closed medicines intended for Israeli hostages."',
'thumbnail': r're:https?://.+/.+\.jpg',
'timestamp': 1710270205000,
},
}, {
# single video article embedded with data-media-vpid
'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
@ -1255,6 +1266,45 @@ def extract_all(pattern):
lambda s: self._parse_json(s, playlist_id, fatal=False),
re.findall(pattern, webpage))))
# US accessed article with single embedded video (e.g.
# https://www.bbc.com/news/uk-68546268)
video_id = self._match_id(url)
next_data = self._search_nextjs_data(webpage, video_id)['props']['pageProps']['page']
video_data = None
timestamp = None
for key in next_data:
for item in (try_get(next_data, lambda x: x[key]['contents'], list) or []):
if item.get('type') == 'video':
video_data = item
elif item.get('type') == 'timestamp':
timestamp = item
if video_data:
for item in (try_get(video_data, lambda x: x['model']['blocks'], list) or []):
if item.get('type') == 'media':
for subtype in (try_get(item, lambda x: x['model']['blocks'], list) or []):
if subtype.get('type') == 'mediaMetadata':
model = subtype.get('model')
if model:
item_id = try_get(model, lambda x: x['versions'][0]['versionId'])
item_thumbnail = model.get('imageUrl')
item_title = model.get('title')
formats, subtitles = self._download_media_selector(item_id)
synopses = model.get('synopses') or {}
item_time = None
if timestamp:
item_time = try_get(timestamp, lambda x: x['model']['timestamp'])
entries.append({
'id': item_id,
'title': item_title,
'thumbnail': item_thumbnail,
'formats': formats,
'subtitles': subtitles,
'timestamp': item_time,
'description': dict_get(synopses, ('long', 'medium', 'short'))
})
return self.playlist_result(
entries, playlist_id, playlist_title, playlist_description)
# Multiple video article (e.g.
# http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX