X-Git-Url: https://jfr.im/git/yt-dlp.git/blobdiff_plain/54c2521ca656f1a0897d21cbdf34cbad4e51735b..315b3544296bb83012e20ee3af9d3cbf5600dd1c:/yt_dlp/extractor/cbc.py
diff --git a/yt_dlp/extractor/cbc.py b/yt_dlp/extractor/cbc.py
index 392c77884..ff320dd68 100644
--- a/yt_dlp/extractor/cbc.py
+++ b/yt_dlp/extractor/cbc.py
@@ -1,24 +1,25 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-import json
import base64
+import json
+import re
import time
+import urllib.parse
+import xml.etree.ElementTree
from .common import InfoExtractor
from ..compat import (
compat_str,
)
from ..utils import (
+ ExtractorError,
int_or_none,
join_nonempty,
js_to_json,
orderedSet,
+ parse_iso8601,
smuggle_url,
strip_or_none,
+ traverse_obj,
try_get,
- ExtractorError,
)
@@ -66,6 +67,7 @@ class CBCIE(InfoExtractor):
'uploader': 'CBCC-NEW',
'timestamp': 255977160,
},
+ 'skip': '404 Not Found',
}, {
# multiple iframes
'url': 'http://www.cbc.ca/natureofthings/blog/birds-eye-view-from-vancouvers-burrard-street-bridge-how-we-got-the-shot',
@@ -97,7 +99,7 @@ class CBCIE(InfoExtractor):
# multiple CBC.APP.Caffeine.initInstance(...)
'url': 'http://www.cbc.ca/news/canada/calgary/dog-indoor-exercise-winter-1.3928238',
'info_dict': {
- 'title': 'Keep Rover active during the deep freeze with doggie pushups and other fun indoor tasks',
+ 'title': 'Keep Rover active during the deep freeze with doggie pushups and other fun indoor tasks', # FIXME
'id': 'dog-indoor-exercise-winter-1.3928238',
'description': 'md5:c18552e41726ee95bd75210d1ca9194c',
},
@@ -127,9 +129,9 @@ def _extract_player_init(self, player_init, display_id):
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
- title = self._og_search_title(webpage, default=None) or self._html_search_meta(
- 'twitter:title', webpage, 'title', default=None) or self._html_search_regex(
- r'([^<]+)', webpage, 'title', fatal=False)
+ title = (self._og_search_title(webpage, default=None)
+ or self._html_search_meta('twitter:title', webpage, 'title', default=None)
+ or self._html_extract_title(webpage))
entries = [
self._extract_player_init(player_init, display_id)
for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)]
@@ -149,7 +151,7 @@ def _real_extract(self, url):
class CBCPlayerIE(InfoExtractor):
IE_NAME = 'cbc.ca:player'
- _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P\d+)'
+ _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P(?:\d\.)?\d+)'
_TESTS = [{
'url': 'http://www.cbc.ca/player/play/2683190193',
'md5': '64d25f841ddf4ddb28a235338af32e2c',
@@ -162,10 +164,53 @@ class CBCPlayerIE(InfoExtractor):
'upload_date': '20160210',
'uploader': 'CBCC-NEW',
},
- 'skip': 'Geo-restricted to Canada',
+ 'skip': 'Geo-restricted to Canada and no longer available',
+ }, {
+ 'url': 'http://www.cbc.ca/i/caffeine/syndicate/?mediaId=2657631896',
+ 'md5': 'e5e708c34ae6fca156aafe17c43e8b75',
+ 'info_dict': {
+ 'id': '2657631896',
+ 'ext': 'mp3',
+ 'title': 'CBC Montreal is organizing its first ever community hackathon!',
+ 'description': 'md5:dd3b692f0a139b0369943150bd1c46a9',
+ 'timestamp': 1425704400,
+ 'upload_date': '20150307',
+ 'uploader': 'CBCC-NEW',
+ 'thumbnail': 'http://thumbnails.cbc.ca/maven_legacy/thumbnails/sonali-karnick-220.jpg',
+ 'chapters': [],
+ 'duration': 494.811,
+ 'categories': ['AudioMobile/All in a Weekend Montreal'],
+ 'tags': 'count:8',
+ 'location': 'Quebec',
+ 'series': 'All in a Weekend Montreal',
+ 'season': 'Season 2015',
+ 'season_number': 2015,
+ 'media_type': 'Excerpt',
+ },
+ }, {
+ 'url': 'http://www.cbc.ca/i/caffeine/syndicate/?mediaId=2164402062',
+ 'md5': '33fcd8f6719b9dd60a5e73adcb83b9f6',
+ 'info_dict': {
+ 'id': '2164402062',
+ 'ext': 'mp4',
+ 'title': 'Cancer survivor four times over',
+ 'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.',
+ 'timestamp': 1320410746,
+ 'upload_date': '20111104',
+ 'uploader': 'CBCC-NEW',
+ 'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/277/67/cancer_852x480_2164412612.jpg',
+ 'chapters': [],
+ 'duration': 186.867,
+ 'series': 'CBC News: Windsor at 6:00',
+ 'categories': ['News/Canada/Windsor'],
+ 'location': 'Windsor',
+ 'tags': ['cancer'],
+ 'creators': ['Allison Johnson'],
+ 'media_type': 'Excerpt',
+ },
}, {
# Redirected from http://www.cbc.ca/player/AudioMobile/All%20in%20a%20Weekend%20Montreal/ID/2657632011/
- 'url': 'http://www.cbc.ca/player/play/2657631896',
+ 'url': 'https://www.cbc.ca/player/play/1.2985700',
'md5': 'e5e708c34ae6fca156aafe17c43e8b75',
'info_dict': {
'id': '2657631896',
@@ -175,9 +220,19 @@ class CBCPlayerIE(InfoExtractor):
'timestamp': 1425704400,
'upload_date': '20150307',
'uploader': 'CBCC-NEW',
+ 'thumbnail': 'http://thumbnails.cbc.ca/maven_legacy/thumbnails/sonali-karnick-220.jpg',
+ 'chapters': [],
+ 'duration': 494.811,
+ 'categories': ['AudioMobile/All in a Weekend Montreal'],
+ 'tags': 'count:8',
+ 'location': 'Quebec',
+ 'series': 'All in a Weekend Montreal',
+ 'season': 'Season 2015',
+ 'season_number': 2015,
+ 'media_type': 'Excerpt',
},
}, {
- 'url': 'http://www.cbc.ca/player/play/2164402062',
+ 'url': 'https://www.cbc.ca/player/play/1.1711287',
'md5': '33fcd8f6719b9dd60a5e73adcb83b9f6',
'info_dict': {
'id': '2164402062',
@@ -187,11 +242,60 @@ class CBCPlayerIE(InfoExtractor):
'timestamp': 1320410746,
'upload_date': '20111104',
'uploader': 'CBCC-NEW',
+ 'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/277/67/cancer_852x480_2164412612.jpg',
+ 'chapters': [],
+ 'duration': 186.867,
+ 'series': 'CBC News: Windsor at 6:00',
+ 'categories': ['News/Canada/Windsor'],
+ 'location': 'Windsor',
+ 'tags': ['cancer'],
+ 'creators': ['Allison Johnson'],
+ 'media_type': 'Excerpt',
},
+ }, {
+ # Has subtitles
+ # These broadcasts expire after ~1 month, can find new test URL here:
+ # https://www.cbc.ca/player/news/TV%20Shows/The%20National/Latest%20Broadcast
+ 'url': 'https://www.cbc.ca/player/play/1.7159484',
+ 'md5': '6ed6cd0fc2ef568d2297ba68a763d455',
+ 'info_dict': {
+ 'id': '2324213316001',
+ 'ext': 'mp4',
+ 'title': 'The National | School boards sue social media giants',
+ 'description': 'md5:4b4db69322fa32186c3ce426da07402c',
+ 'timestamp': 1711681200,
+ 'duration': 2743.400,
+ 'subtitles': {'eng': [{'ext': 'vtt', 'protocol': 'm3u8_native'}]},
+ 'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/607/559/thumbnail.jpeg',
+ 'uploader': 'CBCC-NEW',
+ 'chapters': 'count:5',
+ 'upload_date': '20240329',
+ 'categories': 'count:4',
+ 'series': 'The National - Full Show',
+ 'tags': 'count:1',
+ 'creators': ['News'],
+ 'location': 'Canada',
+ 'media_type': 'Full Program',
+ },
+ }, {
+ 'url': 'cbcplayer:1.7159484',
+ 'only_matching': True,
+ }, {
+ 'url': 'cbcplayer:2164402062',
+ 'only_matching': True,
+ }, {
+ 'url': 'http://www.cbc.ca/player/play/2657631896',
+ 'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
+ if '.' in video_id:
+ webpage = self._download_webpage(f'https://www.cbc.ca/player/play/{video_id}', video_id)
+ video_id = self._search_json(
+ r'window\.__INITIAL_STATE__\s*=', webpage,
+ 'initial state', video_id)['video']['currentClip']['mediaId']
+
return {
'_type': 'url_transparent',
'ie_key': 'ThePlatform',
@@ -200,12 +304,45 @@ def _real_extract(self, url):
'force_smil_url': True
}),
'id': video_id,
+ '_format_sort_fields': ('res', 'proto') # Prioritize direct http formats over HLS
+ }
+
+
+class CBCPlayerPlaylistIE(InfoExtractor):
+ IE_NAME = 'cbc.ca:player:playlist'
+ _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?:player/)(?!play/)(?P[^?#]+)'
+ _TESTS = [{
+ 'url': 'https://www.cbc.ca/player/news/TV%20Shows/The%20National/Latest%20Broadcast',
+ 'playlist_mincount': 25,
+ 'info_dict': {
+ 'id': 'news/tv shows/the national/latest broadcast',
+ }
+ }, {
+ 'url': 'https://www.cbc.ca/player/news/Canada/North',
+ 'playlist_mincount': 25,
+ 'info_dict': {
+ 'id': 'news/canada/north',
}
+ }]
+
+ def _real_extract(self, url):
+ playlist_id = urllib.parse.unquote(self._match_id(url)).lower()
+ webpage = self._download_webpage(url, playlist_id)
+ json_content = self._search_json(
+ r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', playlist_id)
+
+ def entries():
+ for video_id in traverse_obj(json_content, (
+ 'video', 'clipsByCategory', lambda k, _: k.lower() == playlist_id, 'items', ..., 'id'
+ )):
+ yield self.url_result(f'https://www.cbc.ca/player/play/{video_id}', CBCPlayerIE)
+
+ return self.playlist_result(entries(), playlist_id)
class CBCGemIE(InfoExtractor):
IE_NAME = 'gem.cbc.ca'
- _VALID_URL = r'https?://gem\.cbc\.ca/media/(?P[0-9a-z-]+/s[0-9]+[a-z][0-9]+)'
+ _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P[0-9a-z-]+/s[0-9]+[a-z][0-9]+)'
_TESTS = [{
# This is a normal, public, TV show video
'url': 'https://gem.cbc.ca/media/schitts-creek/s06e01',
@@ -248,6 +385,9 @@ class CBCGemIE(InfoExtractor):
},
'params': {'format': 'bv'},
'skip': 'Geo-restricted to Canada',
+ }, {
+ 'url': 'https://gem.cbc.ca/nadiyas-family-favourites/s01e01',
+ 'only_matching': True,
}]
_GEO_COUNTRIES = ['CA']
@@ -278,12 +418,12 @@ def _new_claims_token(self, email, password):
data = json.dumps({'jwt': sig}).encode()
headers = {'content-type': 'application/json', 'ott-device-type': 'web'}
resp = self._download_json('https://services.radio-canada.ca/ott/cbc-api/v2/token',
- None, data=data, headers=headers)
+ None, data=data, headers=headers, expected_status=426)
cbc_access_token = resp['accessToken']
headers = {'content-type': 'application/json', 'ott-device-type': 'web', 'ott-access-token': cbc_access_token}
resp = self._download_json('https://services.radio-canada.ca/ott/cbc-api/v2/profile',
- None, headers=headers)
+ None, headers=headers, expected_status=426)
return resp['claimsToken']
def _get_claims_token_expiry(self):
@@ -307,13 +447,13 @@ def claims_token_valid(self):
def _get_claims_token(self, email, password):
if not self.claims_token_valid():
self._claims_token = self._new_claims_token(email, password)
- self._downloader.cache.store(self._NETRC_MACHINE, 'claims_token', self._claims_token)
+ self.cache.store(self._NETRC_MACHINE, 'claims_token', self._claims_token)
return self._claims_token
def _real_initialize(self):
if self.claims_token_valid():
return
- self._claims_token = self._downloader.cache.load(self._NETRC_MACHINE, 'claims_token')
+ self._claims_token = self.cache.load(self._NETRC_MACHINE, 'claims_token')
def _find_secret_formats(self, formats, video_id):
""" Find a valid video url and convert it to the secret variant """
@@ -325,7 +465,7 @@ def _find_secret_formats(self, formats, video_id):
url = re.sub(r'(Manifest\(.*?),format=[\w-]+(.*?\))', r'\1\2', base_url)
secret_xml = self._download_xml(url, video_id, note='Downloading secret XML', fatal=False)
- if not secret_xml:
+ if not isinstance(secret_xml, xml.etree.ElementTree.Element):
return
for child in secret_xml:
@@ -340,7 +480,8 @@ def _find_secret_formats(self, formats, video_id):
yield {
**base_format,
'format_id': join_nonempty('sec', height),
- 'url': re.sub(r'(QualityLevels\()\d+(\))', fr'\<1>{bitrate}\2', base_url),
+ # Note: \g<1> is necessary instead of \1 since bitrate is a number
+ 'url': re.sub(r'(QualityLevels\()\d+(\))', fr'\g<1>{bitrate}\2', base_url),
'width': int_or_none(video_quality.attrib.get('MaxWidth')),
'tbr': bitrate / 1000.0,
'height': height,
@@ -348,7 +489,9 @@ def _find_secret_formats(self, formats, video_id):
def _real_extract(self, url):
video_id = self._match_id(url)
- video_info = self._download_json('https://services.radio-canada.ca/ott/cbc-api/v2/assets/' + video_id, video_id)
+ video_info = self._download_json(
+ f'https://services.radio-canada.ca/ott/cbc-api/v2/assets/{video_id}',
+ video_id, expected_status=426)
email, password = self._get_login_info()
if email and password:
@@ -382,8 +525,6 @@ def _real_extract(self, url):
if 'descriptive' in format['format_id'].lower():
format['preference'] = -2
- self._sort_formats(formats)
-
return {
'id': video_id,
'title': video_info['title'],
@@ -405,7 +546,7 @@ def _real_extract(self, url):
class CBCGemPlaylistIE(InfoExtractor):
IE_NAME = 'gem.cbc.ca:playlist'
- _VALID_URL = r'https?://gem\.cbc\.ca/media/(?P(?P[0-9a-z-]+)/s(?P[0-9]+))/?(?:[?#]|$)'
+ _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P(?P[0-9a-z-]+)/s(?P[0-9]+))/?(?:[?#]|$)'
_TESTS = [{
# TV show playlist, all public videos
'url': 'https://gem.cbc.ca/media/schitts-creek/s06',
@@ -414,7 +555,14 @@ class CBCGemPlaylistIE(InfoExtractor):
'id': 'schitts-creek/s06',
'title': 'Season 6',
'description': 'md5:6a92104a56cbeb5818cc47884d4326a2',
+ 'series': 'Schitt\'s Creek',
+ 'season_number': 6,
+ 'season': 'Season 6',
+ 'thumbnail': 'https://images.radio-canada.ca/v1/synps-cbc/season/perso/cbc_schitts_creek_season_06_carousel_v03.jpg?impolicy=ott&im=Resize=(_Size_)&quality=75',
},
+ }, {
+ 'url': 'https://gem.cbc.ca/schitts-creek/s06',
+ 'only_matching': True,
}]
_API_BASE = 'https://services.radio-canada.ca/ott/cbc-api/v2/shows/'
@@ -422,7 +570,7 @@ def _real_extract(self, url):
match = self._match_valid_url(url)
season_id = match.group('id')
show = match.group('show')
- show_info = self._download_json(self._API_BASE + show, season_id)
+ show_info = self._download_json(self._API_BASE + show, season_id, expected_status=426)
season = int(match.group('season'))
season_info = next((s for s in show_info['seasons'] if s.get('season') == season), None)
@@ -474,49 +622,90 @@ def _real_extract(self, url):
class CBCGemLiveIE(InfoExtractor):
IE_NAME = 'gem.cbc.ca:live'
- _VALID_URL = r'https?://gem\.cbc\.ca/live/(?P\d+)'
- _TEST = {
- 'url': 'https://gem.cbc.ca/live/920604739687',
- 'info_dict': {
- 'title': 'Ottawa',
- 'description': 'The live TV channel and local programming from Ottawa',
- 'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/CBC_OTT_VMS/Live_Channel_Static_Images/Ottawa_2880x1620.jpg',
- 'is_live': True,
- 'id': 'AyqZwxRqh8EH',
- 'ext': 'mp4',
- 'timestamp': 1492106160,
- 'upload_date': '20170413',
- 'uploader': 'CBCC-NEW',
+ _VALID_URL = r'https?://gem\.cbc\.ca/live(?:-event)?/(?P\d+)'
+ _TESTS = [
+ {
+ 'url': 'https://gem.cbc.ca/live/920604739687',
+ 'info_dict': {
+ 'title': 'Ottawa',
+ 'description': 'The live TV channel and local programming from Ottawa',
+ 'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/CBC_OTT_VMS/Live_Channel_Static_Images/Ottawa_2880x1620.jpg',
+ 'is_live': True,
+ 'id': 'AyqZwxRqh8EH',
+ 'ext': 'mp4',
+ 'timestamp': 1492106160,
+ 'upload_date': '20170413',
+ 'uploader': 'CBCC-NEW',
+ },
+ 'skip': 'Live might have ended',
},
- 'skip': 'Live might have ended',
- }
-
- # It's unclear where the chars at the end come from, but they appear to be
- # constant. Might need updating in the future.
- # There are two URLs, some livestreams are in one, and some
- # in the other. The JSON schema is the same for both.
- _API_URLS = ['https://tpfeed.cbc.ca/f/ExhSPC/t_t3UKJR6MAT', 'https://tpfeed.cbc.ca/f/ExhSPC/FNiv9xQx_BnT']
+ {
+ 'url': 'https://gem.cbc.ca/live/44',
+ 'info_dict': {
+ 'id': '44',
+ 'ext': 'mp4',
+ 'is_live': True,
+ 'title': r're:^Ottawa [0-9\-: ]+',
+ 'description': 'The live TV channel and local programming from Ottawa',
+ 'live_status': 'is_live',
+ 'thumbnail': r're:https://images.gem.cbc.ca/v1/cbc-gem/live/.*'
+ },
+ 'params': {'skip_download': True},
+ 'skip': 'Live might have ended',
+ },
+ {
+ 'url': 'https://gem.cbc.ca/live-event/10835',
+ 'info_dict': {
+ 'id': '10835',
+ 'ext': 'mp4',
+ 'is_live': True,
+ 'title': r're:^The National \| Bidenâs trip wraps up, Paltrow testifies, Bird flu [0-9\-: ]+',
+ 'description': 'March 24, 2023 | President Bidenâs Ottawa visit ends with big pledges from both countries. Plus, Gwyneth Paltrow testifies in her ski collision trial.',
+ 'live_status': 'is_live',
+ 'thumbnail': r're:https://images.gem.cbc.ca/v1/cbc-gem/live/.*',
+ 'timestamp': 1679706000,
+ 'upload_date': '20230325',
+ },
+ 'params': {'skip_download': True},
+ 'skip': 'Live might have ended',
+ }
+ ]
def _real_extract(self, url):
video_id = self._match_id(url)
+ webpage = self._download_webpage(url, video_id)
+ video_info = self._search_nextjs_data(webpage, video_id)['props']['pageProps']['data']
- for api_url in self._API_URLS:
- video_info = next((
- stream for stream in self._download_json(api_url, video_id)['entries']
- if stream.get('guid') == video_id), None)
- if video_info:
- break
- else:
+ # Two types of metadata JSON
+ if not video_info.get('formattedIdMedia'):
+ video_info = traverse_obj(
+ video_info, (('freeTv', ('streams', ...)), 'items', lambda _, v: v['key'] == video_id, {dict}),
+ get_all=False, default={})
+
+ video_stream_id = video_info.get('formattedIdMedia')
+ if not video_stream_id:
raise ExtractorError('Couldn\'t find video metadata, maybe this livestream is now offline', expected=True)
+ stream_data = self._download_json(
+ 'https://services.radio-canada.ca/media/validation/v2/', video_id, query={
+ 'appCode': 'mpx',
+ 'connectionType': 'hd',
+ 'deviceType': 'ipad',
+ 'idMedia': video_stream_id,
+ 'multibitrate': 'true',
+ 'output': 'json',
+ 'tech': 'hls',
+ 'manifestType': 'desktop',
+ })
+
return {
- '_type': 'url_transparent',
- 'ie_key': 'ThePlatform',
- 'url': video_info['content'][0]['url'],
'id': video_id,
- 'title': video_info.get('title'),
- 'description': video_info.get('description'),
- 'tags': try_get(video_info, lambda x: x['keywords'].split(', ')),
- 'thumbnail': video_info.get('cbc$staticImage'),
+ 'formats': self._extract_m3u8_formats(stream_data['url'], video_id, 'mp4', live=True),
'is_live': True,
+ **traverse_obj(video_info, {
+ 'title': 'title',
+ 'description': 'description',
+ 'thumbnail': ('images', 'card', 'url'),
+ 'timestamp': ('airDate', {parse_iso8601}),
+ })
}