artem/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2026-04-23 00:56:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix extractor harder

Browse Source 
Authored by: bashonly
This commit is contained in:
bashonly 2024-11-17 14:58:33 -06:00
parent b91afadd1a
commit d470a4c4fb
No known key found for this signature in database
GPG Key ID: 783F096F253D15B0
1 changed files with 28 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
yt_dlp/extractor/ctvnews.py
View File

@ -1,5 +1,6 @@
import json import json
import re import re
import urllib.parse
from .common import InfoExtractor from .common import InfoExtractor
from .ninecninemedia import NineCNineMediaIE from .ninecninemedia import NineCNineMediaIE
@ -15,7 +16,8 @@ class CTVNewsIE(InfoExtractor):
rf'{_BASE_REGEX}video/c{_VIDEO_ID_RE}', rf'{_BASE_REGEX}video/c{_VIDEO_ID_RE}',
rf'{_BASE_REGEX}video(?:-gallery)?/?\?clipId={_VIDEO_ID_RE}', rf'{_BASE_REGEX}video(?:-gallery)?/?\?clipId={_VIDEO_ID_RE}',
rf'{_BASE_REGEX}video/?\?(?:playlist|bin)Id={_PLAYLIST_ID_RE}', rf'{_BASE_REGEX}video/?\?(?:playlist|bin)Id={_PLAYLIST_ID_RE}',
rf'{_BASE_REGEX}(?!video/)[^?#]*?{_PLAYLIST_ID_RE}', rf'{_BASE_REGEX}(?!video/)[^?#]*?{_PLAYLIST_ID_RE}/?(?:$|[?#])',
rf'{_BASE_REGEX}(?!video/)[^?#]+\?binId={_PLAYLIST_ID_RE}',
] ]
_TESTS = [{ _TESTS = [{
'url': 'http://www.ctvnews.ca/video?clipId=901995', 'url': 'http://www.ctvnews.ca/video?clipId=901995',
@ -114,6 +116,27 @@ class CTVNewsIE(InfoExtractor):
'tags': [], 'tags': [],
'upload_date': '20150401', 'upload_date': '20150401',
}, },
}, {
'url': 'https://ottawa.ctvnews.ca/features/regional-contact/regional-contact-archive?binId=1.1164587#3023759',
'md5': 'a14c0603557decc6531260791c23cc5e',
'info_dict': {
'id': '3023759',
'ext': 'flv',
'season_number': 2024,
'timestamp': 1731798000,
'season': '2024',
'episode': 'Episode 125',
'description': 'CTV News Ottawa at Six',
'duration': 2712.076,
'episode_number': 125,
'upload_date': '20241116',
'title': 'CTV News Ottawa at Six for Saturday, November 16, 2024',
'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
'categories': [],
'tags': [],
'series': 'CTV News Ottawa at Six',
'season_id': '92667',
},
}, { }, {
'url': 'http://www.ctvnews.ca/1.810401', 'url': 'http://www.ctvnews.ca/1.810401',
'only_matching': True, 'only_matching': True,
@ -131,7 +154,10 @@ class CTVNewsIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
page_id = self._match_id(url) page_id = self._match_id(url)
if page_id.isdecimal(): if mobj := re.fullmatch(self._VIDEO_ID_RE, urllib.parse.urlparse(url).fragment):
page_id = mobj.group('id')
if re.fullmatch(self._VIDEO_ID_RE, page_id):
return self._ninecninemedia_url_result(page_id) return self._ninecninemedia_url_result(page_id)
webpage = self._download_webpage(f'https://www.ctvnews.ca/{page_id}', page_id, query={ webpage = self._download_webpage(f'https://www.ctvnews.ca/{page_id}', page_id, query={