Compare commits

...

11 Commits

Author SHA1 Message Date
kclauhk
8e905c2656
Merge 48416c3718 into f2a4983df7 2024-11-13 01:47:05 +01:00
Jackson Humphrey
f2a4983df7
[ie/archive.org] Fix comments extraction (#11527)
Closes #11526
Authored by: jshumphrey
2024-11-12 23:26:18 +00:00
bashonly
bacc31b05a
[ie/facebook] Fix formats extraction (#11513)
Closes #11497
Authored by: bashonly
2024-11-12 23:23:10 +00:00
manav_chaudhary
a9f85670d0
[ie/Chaturbate] Support alternate domains (#10595)
Closes #10594
Authored by: manavchaudhary1
2024-11-11 23:41:56 +01:00
kclauhk
48416c3718 remove trailing whitespace 2024-10-31 12:25:11 +08:00
kclauhk
2fd16fdc6e ver contains instead of exact match 2024-10-31 12:12:25 +08:00
kclauhk
245ca4e515 use one ie_key only 2024-10-31 02:42:58 +08:00
kclauhk
0f49100783 correct file ext 2024-09-28 20:39:10 +08:00
kclauhk
05b4719f6c refactor
- avoid repeatedly calling `self._configuration_arg`
- avoid unnecessary looping when 'ver' argument is not provided
- add/change error messages
2024-09-28 17:00:22 +08:00
kclauhk
52f73ffc61 use geo_bypass_country instead and modify _REQUEST_HEADERS 2024-09-21 15:23:56 +08:00
kclauhk
83dedc8369 [ie/extrememusic] Add extractor 2024-09-17 03:29:56 +08:00
5 changed files with 426 additions and 9 deletions

View File

@ -622,6 +622,11 @@ from .europeantour import EuropeanTourIE
from .eurosport import EurosportIE
from .euscreen import EUScreenIE
from .expressen import ExpressenIE
from .extrememusic import (
ExtremeMusicAIE,
ExtremeMusicIE,
ExtremeMusicPIE,
)
from .eyedotv import EyedoTVIE
from .facebook import (
FacebookAdsIE,

View File

@ -205,6 +205,26 @@ class ArchiveOrgIE(InfoExtractor):
},
},
],
}, {
# The reviewbody is None for one of the reviews; just need to extract data without crashing
'url': 'https://archive.org/details/gd95-04-02.sbd.11622.sbeok.shnf/gd95-04-02d1t04.shn',
'info_dict': {
'id': 'gd95-04-02.sbd.11622.sbeok.shnf/gd95-04-02d1t04.shn',
'ext': 'mp3',
'title': 'Stuck Inside of Mobile with the Memphis Blues Again',
'creators': ['Grateful Dead'],
'duration': 338.31,
'track': 'Stuck Inside of Mobile with the Memphis Blues Again',
'description': 'md5:764348a470b986f1217ffd38d6ac7b72',
'display_id': 'gd95-04-02d1t04.shn',
'location': 'Pyramid Arena',
'uploader': 'jon@archive.org',
'album': '1995-04-02 - Pyramid Arena',
'upload_date': '20040519',
'track_number': 4,
'release_date': '19950402',
'timestamp': 1084927901,
},
}]
@staticmethod
@ -335,7 +355,7 @@ class ArchiveOrgIE(InfoExtractor):
info['comments'].append({
'id': review.get('review_id'),
'author': review.get('reviewer'),
'text': str_or_none(review.get('reviewtitle'), '') + '\n\n' + review.get('reviewbody'),
'text': join_nonempty('reviewtitle', 'reviewbody', from_dict=review, delim='\n\n'),
'timestamp': unified_timestamp(review.get('createdate')),
'parent': 'root'})

View File

@ -9,7 +9,7 @@ from ..utils import (
class ChaturbateIE(InfoExtractor):
_VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.com/(?:fullvideo/?\?.*?\bb=)?(?P<id>[^/?&#]+)'
_VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.(?P<tld>com|eu|global)/(?:fullvideo/?\?.*?\bb=)?(?P<id>[^/?&#]+)'
_TESTS = [{
'url': 'https://www.chaturbate.com/siswet19/',
'info_dict': {
@ -29,15 +29,24 @@ class ChaturbateIE(InfoExtractor):
}, {
'url': 'https://en.chaturbate.com/siswet19/',
'only_matching': True,
}, {
'url': 'https://chaturbate.eu/siswet19/',
'only_matching': True,
}, {
'url': 'https://chaturbate.eu/fullvideo/?b=caylin',
'only_matching': True,
}, {
'url': 'https://chaturbate.global/siswet19/',
'only_matching': True,
}]
_ROOM_OFFLINE = 'Room is currently offline'
def _real_extract(self, url):
video_id = self._match_id(url)
video_id, tld = self._match_valid_url(url).group('id', 'tld')
webpage = self._download_webpage(
f'https://chaturbate.com/{video_id}/', video_id,
f'https://chaturbate.{tld}/{video_id}/', video_id,
headers=self.geo_verification_headers())
found_m3u8_urls = []

View File

@ -0,0 +1,358 @@
import itertools
import re
from .common import InfoExtractor
from ..utils import (
determine_ext,
int_or_none,
join_nonempty,
merge_dicts,
str_or_none,
traverse_obj,
unified_strdate,
url_or_none,
)
class ExtremeMusicBaseIE(InfoExtractor):
_API_URL = 'https://snapi.extrememusic.com'
_REQUEST_HEADERS = None
_REQUIRE_VERSION = []
def _initialize(self, url, video_id, country=None):
self._REQUIRE_VERSION = (self._configuration_arg('ver', ie_key='extrememusic')
or self._configuration_arg('version', ie_key='extrememusic'))
# This site serves different versions of the same playlist id due to geo-restriction
# use user's own country code if no code (geo_bypass_country or pre-defined country code) is provided
if not country:
country = self._download_webpage('https://ipapi.co/country_code', video_id)
self.to_screen(f'Set country code to {country}')
env = self._download_json('https://www.extrememusic.com/env', video_id)
self._REQUEST_HEADERS = {
'Accept': 'application/json',
'Origin': 'https://www.extrememusic.com',
'Referer': url,
'Sec-Fetch-Mode': 'cors',
'X-API-Auth': env['token'],
'X-Site-Id': 4,
'X-Viewer-Country': country.upper(),
}
def _get_album_data(self, album_id, video_id, fatal=True):
album = self._download_json(f'{self._API_URL}/albums/{album_id}', video_id, fatal=fatal,
note='Downloading album data', errnote='Unable to download album data',
headers=self._REQUEST_HEADERS) or {}
if video_id == album_id:
bio = self._download_json(f'{self._API_URL}/albums/{album_id}/bio', video_id, fatal=False,
note='Downloading album data', errnote='Unable to download album data',
headers=self._REQUEST_HEADERS) or {}
return merge_dicts(album, bio)
else:
return album
def _extract_track(self, album_data, track_id=None, version_id=None):
if 'tracks' in album_data and 'track_sounds' in album_data:
if not track_id and version_id:
track_id = traverse_obj(album_data['track_sounds'],
(lambda _, v: v['id'] == int(version_id), 'track_id', {int}), get_all=False)
if track := traverse_obj(album_data['tracks'],
(lambda _, v: v['id'] == int(track_id), {dict}), get_all=False):
info = {**traverse_obj(track, {
'track': ('title', {str}),
'track_number': ('sort_order', {lambda v: v + 1}, {int}),
'track_id': ('track_no', {str}),
'description': ('description', {lambda v: str_or_none(v) or None}),
'artists': ('artists', {lambda v: v or traverse_obj(album_data, ('album', 'artist'))},
{lambda v: (v if isinstance(v, list) else [v]) if v else None}),
'composers': ('composers', ..., 'name'),
'genres': (('genre', 'subgenre'), ..., 'label'),
'tag': ('keywords', ..., 'label'),
'album': ('album_title', {lambda v: str_or_none(v) or None}),
}), **traverse_obj(album_data, {
'album_artists': ('album', 'artist', {lambda v: [v] if v else None}),
'upload_date': ('album', 'created', {unified_strdate}),
})}
entries, thumbnails = [], []
for image in traverse_obj(track, ('images', 'default')):
thumbnails.append(traverse_obj(image, {
'url': ('url', {url_or_none}),
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
}))
if not self._REQUIRE_VERSION:
version_id = version_id or traverse_obj(track, 'default_track_sound_id', ('track_sound_ids', 0))
for sound_id in [version_id] if version_id else track['track_sound_ids']:
if sound := traverse_obj(album_data['track_sounds'],
(lambda _, v: v['id'] == int(sound_id) and v['track_id'] == int(track_id),
{dict}), get_all=False):
if (version_id
or 'all' in self._REQUIRE_VERSION
or any(x in sound['version_type'].lower() for x in self._REQUIRE_VERSION)):
formats = []
for audio_url in traverse_obj(sound, ('assets', 'audio', ('preview_url',
'preview_url_hls'))):
if determine_ext(audio_url) == 'm3u8':
m3u8_url = re.sub(r'\.m3u8\?.*', '/HLS/128_v4.m3u8', audio_url)
for f in self._extract_m3u8_formats(m3u8_url, sound_id, 'm4a', fatal=False):
formats.append({
**f,
'vcodec': 'none',
'perference': -2,
})
else:
formats.append({
'url': audio_url,
'vcodec': 'none',
})
entries.append({
'id': str(sound_id),
'title': join_nonempty('title', 'version_type', from_dict=sound, delim=' - '),
'alt_title': sound['version_type'],
**info,
'thumbnails': thumbnails,
'duration': sound.get('duration'),
'formats': formats,
'webpage_url': f"https://www.extrememusic.com/albums/{track['album_id']}?item={track_id}&ver={sound_id}",
})
if len(entries) > 1:
return {
'id': track_id,
**info,
'entries': entries,
'_type': 'playlist',
}
elif len(entries) == 1:
return entries[0]
else:
self.raise_no_formats('Track data not found', video_id=track_id)
return []
class ExtremeMusicIE(ExtremeMusicBaseIE):
_VALID_URL = r'https?://(?:www\.)?extrememusic\.com/albums/(?P<album>\d+)\?(.*item=(?P<id>\d+))?(.*ver=(?P<ver>\d+))?'
_TESTS = [{
'url': 'https://www.extrememusic.com/albums/15875?item=263381&ver=1265009&sharedTrack=dHJ1ZQ==',
'info_dict': {
'id': '1265009',
'ext': 'mp3',
'title': 'FOLLOW - Instrumental',
'alt_title': 'Instrumental',
'track': 'FOLLOW',
'track_number': 5,
'track_id': 'HPE316_05',
'artists': ['PRAERS'],
'composers': ['Joseph Andrew Banfi', 'Thomas Louis James White'],
'genres': ['POP', 'DREAM', 'INDIE'],
'tag': 'count:7',
'album': 'AVALON',
'album_artists': ['PRAERS'],
'upload_date': '20240729',
'thumbnail': 'https://d2oet5a29f64lj.cloudfront.net/img-data/w/2480/album/600/HPE316.jpg',
'duration': 246,
},
}, {
'url': 'https://www.extrememusic.com/albums/15823?ver=1262087',
'info_dict': {
'id': '1262087',
'ext': 'mp3',
'title': 'MAGICAL HIGHWAY - VOCALS',
'alt_title': 'VOCALS',
'track': 'MAGICAL HIGHWAY',
'track_number': 2,
'track_id': 'ASM0002_02',
'description': 'Full version - a fun, happy and upbeat pop track with a medium - fast tempo - electronic, bouncy, bright',
'composers': ['ENB'],
'genres': ['POP', 'ELECTRO', 'JPOP'],
'tag': 'count:8',
'album': 'TOKYO POPPIN\'',
'upload_date': '20240709',
'thumbnail': 'https://d2oet5a29f64lj.cloudfront.net/img-data/w/2480/album/600/ASM0002.jpg',
'duration': 265,
},
}, {
'url': 'https://www.extrememusic.com/albums/15064?item=254704',
'info_dict': {
'id': '1178851',
'ext': 'mp3',
'title': 'SWEET TOOTH - Full Version',
'alt_title': 'Full Version',
'track': 'SWEET TOOTH',
'track_number': 2,
'track_id': 'HPE263_02',
'artists': ['PILOT PAISLEY-ROSE'],
'composers': ['PILOT PAISLEY ROSE SARACENO', 'SAMUEL JAMES BRANDT'],
'genres': ['POP', 'ELECTRO', 'ROCK'],
'tag': 'count:7',
'album': 'ADDICTED',
'album_artists': ['PILOT PAISLEY-ROSE'],
'upload_date': '20230629',
'thumbnail': 'https://d2oet5a29f64lj.cloudfront.net/img-data/w/2480/album/600/HPE263.jpg',
'duration': 161,
},
}, {
'url': 'https://www.extrememusic.com/albums/1315?item=24795',
'info_dict': {
'id': '61003',
'ext': 'mp3',
'title': 'JOY TO THE WORLD (INST) - Instrumental',
'alt_title': 'Instrumental',
'track': 'JOY TO THE WORLD (INST)',
'track_number': 6,
'track_id': 'XEL016_06',
'composers': ['TRADITIONAL'],
'genres': ['HOLIDAY', 'CHRISTMAS'],
'tag': 'count:5',
'album': 'CHRISTMAS SPARKLE',
'upload_date': '20041001',
'thumbnail': 'https://d2oet5a29f64lj.cloudfront.net/img-data/w/2480/album/600/XEL016.jpg',
'duration': 132,
},
}]
def _real_extract(self, url):
album_id, track_id, version_id = self._match_valid_url(url).group('album', 'id', 'ver')
self._initialize(url, version_id or track_id, self.get_param('geo_bypass_country') or 'DE')
album_data = self._get_album_data(album_id, version_id or track_id)
if result := self._extract_track(album_data, track_id, version_id):
return result
else:
self.raise_no_formats('No formats were found')
class ExtremeMusicAIE(ExtremeMusicBaseIE):
IE_NAME = 'ExtremeMusic:album'
_VALID_URL = r'https?://(?:www\.)?extrememusic\.com/albums/(?P<id>\d+)(?!.*(item|ver)=)'
_TESTS = [{
'url': 'https://www.extrememusic.com/albums/6778',
'info_dict': {
'id': '6778',
'album': 'Ethereal Voices',
},
'playlist_count': 11,
}, {
'url': 'https://www.extrememusic.com/albums/15835',
'info_dict': {
'id': '15835',
'album': 'BIGGEST BANG',
'description': 'Minus Aura, a minimalist duo who create deep drama and emotion to put you under their spell.',
'artists': ['MINUS AURA'],
'genres': ['ELECTRONICA', 'POP', 'SYNTH'],
'tag': ['ELECTRONIC', 'STRUGGLE'],
},
'playlist_count': 4,
}]
def _real_extract(self, url):
album_id = self._match_id(url)
self._initialize(url, album_id, self.get_param('geo_bypass_country') or 'DE')
album_data = self._get_album_data(album_id, album_id)
entries = []
for track_id in traverse_obj(album_data, ('tracks', ..., 'id')):
if track := self._extract_track(album_data, track_id=track_id):
if track.get('entries'):
entries.extend(track['entries'])
else:
entries.append(track)
if entries:
subgenres = traverse_obj(album_data, ('album', 'subgenres', {str_or_none}))
return merge_dicts(traverse_obj(album_data.get('album'), {
'id': ('id', {lambda v: str(v)}),
'album': ('title', {str_or_none}),
'description': ('description', {lambda v: str_or_none(v) or None}),
'artists': ('artist', {lambda v: [v] if v else None}),
'genres': ('genres', {str_or_none}, {lambda v: join_nonempty(v, subgenres, delim=', ')},
{lambda v: v.split(', ') if v else None}),
'tag': ('keywords', {lambda v: v.split(', ') if v else None}),
}), {
'description': traverse_obj(album_data, ('bio', 'description', {lambda v: str_or_none(v) or None})),
'entries': entries,
'_type': 'playlist',
})
else:
self.raise_no_formats('No formats were found')
class ExtremeMusicPIE(ExtremeMusicBaseIE):
IE_NAME = 'ExtremeMusic:playlist'
_VALID_URL = r'https?://(?:www\.)?extrememusic\.com/playlists/(?P<id>[^?]+)'
_TESTS = [{
'url': 'https://www.extrememusic.com/playlists/Kf3fAppAKK2UpAUUp7KK1pBDBMrC62c_Kf8UKAAppUUKppK2UAp92K7Appp8xMx',
'info_dict': {
'id': 'Kf3fAppAKK2UpAUUp7KK1pBDBMrC62c_Kf8UKAAppUUKppK2UAp92K7Appp8xMx',
'title': 'NICE',
'thumbnail': 'https://d2oet5a29f64lj.cloudfront.net/img-data/w/2480/featureditem/square/thumbnail_PLAYLIST_Nice-square-(formerly ChristmasTraditional).jpg',
},
'playlist_mincount': 29,
'expected_warnings': ['This playlist has geo-restricted items. Try using --xff to specify a different country code, e.g. DE'],
}, {
'url': 'https://www.extrememusic.com/playlists/fUKKU5KAfK61pAAKp4U4KpKUxsRk2ki_fU117KpUUAAUKAUfpA6UAfAKK8Ul5ji',
'info_dict': {
'id': 'fUKKU5KAfK61pAAKp4U4KpKUxsRk2ki_fU117KpUUAAUKAUfpA6UAfAKK8Ul5ji',
'title': 'NEO CLASSICAL',
'thumbnail': 'https://d2oet5a29f64lj.cloudfront.net/img-data/w/2480/featureditem/square/NeoClassical.jpg',
},
'playlist_mincount': 50,
}]
def _real_extract(self, url):
playlist_id = self._match_id(url)
self._initialize(url, playlist_id, self.get_param('geo_bypass_country'))
def playlist_query(playlist_id, offset, limit):
# playlist api: https://snapi.extrememusic.com/playlists?id={playlist_id}&range={offset}%2C{limit}'
return self._download_json(
'https://snapi.extrememusic.com/playlists', playlist_id,
note=f'Downloading item {offset + 1}-{offset + limit}', query={
'id': playlist_id,
'range': f'{offset},{limit}',
}, headers=self._REQUEST_HEADERS)
thumbnails, entries = [], []
album_data, track_done, limit = {}, [], 50
for i in itertools.count():
playlist = playlist_query(playlist_id, i * limit, limit)
if len(playlist['playlist_items']) == 0:
break
else:
track_ids = traverse_obj(playlist, ('playlist_items', ..., 'track_id'))
for track_id in list(dict.fromkeys(track_ids)):
if track_id not in track_done:
album_id = traverse_obj(playlist,
('tracks', lambda _, v: v['id'] == track_id, 'album_id', {int}), get_all=False)
if album_id not in album_data:
album_data[album_id] = self._get_album_data(album_id, track_id, fatal=False)
playlist['album'] = traverse_obj(album_data, (album_id, 'album', {dict}))
if track := self._extract_track(playlist, track_id=track_id):
if track.get('entries'):
entries.extend(track['entries'])
else:
entries.append(track)
track_done.append(track_id)
if len(track_done) >= playlist['playlist']['playlist_items_count']:
break
if entries:
if len(track_done) < playlist['playlist']['playlist_items_count']:
self.report_warning('This playlist has geo-restricted items. Try using --xff to specify a different country code, e.g. DE')
for image in traverse_obj(playlist['playlist'], ('images', 'square')):
thumbnails.append(traverse_obj(image, {
'url': ('url', {url_or_none}),
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
}))
return {k: v for k, v in {
'id': playlist['playlist']['id'],
'title': playlist['playlist']['title'],
'thumbnail': traverse_obj(thumbnails, (0, 'url', {url_or_none})),
'thumbnails': thumbnails,
'uploader': playlist['playlist']['owner_name'],
'entries': entries,
'_type': 'playlist',
}.items() if v}
else:
self.raise_no_formats('No formats were found')

View File

@ -563,13 +563,13 @@ class FacebookIE(InfoExtractor):
return extract_video_data(try_get(
js_data, lambda x: x['jsmods']['instances'], list) or [])
def extract_dash_manifest(video, formats):
def extract_dash_manifest(vid_data, formats, mpd_url=None):
dash_manifest = traverse_obj(
video, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', expected_type=str)
vid_data, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', 'manifest_xml', expected_type=str)
if dash_manifest:
formats.extend(self._parse_mpd_formats(
compat_etree_fromstring(urllib.parse.unquote_plus(dash_manifest)),
mpd_url=url_or_none(video.get('dash_manifest_url'))))
mpd_url=url_or_none(video.get('dash_manifest_url')) or mpd_url))
def process_formats(info):
# Downloads with browser's User-Agent are rate limited. Working around
@ -619,9 +619,12 @@ class FacebookIE(InfoExtractor):
video = video['creation_story']
video['owner'] = traverse_obj(video, ('short_form_video_context', 'video_owner'))
video.update(reel_info)
fmt_data = traverse_obj(video, ('videoDeliveryLegacyFields', {dict})) or video
formats = []
q = qualities(['sd', 'hd'])
# Legacy formats extraction
fmt_data = traverse_obj(video, ('videoDeliveryLegacyFields', {dict})) or video
for key, format_id in (('playable_url', 'sd'), ('playable_url_quality_hd', 'hd'),
('playable_url_dash', ''), ('browser_native_hd_url', 'hd'),
('browser_native_sd_url', 'sd')):
@ -629,7 +632,7 @@ class FacebookIE(InfoExtractor):
if not playable_url:
continue
if determine_ext(playable_url) == 'mpd':
formats.extend(self._extract_mpd_formats(playable_url, video_id))
formats.extend(self._extract_mpd_formats(playable_url, video_id, fatal=False))
else:
formats.append({
'format_id': format_id,
@ -638,6 +641,28 @@ class FacebookIE(InfoExtractor):
'url': playable_url,
})
extract_dash_manifest(fmt_data, formats)
# New videoDeliveryResponse formats extraction
fmt_data = traverse_obj(video, ('videoDeliveryResponseFragment', 'videoDeliveryResponseResult'))
mpd_urls = traverse_obj(fmt_data, ('dash_manifest_urls', ..., 'manifest_url', {url_or_none}))
dash_manifests = traverse_obj(fmt_data, ('dash_manifests', lambda _, v: v['manifest_xml']))
for idx, dash_manifest in enumerate(dash_manifests):
extract_dash_manifest(dash_manifest, formats, mpd_url=traverse_obj(mpd_urls, idx))
if not dash_manifests:
# Only extract from MPD URLs if the manifests are not already provided
for mpd_url in mpd_urls:
formats.extend(self._extract_mpd_formats(mpd_url, video_id, fatal=False))
for prog_fmt in traverse_obj(fmt_data, ('progressive_urls', lambda _, v: v['progressive_url'])):
format_id = traverse_obj(prog_fmt, ('metadata', 'quality', {str.lower}))
formats.append({
'format_id': format_id,
# sd, hd formats w/o resolution info should be deprioritized below DASH
'quality': q(format_id) - 3,
'url': prog_fmt['progressive_url'],
})
for m3u8_url in traverse_obj(fmt_data, ('hls_playlist_urls', ..., 'hls_playlist_url', {url_or_none})):
formats.extend(self._extract_m3u8_formats(m3u8_url, video_id, 'mp4', fatal=False, m3u8_id='hls'))
if not formats:
# Do not append false positive entry w/o any formats
return