Compare commits

...

4 Commits

Author SHA1 Message Date
Mozi
4ac63203e6
Merge 0a08a50f10 into f2a4983df7 2024-11-13 15:07:56 +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
Mozi
0a08a50f10 [ie/matchtv:video] Add extractor 2024-09-04 17:28:11 +00:00
4 changed files with 155 additions and 7 deletions

View File

@ -1091,7 +1091,11 @@ from .markiza import (
) )
from .massengeschmacktv import MassengeschmackTVIE from .massengeschmacktv import MassengeschmackTVIE
from .masters import MastersIE from .masters import MastersIE
from .matchtv import MatchTVIE from .matchtv import (
MatchTVFeedIE,
MatchTVIE,
MatchTVVideoIE,
)
from .mbn import MBNIE from .mbn import MBNIE
from .mdr import MDRIE from .mdr import MDRIE
from .medaltv import MedalTVIE from .medaltv import MedalTVIE

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 @staticmethod
@ -335,7 +355,7 @@ class ArchiveOrgIE(InfoExtractor):
info['comments'].append({ info['comments'].append({
'id': review.get('review_id'), 'id': review.get('review_id'),
'author': review.get('reviewer'), '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')), 'timestamp': unified_timestamp(review.get('createdate')),
'parent': 'root'}) 'parent': 'root'})

View File

@ -563,13 +563,13 @@ class FacebookIE(InfoExtractor):
return extract_video_data(try_get( return extract_video_data(try_get(
js_data, lambda x: x['jsmods']['instances'], list) or []) 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( 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: if dash_manifest:
formats.extend(self._parse_mpd_formats( formats.extend(self._parse_mpd_formats(
compat_etree_fromstring(urllib.parse.unquote_plus(dash_manifest)), 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): def process_formats(info):
# Downloads with browser's User-Agent are rate limited. Working around # Downloads with browser's User-Agent are rate limited. Working around
@ -619,9 +619,12 @@ class FacebookIE(InfoExtractor):
video = video['creation_story'] video = video['creation_story']
video['owner'] = traverse_obj(video, ('short_form_video_context', 'video_owner')) video['owner'] = traverse_obj(video, ('short_form_video_context', 'video_owner'))
video.update(reel_info) video.update(reel_info)
fmt_data = traverse_obj(video, ('videoDeliveryLegacyFields', {dict})) or video
formats = [] formats = []
q = qualities(['sd', 'hd']) 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'), for key, format_id in (('playable_url', 'sd'), ('playable_url_quality_hd', 'hd'),
('playable_url_dash', ''), ('browser_native_hd_url', 'hd'), ('playable_url_dash', ''), ('browser_native_hd_url', 'hd'),
('browser_native_sd_url', 'sd')): ('browser_native_sd_url', 'sd')):
@ -629,7 +632,7 @@ class FacebookIE(InfoExtractor):
if not playable_url: if not playable_url:
continue continue
if determine_ext(playable_url) == 'mpd': 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: else:
formats.append({ formats.append({
'format_id': format_id, 'format_id': format_id,
@ -638,6 +641,28 @@ class FacebookIE(InfoExtractor):
'url': playable_url, 'url': playable_url,
}) })
extract_dash_manifest(fmt_data, formats) 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: if not formats:
# Do not append false positive entry w/o any formats # Do not append false positive entry w/o any formats
return return

View File

@ -1,4 +1,11 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import (
determine_ext,
extract_attributes,
join_nonempty,
xpath_text,
)
from ..utils.traversal import traverse_obj
class MatchTVIE(InfoExtractor): class MatchTVIE(InfoExtractor):
@ -33,3 +40,95 @@ class MatchTVIE(InfoExtractor):
'is_live': True, 'is_live': True,
'formats': self._extract_m3u8_formats(video_url, video_id, 'mp4', live=True), 'formats': self._extract_m3u8_formats(video_url, video_id, 'mp4', live=True),
} }
# WebcasterIE
class MatchTVVideoIE(InfoExtractor):
_GEO_COUNTRIES = ['RU']
_VALID_URL = r'https?://[.\w-]+/(?:quote|media)/start/free_(?P<id>[^/]+)'
_TESTS = []
def _real_extract(self, url):
video_id = self._match_id(url)
video = self._download_xml(url, video_id)
title = xpath_text(video, './/event_name', 'event name', fatal=True)
formats = []
for format_id in (None, 'noise'):
track_tag = join_nonempty('track', format_id, delim='_')
for track in video.findall(f'.//iphone/{track_tag}'):
track_url = track.text
if not track_url:
continue
if determine_ext(track_url) == 'm3u8':
m3u8_formats = self._extract_m3u8_formats(
track_url, video_id, 'mp4',
entry_protocol='m3u8_native',
m3u8_id=join_nonempty('hls', format_id, delim='-'), fatal=False)
for f in m3u8_formats:
f.update({
'source_preference': 0 if format_id == 'noise' else 1,
'format_note': track.get('title'),
})
formats.extend(m3u8_formats)
thumbnail = xpath_text(video, './/image', 'thumbnail')
return {
'id': video_id,
'title': title,
'thumbnail': thumbnail,
'formats': formats,
}
# WebcasterFeedIE
class MatchTVFeedIE(InfoExtractor):
_GEO_COUNTRIES = ['RU']
_VALID_URL = r'https?://[.\w-]+/feed/start/free_(?P<id>[^/]+)'
_EMBED_REGEX = [r'<(?:object|a|span[^>]+class=["\']webcaster-player["\'])[^>]+data(?:-config)?=(["\']).*?config=(?P<url>https?://(?:(?!\1).)+)\1']
_TESTS = []
_WEBPAGE_TESTS = [{
'url': 'https://matchtv.ru/football/matchtvvideo_NI1593368_clip_Zolotoj_dubl_Cherchesova_Specialnyj_reportazh',
'info_dict': {
'id': '675ea0e4b4b1d54d21f9b52db6624199',
'ext': 'mp4',
'title': '«Золотой дубль Черчесова». Специальный репортаж',
'thumbnail': r're:https?://[\w-]+.video.matchtv.ru/fc/[\w-]+/thumbnails/events/920749/135154185.jpg',
},
}, {
'url': 'https://matchtv.ru/football/rossija/kubok_rossii/matchtvvideo_NI2100168_translation_FONBET_Kubok_Rossii_Tekstilshhik___Spartak_Kostroma',
'info_dict': {
'id': 'b6570efa80dc28df18523237d3f14a5b',
'ext': 'mp4',
'title': 'FONBET Кубок России по футболу сезона 2024 - 2025 гг. Текстильщик - Спартак Кострома',
'thumbnail': r're:https?://[\w-]+.video.matchtv.ru/fc/[\w-]+/thumbnails/events/1202122/1039728778.jpg',
},
}, {
'url': 'https://matchtv.ru/biathlon/matchtvvideo_NI1938496_translation_Letnij_biatlon_Alfa_Bank_Kubok_Sodruzhestva_Sprint_Muzhchiny',
'info_dict': {
'id': '20975a4cd84acdb55a0b5521277d0402',
'ext': 'mp4',
'title': 'Летний биатлон. Альфа-Банк Кубок Содружества. Спринт. Мужчины',
'thumbnail': r're:https?://[\w-]+.video.matchtv.ru/fc/[\w-]+/thumbnails/events/1101266/590556538.jpg',
},
}]
def _extract_from_webpage(self, url, webpage):
yield from super()._extract_from_webpage(url, webpage)
yield from traverse_obj(self._yield_json_ld(webpage, None), (
lambda _, v: v['@type'] == 'VideoObject', 'url',
{extract_attributes}, 'src', {self.url_result}))
def _real_extract(self, url):
video_id = self._match_id(url)
feed = self._download_xml(url, video_id)
video_url = xpath_text(
feed, ('video_hd', 'video'), 'video url', fatal=True)
return self.url_result(video_url)