[ie/espn] Add subtitles extraction; fix clip regex in articles

No video is extractable from any "only_matching" url in "ESPNArticleIE".
This commit is contained in:
Mozi 2024-09-05 16:50:54 +00:00
parent e8e6a982a1
commit 0743fbd6e9
1 changed files with 25 additions and 11 deletions

View File

@ -113,6 +113,7 @@ class ESPNIE(OnceIE):
format_urls = set() format_urls = set()
formats = [] formats = []
subtitles = {}
def traverse_source(source, base_source_id=None): def traverse_source(source, base_source_id=None):
for src_id, src_item in source.items(): for src_id, src_item in source.items():
@ -140,9 +141,11 @@ class ESPNIE(OnceIE):
formats.extend(self._extract_f4m_formats( formats.extend(self._extract_f4m_formats(
source_url, video_id, f4m_id=source_id, fatal=False)) source_url, video_id, f4m_id=source_id, fatal=False))
elif ext == 'm3u8': elif ext == 'm3u8':
formats.extend(self._extract_m3u8_formats( m3u8_frmts, m3u8_subs = self._extract_m3u8_formats_and_subtitles(
source_url, video_id, 'mp4', entry_protocol='m3u8_native', source_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id=source_id, fatal=False)) m3u8_id=source_id, fatal=False)
formats.extend(m3u8_frmts)
self._merge_subtitles(m3u8_subs, target=subtitles)
else: else:
f = { f = {
'url': source_url, 'url': source_url,
@ -176,12 +179,26 @@ class ESPNIE(OnceIE):
'timestamp': timestamp, 'timestamp': timestamp,
'duration': duration, 'duration': duration,
'formats': formats, 'formats': formats,
'subtitles': subtitles,
} }
class ESPNArticleIE(InfoExtractor): class ESPNArticleIE(InfoExtractor):
_VALID_URL = r'https?://(?:espn\.go|(?:www\.)?espn)\.com/(?:[^/]+/)*(?P<id>[^/]+)' _VALID_URL = r'https?://(?:espn\.go|(?:www\.)?espn)\.com/(?:[^/]+/)*(?P<id>[^/]+)'
_TESTS = [{ _TESTS = [{
'url': 'https://www.espn.com/college-football/game/_/gameId/401520427',
'info_dict': {
'id': '401520427',
'title': 'Alabama 27-24 Auburn (Nov 25, 2023) Final Score - ESPN',
'description': 'Game summary of the Alabama Crimson Tide vs. Auburn Tigers NCAAF game, final score 27-24, from November 25, 2023 on ESPN.',
'entries': [{
'id': '38979520',
}, {
'id': '38981707',
}],
},
'playlist_count': 2,
}, {
'url': 'http://espn.go.com/nba/recap?gameId=400793786', 'url': 'http://espn.go.com/nba/recap?gameId=400793786',
'only_matching': True, 'only_matching': True,
}, { }, {
@ -200,16 +217,13 @@ class ESPNArticleIE(InfoExtractor):
return False if (ESPNIE.suitable(url) or WatchESPNIE.suitable(url)) else super().suitable(url) return False if (ESPNIE.suitable(url) or WatchESPNIE.suitable(url)) else super().suitable(url)
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) playlist_id = self._match_id(url)
webpage = self._download_webpage(url, playlist_id)
webpage = self._download_webpage(url, video_id) return self.playlist_result(traverse_obj(re.finditer(
r'class=(["\']).*?Media.*?\1[^>]+data-videoid=["\'](?P<id>\d+)', webpage), (..., 'id', {
video_id = self._search_regex( lambda x: self.url_result(f'http://espn.go.com/video/clip?id={x}', ESPNIE.ie_key(), x),
r'class=(["\']).*?video-play-button.*?\1[^>]+data-id=["\'](?P<id>\d+)', })), playlist_id, self._html_extract_title(webpage), self._html_search_meta('description', webpage))
webpage, 'video id', group='id')
return self.url_result(
f'http://espn.go.com/video/clip?id={video_id}', ESPNIE.ie_key())
class FiveThirtyEightIE(InfoExtractor): class FiveThirtyEightIE(InfoExtractor):