[condenast] improve embed support

This commit is contained in:
Remita Amine 2017-05-12 16:17:32 +01:00
parent 116283ff64
commit 6c52477f59
2 changed files with 54 additions and 43 deletions

View File

@ -16,7 +16,6 @@
mimetype2ext, mimetype2ext,
orderedSet, orderedSet,
parse_iso8601, parse_iso8601,
remove_end,
) )
@ -50,10 +49,17 @@ class CondeNastIE(InfoExtractor):
'wmagazine': 'W Magazine', 'wmagazine': 'W Magazine',
} }
_VALID_URL = r'https?://(?:video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed(?:js)?)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys()) _VALID_URL = r'''(?x)https?://(?:video|www|player(?:-backend)?)\.(?:%s)\.com/
(?:
(?:
embed(?:js)?|
(?:script|inline)/video
)/(?P<id>[0-9a-f]{24})(?:/(?P<player_id>[0-9a-f]{24}))?(?:.+?\btarget=(?P<target>[^&]+))?|
(?P<type>watch|series|video)/(?P<display_id>[^/?#]+)
)''' % '|'.join(_SITES.keys())
IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values())) IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed(?:js)?)/.+?' % '|'.join(_SITES.keys()) EMBED_URL = r'(?:https?:)?//player(?:-backend)?\.(?:%s)\.com/(?:embed(?:js)?|(?:script|inline)/video)/.+?' % '|'.join(_SITES.keys())
_TESTS = [{ _TESTS = [{
'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led', 'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
@ -89,6 +95,12 @@ class CondeNastIE(InfoExtractor):
'upload_date': '20150916', 'upload_date': '20150916',
'timestamp': 1442434955, 'timestamp': 1442434955,
} }
}, {
'url': 'https://player.cnevids.com/inline/video/59138decb57ac36b83000005.js?target=js-cne-player',
'only_matching': True,
}, {
'url': 'http://player-backend.cnevids.com/script/video/59138decb57ac36b83000005.js',
'only_matching': True,
}] }]
def _extract_series(self, url, webpage): def _extract_series(self, url, webpage):
@ -104,7 +116,7 @@ def _extract_series(self, url, webpage):
entries = [self.url_result(build_url(path), 'CondeNast') for path in paths] entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
return self.playlist_result(entries, playlist_title=title) return self.playlist_result(entries, playlist_title=title)
def _extract_video(self, webpage, url_type): def _extract_video_params(self, webpage):
query = {} query = {}
params = self._search_regex( params = self._search_regex(
r'(?s)var params = {(.+?)}[;,]', webpage, 'player params', default=None) r'(?s)var params = {(.+?)}[;,]', webpage, 'player params', default=None)
@ -123,17 +135,30 @@ def _extract_video(self, webpage, url_type):
'playerId': params['data-player'], 'playerId': params['data-player'],
'target': params['id'], 'target': params['id'],
}) })
video_id = query['videoId'] return query
def _extract_video(self, params):
video_id = params['videoId']
video_info = None video_info = None
if params.get('playerId'):
info_page = self._download_json( info_page = self._download_json(
'http://player.cnevids.com/player/video.js', 'http://player.cnevids.com/player/video.js',
video_id, 'Downloading video info', fatal=False, query=query) video_id, 'Downloading video info', fatal=False, query=params)
if info_page: if info_page:
video_info = info_page.get('video') video_info = info_page.get('video')
if not video_info: if not video_info:
info_page = self._download_webpage( info_page = self._download_webpage(
'http://player.cnevids.com/player/loader.js', 'http://player.cnevids.com/player/loader.js',
video_id, 'Downloading loader info', query=query) video_id, 'Downloading loader info', query=params)
else:
info_page = self._download_webpage(
'https://player.cnevids.com/inline/video/%s.js' % video_id,
video_id, 'Downloading inline info', query={
'target': params.get('target', 'embedplayer')
})
if not video_info:
video_info = self._parse_json( video_info = self._parse_json(
self._search_regex( self._search_regex(
r'(?s)var\s+config\s*=\s*({.+?});', info_page, 'config'), r'(?s)var\s+config\s*=\s*({.+?});', info_page, 'config'),
@ -161,9 +186,7 @@ def _extract_video(self, webpage, url_type):
}) })
self._sort_formats(formats) self._sort_formats(formats)
info = self._search_json_ld( return {
webpage, video_id, fatal=False) if url_type != 'embed' else {}
info.update({
'id': video_id, 'id': video_id,
'formats': formats, 'formats': formats,
'title': title, 'title': title,
@ -174,22 +197,26 @@ def _extract_video(self, webpage, url_type):
'series': video_info.get('series_title'), 'series': video_info.get('series_title'),
'season': video_info.get('season_title'), 'season': video_info.get('season_title'),
'timestamp': parse_iso8601(video_info.get('premiere_date')), 'timestamp': parse_iso8601(video_info.get('premiere_date')),
}) 'categories': video_info.get('categories'),
return info }
def _real_extract(self, url): def _real_extract(self, url):
site, url_type, item_id = re.match(self._VALID_URL, url).groups() video_id, player_id, target, url_type, display_id = re.match(self._VALID_URL, url).groups()
# Convert JS embed to regular embed if video_id:
if url_type == 'embedjs': return self._extract_video({
parsed_url = compat_urlparse.urlparse(url) 'videoId': video_id,
url = compat_urlparse.urlunparse(parsed_url._replace( 'playerId': player_id,
path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/'))) 'target': target,
url_type = 'embed' })
webpage = self._download_webpage(url, item_id) webpage = self._download_webpage(url, display_id)
if url_type == 'series': if url_type == 'series':
return self._extract_series(url, webpage) return self._extract_series(url, webpage)
else: else:
return self._extract_video(webpage, url_type) params = self._extract_video_params(webpage)
info = self._search_json_ld(
webpage, display_id, fatal=False)
info.update(self._extract_video(params))
return info

View File

@ -2146,22 +2146,6 @@ def _real_extract(self, url):
if svt_url: if svt_url:
return self.url_result(svt_url, 'SVT') return self.url_result(svt_url, 'SVT')
# Look for embedded condenast player
matches = re.findall(
r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
webpage)
if matches:
return {
'_type': 'playlist',
'entries': [{
'_type': 'url',
'ie_key': 'CondeNast',
'url': ma,
} for ma in matches],
'title': video_title,
'id': video_id,
}
# Look for Bandcamp pages with custom domain # Look for Bandcamp pages with custom domain
mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage) mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
if mobj is not None: if mobj is not None: