mirror of
https://github.com/ytdl-org/youtube-dl.git
synced 2024-11-19 20:10:25 +00:00
Support Series page
This commit is contained in:
parent
8f6a09b921
commit
65712d99c4
@ -9,7 +9,9 @@ import sys
|
||||
from .common import InfoExtractor
|
||||
from ..compat import (
|
||||
compat_b64decode,
|
||||
compat_parse_qs,
|
||||
compat_struct_unpack,
|
||||
compat_urllib_parse_urlparse,
|
||||
)
|
||||
from ..utils import (
|
||||
determine_ext,
|
||||
@ -25,8 +27,8 @@ _bytes_to_chr = (lambda x: x) if sys.version_info[0] == 2 else (lambda x: map(ch
|
||||
|
||||
|
||||
class RTVEALaCartaIE(InfoExtractor):
|
||||
IE_NAME = 'rtve.es:alacarta'
|
||||
IE_DESC = 'RTVE a la carta'
|
||||
IE_NAME = 'rtve.es:play'
|
||||
IE_DESC = 'RTVE Play'
|
||||
_VALID_URL = r'https?://(?:www\.)?rtve\.es/(m/)?((alacarta|playz?)/videos|filmoteca)/[^/]+/[^/]+/(?P<id>\d+)'
|
||||
|
||||
_TESTS = [{
|
||||
@ -90,6 +92,12 @@ class RTVEALaCartaIE(InfoExtractor):
|
||||
}, {
|
||||
'url': 'http://www.rtve.es/filmoteca/no-do/not-1-introduccion-primer-noticiario-espanol/1465256/',
|
||||
'only_matching': True,
|
||||
}, {
|
||||
'url': 'https://www.rtve.es/play/videos/modulos/capitulos/11332/?currentpage=pf_serie',
|
||||
'info_dict': {
|
||||
'id': '11332',
|
||||
},
|
||||
'playlist_mincount': 20,
|
||||
}]
|
||||
|
||||
def _real_initialize(self):
|
||||
@ -164,8 +172,21 @@ class RTVEALaCartaIE(InfoExtractor):
|
||||
self._sort_formats(formats)
|
||||
return formats
|
||||
|
||||
def _extract_playlist(self, url, playlist_id):
|
||||
|
||||
webpage = self._download_webpage(url, playlist_id)
|
||||
|
||||
matches = re.findall(r'''<a\b[^>]*\bhref\s*=\s*["'](%s)''' % (self._VALID_URL, ), webpage)
|
||||
|
||||
return self.playlist_from_matches(matches, playlist_id=playlist_id, getter=lambda x: x[0], ie=self.ie_key())
|
||||
|
||||
def _real_extract(self, url):
|
||||
video_id = self._match_id(url)
|
||||
|
||||
qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
|
||||
if 'pf_serie' == qs.get('currentpage', [None])[-1]:
|
||||
return self._extract_playlist(url, video_id)
|
||||
|
||||
info = self._download_json(
|
||||
'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
|
||||
video_id)['page']['items'][0]
|
||||
|
Loading…
Reference in New Issue
Block a user