diff options
Diffstat (limited to 'youtube_dl')
| -rw-r--r-- | youtube_dl/extractor/pbs.py | 73 |
1 files changed, 54 insertions, 19 deletions
diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index 25f019231..7444b7b5b 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -1,3 +1,5 @@ +from __future__ import unicode_literals + import re import json @@ -5,30 +7,63 @@ from .common import InfoExtractor class PBSIE(InfoExtractor): - _VALID_URL = r'https?://video\.pbs\.org/video/(?P<id>\d+)/?' + _VALID_URL = r'''(?x)https?:// + (?: + # Direct video URL + video\.pbs\.org/video/(?P<id>[0-9]+)/? | + # Article with embedded player + (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+)/?(?:$|[?\#]) | + # Player + video\.pbs\.org/partnerplayer/(?P<player_id>[^/]+)/ + ) + ''' _TEST = { - u'url': u'http://video.pbs.org/video/2365006249/', - u'file': u'2365006249.mp4', - u'md5': 'ce1888486f0908d555a8093cac9a7362', - u'info_dict': { - u'title': u'A More Perfect Union', - u'description': u'md5:ba0c207295339c8d6eced00b7c363c6a', - u'duration': 3190, + 'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/', + 'md5': 'ce1888486f0908d555a8093cac9a7362', + 'info_dict': { + 'id': '2365006249', + 'ext': 'mp4', + 'title': 'A More Perfect Union', + 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a', + 'duration': 3190, }, } def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + + presumptive_id = mobj.group('presumptive_id') + display_id = presumptive_id + if presumptive_id: + webpage = self._download_webpage(url, display_id) + url = self._search_regex( + r'<iframe\s+id=["\']partnerPlayer["\'].*?\s+src=["\'](.*?)["\']>', + webpage, 'player URL') + mobj = re.match(self._VALID_URL, url) + + player_id = mobj.group('player_id') + if not display_id: + display_id = player_id + if player_id: + player_page = self._download_webpage( + url, display_id, note='Downloading player page', + errnote='Could not download player page') + video_id = self._search_regex( + r'<div\s+id="video_([0-9]+)"', player_page, 'video ID') + else: + video_id = mobj.group('id') + display_id = video_id + info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id - info_page = self._download_webpage(info_url, video_id) - info =json.loads(info_page) - return {'id': video_id, - 'title': info['title'], - 'url': info['alternate_encoding']['url'], - 'ext': 'mp4', - 'description': info['program'].get('description'), - 'thumbnail': info.get('image_url'), - 'duration': info.get('duration'), - } + info = self._download_json(info_url, display_id) + + return { + 'id': video_id, + 'title': info['title'], + 'url': info['alternate_encoding']['url'], + 'ext': 'mp4', + 'description': info['program'].get('description'), + 'thumbnail': info.get('image_url'), + 'duration': info.get('duration'), + } |