From 94c4abce7fd55d076e9d9529f92696fe25ef2d17 Mon Sep 17 00:00:00 2001 From: "Sergey M." Date: Thu, 6 Feb 2014 21:16:41 +0700 Subject: [PATCH] [nfb] Add support for nfb.ca (Closes #2069) --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/nfb.py | 76 ++++++++++++++++++++++++++++++++ 2 files changed, 77 insertions(+) create mode 100644 youtube_dl/extractor/nfb.py diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 4e0501ec3..7a97c3279 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -144,6 +144,7 @@ from .nba import NBAIE from .nbc import NBCNewsIE from .ndtv import NDTVIE from .newgrounds import NewgroundsIE +from .nfb import NFBIE from .nhl import NHLIE, NHLVideocenterIE from .niconico import NiconicoIE from .ninegag import NineGagIE diff --git a/youtube_dl/extractor/nfb.py b/youtube_dl/extractor/nfb.py new file mode 100644 index 000000000..722bd8c2e --- /dev/null +++ b/youtube_dl/extractor/nfb.py @@ -0,0 +1,76 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_request, + compat_urllib_parse, +) + + +class NFBIE(InfoExtractor): + IE_NAME = 'nfb' + IE_DESC = 'National Film Board of Canada' + _VALID_URL = r'https?://(?:www\.)?nfb\.ca/film/(?P[\da-z_-]+)' + + _TEST = { + 'url': 'https://www.nfb.ca/film/qallunaat_why_white_people_are_funny', + 'info_dict': { + 'id': 'qallunaat_why_white_people_are_funny', + 'ext': 'mp4', + 'title': 'Qallunaat! Why White People Are Funny ', + 'description': 'md5:836d8aff55e087d04d9f6df554d4e038', + 'duration': 3128, + 'uploader': 'Mark Sandiford', + 'uploader_id': 'mark-sandiford', + }, + 'params': { + # rtmp download + 'skip_download': True, + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + page = self._download_webpage(url, video_id, 'Downloading film page') + + uploader_id = self._html_search_regex(r'([^<]+)', + page, 'director name', fatal=False) + + request = compat_urllib_request.Request('https://www.nfb.ca/film/%s/player_config' % video_id, + compat_urllib_parse.urlencode({'getConfig': 'true'})) + request.add_header('Content-Type', 'application/x-www-form-urlencoded') + request.add_header('X-NFB-Referer', 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf') + + config = self._download_xml(request, video_id, 'Downloading player config XML') + + thumbnail = config.find("./player/stream/media[@type='posterImage']/assets/asset[@quality='high']/default/url").text + video = config.find("./player/stream/media[@type='video']") + duration = int(video.get('duration')) + title = video.find('title').text + description = video.find('description').text + + # It seems assets always go from lower to better quality, so no need to sort + formats = [{ + 'url': x.find('default/streamerURI').text + '/', + 'play_path': x.find('default/url').text, + 'rtmp_live': False, + 'ext': 'mp4', + 'format_id': x.get('quality'), + } for x in video.findall('assets/asset')] + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'formats': formats, + } \ No newline at end of file