From 6140baf4e1aa7015ac6b43e93236ac6a5813569e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Thu, 19 Feb 2015 18:17:31 +0100 Subject: [nationalgeographic] Add extractor (closes #4960) --- youtube_dl/extractor/nationalgeographic.py | 38 ++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) create mode 100644 youtube_dl/extractor/nationalgeographic.py (limited to 'youtube_dl/extractor/nationalgeographic.py') diff --git a/youtube_dl/extractor/nationalgeographic.py b/youtube_dl/extractor/nationalgeographic.py new file mode 100644 index 0000000..c18640c --- /dev/null +++ b/youtube_dl/extractor/nationalgeographic.py @@ -0,0 +1,38 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + smuggle_url, + url_basename, +) + + +class NationalGeographicIE(InfoExtractor): + _VALID_URL = r'http://video\.nationalgeographic\.com/video/.*?' + + _TEST = { + 'url': 'http://video.nationalgeographic.com/video/news/150210-news-crab-mating-vin?source=featuredvideo', + 'info_dict': { + 'id': '4DmDACA6Qtk_', + 'ext': 'flv', + 'title': 'Mating Crabs Busted by Sharks', + 'description': 'md5:16f25aeffdeba55aaa8ec37e093ad8b3', + }, + 'add_ie': ['ThePlatform'], + } + + def _real_extract(self, url): + name = url_basename(url) + + webpage = self._download_webpage(url, name) + feed_url = self._search_regex(r'data-feed-url="([^"]+)"', webpage, 'feed url') + guid = self._search_regex(r'data-video-guid="([^"]+)"', webpage, 'guid') + + feed = self._download_xml('%s?byGuid=%s' % (feed_url, guid), name) + content = feed.find('.//{http://search.yahoo.com/mrss/}content') + theplatform_id = url_basename(content.attrib.get('url')) + + return self.url_result(smuggle_url( + 'http://link.theplatform.com/s/ngs/%s?format=SMIL&formats=MPEG4&manifest=f4m' % theplatform_id, + # For some reason, the normal links don't work and we must force the use of f4m + {'force_smil_url': True})) -- cgit v1.1