[youtube] fix extraction for embed restricted live streams(fixes #16433)

[youtube-dl] / youtube_dl / extractor / nationalgeographic.py

diff --git a/youtube_dl/extractor/nationalgeographic.py b/youtube_dl/extractor/nationalgeographic.py
index 890e8d5bcea3b869217f337f95de1ac69e4b85e7..4d2ee64080710a0a531aaf391eeb116da730cc89 100644 (file)
--- a/youtube_dl/extractor/nationalgeographic.py
+++ b/youtube_dl/extractor/nationalgeographic.py
@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from .adobepass import AdobePassIE
 from .theplatform import ThePlatformIE
 from ..utils import (
     smuggle_url,
@@ -65,13 +66,13 @@ class NationalGeographicVideoIE(InfoExtractor):
         }
 
 
-class NationalGeographicIE(ThePlatformIE):
+class NationalGeographicIE(ThePlatformIE, AdobePassIE):
     IE_NAME = 'natgeo'
-    _VALID_URL = r'https?://channel\.nationalgeographic\.com/(?:wild/)?[^/]+/(?:videos|episodes)/(?P<id>[^/?]+)'
+    _VALID_URL = r'https?://channel\.nationalgeographic\.com/(?:(?:(?:wild/)?[^/]+/)?(?:videos|episodes)|u)/(?P<id>[^/?]+)'
 
     _TESTS = [
         {
-            'url': 'http://channel.nationalgeographic.com/the-story-of-god-with-morgan-freeman/videos/uncovering-a-universal-knowledge/',
+            'url': 'http://channel.nationalgeographic.com/u/kdi9Ld0PN2molUUIMSBGxoeDhD729KRjQcnxtetilWPMevo8ZwUBIDuPR0Q3D2LVaTsk0MPRkRWDB8ZhqWVeyoxfsZZm36yRp1j-zPfsHEyI_EgAeFY/',
             'md5': '518c9aa655686cf81493af5cc21e2a04',
             'info_dict': {
                 'id': 'vKInpacll2pC',
@@ -85,7 +86,7 @@ class NationalGeographicIE(ThePlatformIE):
             'add_ie': ['ThePlatform'],
         },
         {
-            'url': 'http://channel.nationalgeographic.com/wild/destination-wild/videos/the-stunning-red-bird-of-paradise/',
+            'url': 'http://channel.nationalgeographic.com/u/kdvOstqYaBY-vSBPyYgAZRUL4sWUJ5XUUPEhc7ISyBHqoIO4_dzfY3K6EjHIC0hmFXoQ7Cpzm6RkET7S3oMlm6CFnrQwSUwo/',
             'md5': 'c4912f656b4cbe58f3e000c489360989',
             'info_dict': {
                 'id': 'Pok5lWCkiEFA',
@@ -101,6 +102,18 @@ class NationalGeographicIE(ThePlatformIE):
         {
             'url': 'http://channel.nationalgeographic.com/the-story-of-god-with-morgan-freeman/episodes/the-power-of-miracles/',
             'only_matching': True,
+        },
+        {
+            'url': 'http://channel.nationalgeographic.com/videos/treasures-rediscovered/',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://channel.nationalgeographic.com/the-story-of-god-with-morgan-freeman/videos/uncovering-a-universal-knowledge/',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://channel.nationalgeographic.com/wild/destination-wild/videos/the-stunning-red-bird-of-paradise/',
+            'only_matching': True,
         }
     ]
 
@@ -110,28 +123,42 @@ class NationalGeographicIE(ThePlatformIE):
         release_url = self._search_regex(
             r'video_auth_playlist_url\s*=\s*"([^"]+)"',
             webpage, 'release url')
+        theplatform_path = self._search_regex(r'https?://link\.theplatform\.com/s/([^?]+)', release_url, 'theplatform path')
+        video_id = theplatform_path.split('/')[-1]
         query = {
             'mbr': 'true',
-            'switch': 'http',
         }
         is_auth = self._search_regex(r'video_is_auth\s*=\s*"([^"]+)"', webpage, 'is auth', fatal=False)
         if is_auth == 'auth':
             auth_resource_id = self._search_regex(
                 r"video_auth_resourceId\s*=\s*'([^']+)'",
                 webpage, 'auth resource id')
-            query['auth'] = self._extract_mvpd_auth(url, display_id, 'natgeo', auth_resource_id)
-
-        return {
-            '_type': 'url_transparent',
-            'ie_key': 'ThePlatform',
-            'url': smuggle_url(
-                update_url_query(release_url, query),
-                {'force_smil_url': True}),
+            query['auth'] = self._extract_mvpd_auth(url, video_id, 'natgeo', auth_resource_id)
+
+        formats = []
+        subtitles = {}
+        for key, value in (('switch', 'http'), ('manifest', 'm3u')):
+            tp_query = query.copy()
+            tp_query.update({
+                key: value,
+            })
+            tp_formats, tp_subtitles = self._extract_theplatform_smil(
+                update_url_query(release_url, tp_query), video_id, 'Downloading %s SMIL data' % value)
+            formats.extend(tp_formats)
+            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
+        self._sort_formats(formats)
+
+        info = self._extract_theplatform_metadata(theplatform_path, display_id)
+        info.update({
+            'id': video_id,
+            'formats': formats,
+            'subtitles': subtitles,
             'display_id': display_id,
-        }
+        })
+        return info
 
 
-class NationalGeographicEpisodeGuideIE(ThePlatformIE):
+class NationalGeographicEpisodeGuideIE(InfoExtractor):
     IE_NAME = 'natgeo:episodeguide'
     _VALID_URL = r'https?://channel\.nationalgeographic\.com/(?:wild/)?(?P<id>[^/]+)/episode-guide'
     _TESTS = [