youtube-dl

commit 78338f71ca2d96e4bf507c438fbb2751742989b1
parent f5172a308418a54d077e19495a5c560f2bd644e6
Author: Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Date:   Thu, 26 Jun 2014 16:34:36 +0200

[livestream:original] Add support for folder urls (closes #2631)

The webpage only contains shortened links for the videos, since the server
doesn't support HEAD requests, we use an specific extractor for them.

Diffstat:
M
test/test_playlists.py
 | 
9
+++++++++
M
youtube_dl/extractor/__init__.py
 | 
6
+++++-
M
youtube_dl/extractor/common.py
 | 
3
+++
M
youtube_dl/extractor/livestream.py
 | 
52
+++++++++++++++++++++++++++++++++++++++++++++++-----

4 files changed, 64 insertions(+), 6 deletions(-)
diff --git a/test/test_playlists.py b/test/test_playlists.py
@@ -30,6 +30,7 @@ from youtube_dl.extractor import (
     SoundcloudPlaylistIE,
     TeacherTubeClassroomIE,
     LivestreamIE,
+    LivestreamOriginalIE,
     NHLVideocenterIE,
     BambuserChannelIE,
     BandcampAlbumIE,
@@ -155,6 +156,14 @@ class TestPlaylists(unittest.TestCase):
         self.assertEqual(result['title'], 'TEDCity2.0 (English)')
         self.assertTrue(len(result['entries']) >= 4)
 
+    def test_livestreamoriginal_folder(self):
+        dl = FakeYDL()
+        ie = LivestreamOriginalIE(dl)
+        result = ie.extract('https://www.livestream.com/newplay/folder?dirId=a07bf706-d0e4-4e75-a747-b021d84f2fd3')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'a07bf706-d0e4-4e75-a747-b021d84f2fd3')
+        self.assertTrue(len(result['entries']) >= 28)
+
     def test_nhl_videocenter(self):
         dl = FakeYDL()
         ie = NHLVideocenterIE(dl)
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
@@ -147,7 +147,11 @@ from .ku6 import Ku6IE
 from .la7 import LA7IE
 from .lifenews import LifeNewsIE
 from .liveleak import LiveLeakIE
-from .livestream import LivestreamIE, LivestreamOriginalIE
+from .livestream import (
+    LivestreamIE,
+    LivestreamOriginalIE,
+    LivestreamShortenerIE,
+)
 from .lynda import (
     LyndaIE,
     LyndaCourseIE
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
@@ -459,6 +459,9 @@ class InfoExtractor(object):
         if secure: regexes = self._og_regexes('video:secure_url') + regexes
         return self._html_search_regex(regexes, html, name, **kargs)
 
+    def _og_search_url(self, html, **kargs):
+        return self._og_search_property('url', html, **kargs)
+
     def _html_search_meta(self, name, html, display_name=None, fatal=False):
         if display_name is None:
             display_name = name
diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py
@@ -9,6 +9,7 @@ from ..utils import (
     compat_urlparse,
     xpath_with_ns,
     compat_str,
+    orderedSet,
 )
 
 
@@ -64,7 +65,10 @@ class LivestreamIE(InfoExtractor):
 # The original version of Livestream uses a different system
 class LivestreamOriginalIE(InfoExtractor):
     IE_NAME = 'livestream:original'
-    _VALID_URL = r'https?://www\.livestream\.com/(?P<user>[^/]+)/video\?.*?clipId=(?P<id>.*?)(&|$)'
+    _VALID_URL = r'''(?x)https?://www\.livestream\.com/
+        (?P<user>[^/]+)/(?P<type>video|folder)
+        (?:\?.*?Id=|/)(?P<id>.*?)(&|$)
+        '''
     _TEST = {
         'url': 'http://www.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
         'info_dict': {
@@ -78,10 +82,7 @@ class LivestreamOriginalIE(InfoExtractor):
         },
     }
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        user = mobj.group('user')
+    def _extract_video(self, user, video_id):
         api_url = 'http://x{0}x.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id={1}'.format(user, video_id)
 
         info = self._download_xml(api_url, video_id)
@@ -99,3 +100,44 @@ class LivestreamOriginalIE(InfoExtractor):
             'ext': 'flv',
             'thumbnail': thumbnail_url,
         }
+
+    def _extract_folder(self, url, folder_id):
+        webpage = self._download_webpage(url, folder_id)
+        urls = orderedSet(re.findall(r'<a href="(https?://livestre\.am/.*?)"', webpage))
+
+        return {
+            '_type': 'playlist',
+            'id': folder_id,
+            'entries': [{
+                '_type': 'url',
+                'url': video_url,
+            } for video_url in urls],
+        }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        id = mobj.group('id')
+        user = mobj.group('user')
+        url_type = mobj.group('type')
+        if url_type == 'folder':
+            return self._extract_folder(url, id)
+        else:
+            return self._extract_video(user, id)
+
+
+# The server doesn't support HEAD request, the generic extractor can't detect
+# the redirection
+class LivestreamShortenerIE(InfoExtractor):
+    IE_NAME = 'livestream:shortener'
+    IE_DESC = False  # Do not list
+    _VALID_URL = r'https?://livestre\.am/(?P<id>.+)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        id = mobj.group('id')
+        webpage = self._download_webpage(url, id)
+
+        return {
+            '_type': 'url',
+            'url': self._og_search_url(webpage),
+        }