Merge tag 'upstream/2015.11.27.1'

[youtubedl] / youtube_dl / extractor / ted.py

diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py
index d260c91c2172deabb697c3ad242cfda24f395d41..a48d77c309dcd1f9984cd0a6c71b7af574ca5498 100644 (file)
--- a/youtube_dl/extractor/ted.py
+++ b/youtube_dl/extractor/ted.py
@@ -3,17 +3,17 @@ from __future__ import unicode_literals
 import json
 import re
 
 import json
 import re
 

-from .subtitles import SubtitlesInfoExtractor
+from .common import InfoExtractor

 
 

-from ..utils import (
-    compat_str,
-)
+from ..compat import compat_str
+from ..utils import int_or_none

 
 
 
 

-class TEDIE(SubtitlesInfoExtractor):
+class TEDIE(InfoExtractor):
+    IE_NAME = 'ted'

     _VALID_URL = r'''(?x)
         (?P<proto>https?://)
     _VALID_URL = r'''(?x)
         (?P<proto>https?://)

-        (?P<type>www|embed)(?P<urlmain>\.ted\.com/
+        (?P<type>www|embed(?:-ssl)?)(?P<urlmain>\.ted\.com/

         (
             (?P<type_playlist>playlists(?:/\d+)?) # We have a playlist
             |
         (
             (?P<type_playlist>playlists(?:/\d+)?) # We have a playlist
             |


@@ -27,17 +27,18 @@ class TEDIE(SubtitlesInfoExtractor):
         '''
     _TESTS = [{
         'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
         '''
     _TESTS = [{
         'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',

-        'md5': '4ea1dada91e4174b53dac2bb8ace429d',
+        'md5': 'fc94ac279feebbce69f21c0c6ee82810',

         'info_dict': {
             'id': '102',
             'ext': 'mp4',
             'title': 'The illusion of consciousness',
             'description': ('Philosopher Dan Dennett makes a compelling '
         'info_dict': {
             'id': '102',
             'ext': 'mp4',
             'title': 'The illusion of consciousness',
             'description': ('Philosopher Dan Dennett makes a compelling '

-                'argument that not only don\'t we understand our own '
-                'consciousness, but that half the time our brains are '
-                'actively fooling us.'),
+                            'argument that not only don\'t we understand our own '
+                            'consciousness, but that half the time our brains are '
+                            'actively fooling us.'),

             'uploader': 'Dan Dennett',
             'width': 854,
             'uploader': 'Dan Dennett',
             'width': 854,

+            'duration': 1308,

         }
     }, {
         'url': 'http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms',
         }
     }, {
         'url': 'http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms',


@@ -51,13 +52,52 @@ class TEDIE(SubtitlesInfoExtractor):
         }
     }, {
         'url': 'http://www.ted.com/talks/gabby_giffords_and_mark_kelly_be_passionate_be_courageous_be_your_best',
         }
     }, {
         'url': 'http://www.ted.com/talks/gabby_giffords_and_mark_kelly_be_passionate_be_courageous_be_your_best',

-        'md5': '49144e345a899b8cb34d315f3b9cfeeb',

         'info_dict': {
             'id': '1972',
             'ext': 'mp4',
             'title': 'Be passionate. Be courageous. Be your best.',
             'uploader': 'Gabby Giffords and Mark Kelly',
             'description': 'md5:5174aed4d0f16021b704120360f72b92',
         'info_dict': {
             'id': '1972',
             'ext': 'mp4',
             'title': 'Be passionate. Be courageous. Be your best.',
             'uploader': 'Gabby Giffords and Mark Kelly',
             'description': 'md5:5174aed4d0f16021b704120360f72b92',

+            'duration': 1128,
+        },
+    }, {
+        'url': 'http://www.ted.com/playlists/who_are_the_hackers',
+        'info_dict': {
+            'id': '10',
+            'title': 'Who are the hackers?',
+        },
+        'playlist_mincount': 6,
+    }, {
+        # contains a youtube video
+        'url': 'https://www.ted.com/talks/douglas_adams_parrots_the_universe_and_everything',
+        'add_ie': ['Youtube'],
+        'info_dict': {
+            'id': '_ZG8HBuDjgc',
+            'ext': 'mp4',
+            'title': 'Douglas Adams: Parrots the Universe and Everything',
+            'description': 'md5:01ad1e199c49ac640cb1196c0e9016af',
+            'uploader': 'University of California Television (UCTV)',
+            'uploader_id': 'UCtelevision',
+            'upload_date': '20080522',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # YouTube video
+        'url': 'http://www.ted.com/talks/jeffrey_kluger_the_sibling_bond',
+        'add_ie': ['Youtube'],
+        'info_dict': {
+            'id': 'aFBIPO-P7LM',
+            'ext': 'mp4',
+            'title': 'The hidden power of siblings: Jeff Kluger at TEDxAsheville',
+            'description': 'md5:3d7a4f50d95ca5dd67104e2a20f43fe1',
+            'uploader': 'TEDx Talks',
+            'uploader_id': 'TEDxTalks',
+            'upload_date': '20111216',
+        },
+        'params': {
+            'skip_download': True,

         },
     }]
 
         },
     }]
 


@@ -69,12 +109,12 @@ class TEDIE(SubtitlesInfoExtractor):
 
     def _extract_info(self, webpage):
         info_json = self._search_regex(r'q\("\w+.init",({.+})\)</script>',
 
     def _extract_info(self, webpage):
         info_json = self._search_regex(r'q\("\w+.init",({.+})\)</script>',

-            webpage, 'info json')
+                                       webpage, 'info json')

         return json.loads(info_json)
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url, re.VERBOSE)
         return json.loads(info_json)
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url, re.VERBOSE)

-        if m.group('type') == 'embed':
+        if m.group('type').startswith('embed'):

             desktop_url = m.group('proto') + 'www' + m.group('urlmain')
             return self.url_result(desktop_url, 'TED')
         name = m.group('name')
             desktop_url = m.group('proto') + 'www' + m.group('urlmain')
             return self.url_result(desktop_url, 'TED')
         name = m.group('name')


@@ -89,7 +129,7 @@ class TEDIE(SubtitlesInfoExtractor):
         '''Returns the videos of the playlist'''
 
         webpage = self._download_webpage(url, name,
         '''Returns the videos of the playlist'''
 
         webpage = self._download_webpage(url, name,

-            'Downloading playlist webpage')
+                                         'Downloading playlist webpage')

         info = self._extract_info(webpage)
         playlist_info = info['playlist']
 
         info = self._extract_info(webpage)
         playlist_info = info['playlist']
 


@@ -108,6 +148,18 @@ class TEDIE(SubtitlesInfoExtractor):
 
         talk_info = self._extract_info(webpage)['talks'][0]
 
 
         talk_info = self._extract_info(webpage)['talks'][0]
 

+        external = talk_info.get('external')
+        if external:
+            service = external['service']
+            self.to_screen('Found video from %s' % service)
+            ext_url = None
+            if service.lower() == 'youtube':
+                ext_url = external.get('code')
+            return {
+                '_type': 'url',
+                'url': ext_url or external['uri'],
+            }
+

         formats = [{
             'url': format_url,
             'format_id': format_id,
         formats = [{
             'url': format_url,
             'format_id': format_id,


@@ -118,57 +170,92 @@ class TEDIE(SubtitlesInfoExtractor):
                 finfo = self._NATIVE_FORMATS.get(f['format_id'])
                 if finfo:
                     f.update(finfo)
                 finfo = self._NATIVE_FORMATS.get(f['format_id'])
                 if finfo:
                     f.update(finfo)

-        else:
-            # Use rtmp downloads
-            formats = [{
-                'format_id': f['name'],
-                'url': talk_info['streamer'],
-                'play_path': f['file'],
-                'ext': 'flv',
-                'width': f['width'],
-                'height': f['height'],
-                'tbr': f['bitrate'],
-            } for f in talk_info['resources']['rtmp']]
+
+        for format_id, resources in talk_info['resources'].items():
+            if format_id == 'h264':
+                for resource in resources:
+                    bitrate = int_or_none(resource.get('bitrate'))
+                    formats.append({
+                        'url': resource['file'],
+                        'format_id': '%s-%sk' % (format_id, bitrate),
+                        'tbr': bitrate,
+                    })
+            elif format_id == 'rtmp':
+                streamer = talk_info.get('streamer')
+                if not streamer:
+                    continue
+                for resource in resources:
+                    formats.append({
+                        'format_id': '%s-%s' % (format_id, resource.get('name')),
+                        'url': streamer,
+                        'play_path': resource['file'],
+                        'ext': 'flv',
+                        'width': int_or_none(resource.get('width')),
+                        'height': int_or_none(resource.get('height')),
+                        'tbr': int_or_none(resource.get('bitrate')),
+                    })
+            elif format_id == 'hls':
+                hls_formats = self._extract_m3u8_formats(
+                    resources.get('stream'), video_name, 'mp4', m3u8_id=format_id)
+                for f in hls_formats:
+                    if f.get('format_id') == 'hls-meta':
+                        continue
+                    if not f.get('height'):
+                        f['vcodec'] = 'none'
+                    else:
+                        f['acodec'] = 'none'
+                formats.extend(hls_formats)
+
+        audio_download = talk_info.get('audioDownload')
+        if audio_download:
+            formats.append({
+                'url': audio_download,
+                'format_id': 'audio',
+                'vcodec': 'none',
+                'preference': -0.5,
+            })
+

         self._sort_formats(formats)
 
         video_id = compat_str(talk_info['id'])
         self._sort_formats(formats)
 
         video_id = compat_str(talk_info['id'])

-        # subtitles
-        video_subtitles = self.extract_subtitles(video_id, talk_info)
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, talk_info)
-            return

 
         thumbnail = talk_info['thumb']
         if not thumbnail.startswith('http'):
             thumbnail = 'http://' + thumbnail
         return {
             'id': video_id,
 
         thumbnail = talk_info['thumb']
         if not thumbnail.startswith('http'):
             thumbnail = 'http://' + thumbnail
         return {
             'id': video_id,

-            'title': talk_info['title'],
+            'title': talk_info['title'].strip(),

             'uploader': talk_info['speaker'],
             'thumbnail': thumbnail,
             'description': self._og_search_description(webpage),
             'uploader': talk_info['speaker'],
             'thumbnail': thumbnail,
             'description': self._og_search_description(webpage),

-            'subtitles': video_subtitles,
+            'subtitles': self._get_subtitles(video_id, talk_info),

             'formats': formats,
             'formats': formats,

+            'duration': talk_info.get('duration'),

         }
 
         }
 

-    def _get_available_subtitles(self, video_id, talk_info):
+    def _get_subtitles(self, video_id, talk_info):

         languages = [lang['languageCode'] for lang in talk_info.get('languages', [])]
         if languages:
             sub_lang_list = {}
             for l in languages:
         languages = [lang['languageCode'] for lang in talk_info.get('languages', [])]
         if languages:
             sub_lang_list = {}
             for l in languages:

-                url = 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/srt' % (video_id, l)
-                sub_lang_list[l] = url
+                sub_lang_list[l] = [
+                    {
+                        'url': 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/%s' % (video_id, l, ext),
+                        'ext': ext,
+                    }
+                    for ext in ['ted', 'srt']
+                ]

             return sub_lang_list
         else:
             return sub_lang_list
         else:

-            self._downloader.report_warning('video doesn\'t have subtitles')

             return {}
 
     def _watch_info(self, url, name):
         webpage = self._download_webpage(url, name)
 
         config_json = self._html_search_regex(
             return {}
 
     def _watch_info(self, url, name):
         webpage = self._download_webpage(url, name)
 
         config_json = self._html_search_regex(

-            r"data-config='([^']+)", webpage, 'config')
-        config = json.loads(config_json)
+            r'"pages\.jwplayer"\s*,\s*({.+?})\s*\)\s*</script>',
+            webpage, 'config')
+        config = json.loads(config_json)['config']

         video_url = config['video']['url']
         thumbnail = config.get('image', {}).get('url')
 
         video_url = config['video']['url']
         thumbnail = config.get('image', {}).get('url')