debian/copyright: use spaces rather than tabs to start continuation lines.

[youtubedl] / youtube_dl / extractor / voxmedia.py

diff --git a/youtube_dl/extractor/voxmedia.py b/youtube_dl/extractor/voxmedia.py
index b1b32ad44ecfd796e46219a87ea71caa3587face..b318e15d4b4da53fe0b42f3537d0fde66182299c 100644 (file)
--- a/youtube_dl/extractor/voxmedia.py
+++ b/youtube_dl/extractor/voxmedia.py
@@ -2,43 +2,114 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from __future__ import unicode_literals
 
 from .common import InfoExtractor

+from .once import OnceIE

 from ..compat import compat_urllib_parse_unquote
 from ..compat import compat_urllib_parse_unquote

+from ..utils import (
+    ExtractorError,
+    int_or_none,
+)
+
+
+class VoxMediaVolumeIE(OnceIE):
+    _VALID_URL = r'https?://volume\.vox-cdn\.com/embed/(?P<id>[0-9a-f]{9})'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        setup = self._parse_json(self._search_regex(
+            r'setup\s*=\s*({.+});', webpage, 'setup'), video_id)
+        video_data = setup.get('video') or {}
+        info = {
+            'id': video_id,
+            'title': video_data.get('title_short'),
+            'description': video_data.get('description_long') or video_data.get('description_short'),
+            'thumbnail': video_data.get('brightcove_thumbnail')
+        }
+        asset = setup.get('asset') or setup.get('params') or {}
+
+        formats = []
+        hls_url = asset.get('hls_url')
+        if hls_url:
+            formats.extend(self._extract_m3u8_formats(
+                hls_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+        mp4_url = asset.get('mp4_url')
+        if mp4_url:
+            tbr = self._search_regex(r'-(\d+)k\.', mp4_url, 'bitrate', default=None)
+            format_id = 'http'
+            if tbr:
+                format_id += '-' + tbr
+            formats.append({
+                'format_id': format_id,
+                'url': mp4_url,
+                'tbr': int_or_none(tbr),
+            })
+        if formats:
+            self._sort_formats(formats)
+            info['formats'] = formats
+            return info
+
+        for provider_video_type in ('ooyala', 'youtube', 'brightcove'):
+            provider_video_id = video_data.get('%s_id' % provider_video_type)
+            if not provider_video_id:
+                continue
+            if provider_video_type == 'brightcove':
+                info['formats'] = self._extract_once_formats(provider_video_id)
+                self._sort_formats(info['formats'])
+            else:
+                info.update({
+                    '_type': 'url_transparent',
+                    'url': provider_video_id if provider_video_type == 'youtube' else '%s:%s' % (provider_video_type, provider_video_id),
+                    'ie_key': provider_video_type.capitalize(),
+                })
+            return info
+        raise ExtractorError('Unable to find provider video id')

 
 
 class VoxMediaIE(InfoExtractor):
 
 
 class VoxMediaIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:www\.)?(?:theverge|vox|sbnation|eater|polygon|curbed|racked)\.com/(?:[^/]+/)*(?P<id>[^/?]+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:(?:theverge|vox|sbnation|eater|polygon|curbed|racked|funnyordie)\.com|recode\.net)/(?:[^/]+/)*(?P<id>[^/?]+)'

     _TESTS = [{
     _TESTS = [{

+        # Volume embed, Youtube

         'url': 'http://www.theverge.com/2014/6/27/5849272/material-world-how-google-discovered-what-software-is-made-of',
         'url': 'http://www.theverge.com/2014/6/27/5849272/material-world-how-google-discovered-what-software-is-made-of',

-        'md5': '73856edf3e89a711e70d5cf7cb280b37',

         'info_dict': {
         'info_dict': {

-            'id': '11eXZobjrG8DCSTgrNjVinU-YmmdYjhe',
+            'id': 'j4mLW6x17VM',

             'ext': 'mp4',
             'ext': 'mp4',

-            'title': 'Google\'s new material design direction',
-            'description': 'md5:2f44f74c4d14a1f800ea73e1c6832ad2',
+            'title': 'Material world: how Google discovered what software is made of',
+            'description': 'md5:dfc17e7715e3b542d66e33a109861382',
+            'upload_date': '20190710',
+            'uploader_id': 'TheVerge',
+            'uploader': 'The Verge',

         },
         },

-        'add_ie': ['Ooyala'],
+        'add_ie': ['Youtube'],

     }, {
     }, {

-        # data-ooyala-id
+        # Volume embed, Youtube

         'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet',
         'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet',

-        'md5': 'd744484ff127884cd2ba09e3fa604e4b',
+        'md5': '4c8f4a0937752b437c3ebc0ed24802b5',

         'info_dict': {
         'info_dict': {

-            'id': 'RkZXU4cTphOCPDMZg5oEounJyoFI0g-B',
+            'id': 'Gy8Md3Eky38',

             'ext': 'mp4',
             'title': 'The Nexus 6: hands-on with Google\'s phablet',
             'ext': 'mp4',
             'title': 'The Nexus 6: hands-on with Google\'s phablet',

-            'description': 'md5:87a51fe95ff8cea8b5bdb9ac7ae6a6af',
+            'description': 'md5:d9f0216e5fb932dd2033d6db37ac3f1d',
+            'uploader_id': 'TheVerge',
+            'upload_date': '20141021',
+            'uploader': 'The Verge',

         },
         },

-        'add_ie': ['Ooyala'],
+        'add_ie': ['Youtube'],
+        'skip': 'similar to the previous test',

     }, {
     }, {

-        # volume embed
+        # Volume embed, Youtube

         'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill',
         'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill',

-        'md5': '375c483c5080ab8cd85c9c84cfc2d1e4',

         'info_dict': {
         'info_dict': {

-            'id': 'wydzk3dDpmRz7PQoXRsTIX6XTkPjYL0b',
+            'id': 'YCjDnX-Xzhg',

             'ext': 'mp4',
             'ext': 'mp4',

-            'title': 'The new frontier of LGBTQ civil rights, explained',
-            'description': 'md5:0dc58e94a465cbe91d02950f770eb93f',
+            'title': "Mississippi's laws are so bad that its anti-LGBTQ law isn't needed to allow discrimination",
+            'description': 'md5:fc1317922057de31cd74bce91eb1c66c',
+            'uploader_id': 'voxdotcom',
+            'upload_date': '20150915',
+            'uploader': 'Vox',

         },
         },

-        'add_ie': ['Ooyala'],
+        'add_ie': ['Youtube'],
+        'skip': 'similar to the previous test',

     }, {
         # youtube embed
         'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance',
     }, {
         # youtube embed
         'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance',


@@ -53,6 +124,7 @@ class VoxMediaIE(InfoExtractor):
             'uploader': 'Vox',
         },
         'add_ie': ['Youtube'],
             'uploader': 'Vox',
         },
         'add_ie': ['Youtube'],

+        'skip': 'Page no longer contain videos',

     }, {
         # SBN.VideoLinkset.entryGroup multiple ooyala embeds
         'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
     }, {
         # SBN.VideoLinkset.entryGroup multiple ooyala embeds
         'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',


@@ -78,6 +150,18 @@ class VoxMediaIE(InfoExtractor):
                 'description': 'md5:e02d56b026d51aa32c010676765a690d',
             },
         }],
                 'description': 'md5:e02d56b026d51aa32c010676765a690d',
             },
         }],

+        'skip': 'Page no longer contain videos',
+    }, {
+        # volume embed, Brightcove Once
+        'url': 'https://www.recode.net/2014/6/17/11628066/post-post-pc-ceo-the-full-code-conference-video-of-microsofts-satya',
+        'md5': '2dbc77b8b0bff1894c2fce16eded637d',
+        'info_dict': {
+            'id': '1231c973d',
+            'ext': 'mp4',
+            'title': 'Post-Post-PC CEO: The Full Code Conference Video of Microsoft\'s Satya Nadella',
+            'description': 'The longtime veteran was chosen earlier this year as the software giant\'s third leader in its history.',
+        },
+        'add_ie': ['VoxMediaVolume'],

     }]
 
     def _real_extract(self, url):
     }]
 
     def _real_extract(self, url):


@@ -85,9 +169,14 @@ class VoxMediaIE(InfoExtractor):
         webpage = compat_urllib_parse_unquote(self._download_webpage(url, display_id))
 
         def create_entry(provider_video_id, provider_video_type, title=None, description=None):
         webpage = compat_urllib_parse_unquote(self._download_webpage(url, display_id))
 
         def create_entry(provider_video_id, provider_video_type, title=None, description=None):

+            video_url = {
+                'youtube': '%s',
+                'ooyala': 'ooyala:%s',
+                'volume': 'http://volume.vox-cdn.com/embed/%s',
+            }[provider_video_type] % provider_video_id

             return {
                 '_type': 'url_transparent',
             return {
                 '_type': 'url_transparent',

-                'url': provider_video_id if provider_video_type == 'youtube' else '%s:%s' % (provider_video_type, provider_video_id),
+                'url': video_url,

                 'title': title or self._og_search_title(webpage),
                 'description': description or self._og_search_description(webpage),
             }
                 'title': title or self._og_search_title(webpage),
                 'description': description or self._og_search_description(webpage),
             }


@@ -118,17 +207,7 @@ class VoxMediaIE(InfoExtractor):
         volume_uuid = self._search_regex(
             r'data-volume-uuid="([^"]+)"', webpage, 'volume uuid', default=None)
         if volume_uuid:
         volume_uuid = self._search_regex(
             r'data-volume-uuid="([^"]+)"', webpage, 'volume uuid', default=None)
         if volume_uuid:

-            volume_webpage = self._download_webpage(
-                'http://volume.vox-cdn.com/embed/%s' % volume_uuid, volume_uuid)
-            video_data = self._parse_json(self._search_regex(
-                r'Volume\.createVideo\(({.+})\s*,\s*{.*}\s*,\s*\[.*\]\s*,\s*{.*}\);', volume_webpage, 'video data'), volume_uuid)
-            for provider_video_type in ('ooyala', 'youtube'):
-                provider_video_id = video_data.get('%s_id' % provider_video_type)
-                if provider_video_id:
-                    description = video_data.get('description_long') or video_data.get('description_short')
-                    entries.append(create_entry(
-                        provider_video_id, provider_video_type, video_data.get('title_short'), description))
-                    break
+            entries.append(create_entry(volume_uuid, 'volume'))

 
         if len(entries) == 1:
             return entries[0]
 
         if len(entries) == 1:
             return entries[0]