[kinja] add support for Kinja embeds

closes #5756
closes #11282
closes #22237
closes #22384

youtube_dl/extractor/extractors.py

@@ -513,6 +513,7 @@
 from .ketnet import KetnetIE
 from .khanacademy import KhanAcademyIE
 from .kickstarter import KickStarterIE
+from .kinja import KinjaEmbedIE
 from .kinopoisk import KinoPoiskIE
 from .konserthusetplay import KonserthusetPlayIE
 from .kontrtube import KontrTubeIE

youtube_dl/extractor/generic.py

@@ -119,6 +119,7 @@
 from .expressen import ExpressenIE
 from .zype import ZypeIE
 from .odnoklassniki import OdnoklassnikiIE
+from .kinja import KinjaEmbedIE
 
 
 class GenericIE(InfoExtractor):
@@ -1487,16 +1488,18 @@ class GenericIE(InfoExtractor):
                 'timestamp': 1432570283,
             },
         },
-        # OnionStudios embed
+        # Kinja embed
         {
             'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
             'info_dict': {
-                'id': '2855',
+                'id': '106351',
                 'ext': 'mp4',
                 'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
+                'description': 'Migrated from OnionStudios',
                 'thumbnail': r're:^https?://.*\.jpe?g$',
-                'uploader': 'ClickHole',
+                'uploader': 'clickhole',
-                'uploader_id': 'clickhole',
+                'upload_date': '20150527',
+                'timestamp': 1432744860,
             }
         },
         # SnagFilms embed
@@ -2894,6 +2897,12 @@ def _real_extract(self, url):
         if senate_isvp_url:
             return self.url_result(senate_isvp_url, 'SenateISVP')
 
+        # Look for Kinja embeds
+        kinja_embed_urls = KinjaEmbedIE._extract_urls(webpage, url)
+        if kinja_embed_urls:
+            return self.playlist_from_matches(
+                kinja_embed_urls, video_id, video_title)
+
         # Look for OnionStudios embeds
         onionstudios_url = OnionStudiosIE._extract_url(webpage)
         if onionstudios_url:

youtube_dl/extractor/kinja.py

@@ -0,0 +1,221 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_str,
+    compat_urllib_parse_unquote,
+)
+from ..utils import (
+    int_or_none,
+    parse_iso8601,
+    strip_or_none,
+    try_get,
+    unescapeHTML,
+    urljoin,
+)
+
+
+class KinjaEmbedIE(InfoExtractor):
+    IENAME = 'kinja:embed'
+    _DOMAIN_REGEX = r'''(?:[^.]+\.)?
+        (?:
+            avclub|
+            clickhole|
+            deadspin|
+            gizmodo|
+            jalopnik|
+            jezebel|
+            kinja|
+            kotaku|
+            lifehacker|
+            splinternews|
+            the(?:inventory|onion|root|takeout)
+        )\.com'''
+    _COMMON_REGEX = r'''/
+        (?:
+            ajax/inset|
+            embed/video
+        )/iframe\?.*?\bid='''
+    _VALID_URL = r'''(?x)https?://%s%s
+        (?P<type>
+            fb|
+            imgur|
+            instagram|
+            jwp(?:layer)?-video|
+            kinjavideo|
+            mcp|
+            megaphone|
+            ooyala|
+            soundcloud(?:-playlist)?|
+            tumblr-post|
+            twitch-stream|
+            twitter|
+            ustream-channel|
+            vimeo|
+            vine|
+            youtube-(?:list|video)
+        )-(?P<id>[^&]+)''' % (_DOMAIN_REGEX, _COMMON_REGEX)
+    _TESTS = [{
+        'url': 'https://kinja.com/ajax/inset/iframe?id=fb-10103303356633621',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=kinjavideo-100313',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=megaphone-PPY1300931075',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=ooyala-xzMXhleDpopuT0u1ijt_qZj3Va-34pEX%2FZTIxYmJjZDM2NWYzZDViZGRiOWJjYzc5',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=soundcloud-128574047',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=soundcloud-playlist-317413750',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=tumblr-post-160130699814-daydreams-at-midnight',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=twitch-stream-libratus_extra',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=twitter-1068875942473404422',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=ustream-channel-10414700',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=vimeo-120153502',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=vine-5BlvV5qqPrD',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=youtube-list-BCQ3KyrPjgA/PLE6509247C270A72E',
+        'only_matching': True,
+    }, {
+        'url': 'https://kinja.com/ajax/inset/iframe?id=youtube-video-00QyL0AgPAE',
+        'only_matching': True,
+    }]
+    _JWPLATFORM_PROVIDER = ('cdn.jwplayer.com/v2/media/', 'JWPlatform')
+    _PROVIDER_MAP = {
+        'fb': ('facebook.com/video.php?v=', 'Facebook'),
+        'imgur': ('imgur.com/', 'Imgur'),
+        'instagram': ('instagram.com/p/', 'Instagram'),
+        'jwplayer-video': _JWPLATFORM_PROVIDER,
+        'jwp-video': _JWPLATFORM_PROVIDER,
+        'megaphone': ('player.megaphone.fm/', 'Generic'),
+        'ooyala': ('player.ooyala.com/player.js?embedCode=', 'Ooyala'),
+        'soundcloud': ('api.soundcloud.com/tracks/', 'Soundcloud'),
+        'soundcloud-playlist': ('api.soundcloud.com/playlists/', 'SoundcloudPlaylist'),
+        'tumblr-post': ('%s.tumblr.com/post/%s', 'Tumblr'),
+        'twitch-stream': ('twitch.tv/', 'TwitchStream'),
+        'twitter': ('twitter.com/i/cards/tfw/v1/', 'TwitterCard'),
+        'ustream-channel': ('ustream.tv/embed/', 'Ustream'),
+        'vimeo': ('vimeo.com/', 'Vimeo'),
+        'vine': ('vine.co/v/', 'Vine'),
+        'youtube-list': ('youtube.com/embed/%s?list=%s', 'YoutubePlaylist'),
+        'youtube-video': ('youtube.com/embed/', 'Youtube'),
+    }
+
+    @staticmethod
+    def _extract_urls(webpage, url):
+        return [urljoin(url, unescapeHTML(mobj.group('url'))) for mobj in re.finditer(
+            r'(?x)<iframe[^>]+?src=(?P<q>["\'])(?P<url>(?:(?:https?:)?//%s)?%s(?:(?!\1).)+)\1' % (KinjaEmbedIE._DOMAIN_REGEX, KinjaEmbedIE._COMMON_REGEX),
+            webpage)]
+
+    def _real_extract(self, url):
+        video_type, video_id = re.match(self._VALID_URL, url).groups()
+
+        provider = self._PROVIDER_MAP.get(video_type)
+        if provider:
+            video_id = compat_urllib_parse_unquote(video_id)
+            if video_type == 'tumblr-post':
+                video_id, blog = video_id.split('-', 1)
+                result_url = provider[0] % (blog, video_id)
+            elif video_type == 'youtube-list':
+                video_id, playlist_id = video_id.split('/')
+                result_url = provider[0] % (video_id, playlist_id)
+            else:
+                if video_type == 'ooyala':
+                    video_id = video_id.split('/')[0]
+                result_url = provider[0] + video_id
+            return self.url_result('http://' + result_url, provider[1])
+
+        if video_type == 'kinjavideo':
+            data = self._download_json(
+                'https://kinja.com/api/core/video/views/videoById',
+                video_id, query={'videoId': video_id})['data']
+            title = data['title']
+
+            formats = []
+            for k in ('signedPlaylist', 'streaming'):
+                m3u8_url = data.get(k + 'Url')
+                if m3u8_url:
+                    formats.extend(self._extract_m3u8_formats(
+                        m3u8_url, video_id, 'mp4', 'm3u8_native',
+                        m3u8_id='hls', fatal=False))
+            self._sort_formats(formats)
+
+            thumbnail = None
+            poster = data.get('poster') or {}
+            poster_id = poster.get('id')
+            if poster_id:
+                thumbnail = 'https://i.kinja-img.com/gawker-media/image/upload/%s.%s' % (poster_id, poster.get('format') or 'jpg')
+
+            return {
+                'id': video_id,
+                'title': title,
+                'description': strip_or_none(data.get('description')),
+                'formats': formats,
+                'tags': data.get('tags'),
+                'timestamp': int_or_none(try_get(
+                    data, lambda x: x['postInfo']['publishTimeMillis']), 1000),
+                'thumbnail': thumbnail,
+                'uploader': data.get('network'),
+            }
+        else:
+            video_data = self._download_json(
+                'https://api.vmh.univision.com/metadata/v1/content/' + video_id,
+                video_id)['videoMetadata']
+            iptc = video_data['photoVideoMetadataIPTC']
+            title = iptc['title']['en']
+            fmg = video_data.get('photoVideoMetadata_fmg') or {}
+            tvss_domain = fmg.get('tvssDomain') or 'https://auth.univision.com'
+            data = self._download_json(
+                tvss_domain + '/api/v3/video-auth/url-signature-tokens',
+                video_id, query={'mcpids': video_id})['data'][0]
+            formats = []
+
+            rendition_url = data.get('renditionUrl')
+            if rendition_url:
+                formats = self._extract_m3u8_formats(
+                    rendition_url, video_id, 'mp4',
+                    'm3u8_native', m3u8_id='hls', fatal=False)
+
+            fallback_rendition_url = data.get('fallbackRenditionUrl')
+            if fallback_rendition_url:
+                formats.append({
+                    'format_id': 'fallback',
+                    'tbr': int_or_none(self._search_regex(
+                        r'_(\d+)\.mp4', fallback_rendition_url,
+                        'bitrate', default=None)),
+                    'url': fallback_rendition_url,
+                })
+
+            self._sort_formats(formats)
+
+            return {
+                'id': video_id,
+                'title': title,
+                'thumbnail': try_get(iptc, lambda x: x['cloudinaryLink']['link'], compat_str),
+                'uploader': fmg.get('network'),
+                'duration': int_or_none(iptc.get('fileDuration')),
+                'formats': formats,
+                'description': try_get(iptc, lambda x: x['description']['en'], compat_str),
+                'timestamp': parse_iso8601(iptc.get('dateReleased')),
+            }

youtube_dl/extractor/onionstudios.py

@@ -4,13 +4,8 @@
 import re
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import compat_str
-    compat_str,
+from ..utils import js_to_json
-    int_or_none,
-    js_to_json,
-    parse_iso8601,
-    try_get,
-)
 
 
 class OnionStudiosIE(InfoExtractor):
@@ -20,7 +15,7 @@ class OnionStudiosIE(InfoExtractor):
         'url': 'http://www.onionstudios.com/videos/hannibal-charges-forward-stops-for-a-cocktail-2937',
         'md5': '5a118d466d62b5cd03647cf2c593977f',
         'info_dict': {
-            'id': '2937',
+            'id': '3459881',
             'ext': 'mp4',
             'title': 'Hannibal charges forward, stops for a cocktail',
             'description': 'md5:545299bda6abf87e5ec666548c6a9448',
@@ -53,43 +48,6 @@ def _real_extract(self, url):
         mcp_id = compat_str(self._parse_json(self._search_regex(
             r'window\.mcpMapping\s*=\s*({.+?});', webpage,
             'MCP Mapping'), video_id, js_to_json)[video_id]['mcp_id'])
-        video_data = self._download_json(
+        return self.url_result(
-            'https://api.vmh.univision.com/metadata/v1/content/' + mcp_id,
+            'http://kinja.com/ajax/inset/iframe?id=mcp-' + mcp_id,
-            mcp_id)['videoMetadata']
+            'KinjaEmbed', mcp_id)
-        iptc = video_data['photoVideoMetadataIPTC']
-        title = iptc['title']['en']
-        fmg = video_data.get('photoVideoMetadata_fmg') or {}
-        tvss_domain = fmg.get('tvssDomain') or 'https://auth.univision.com'
-        data = self._download_json(
-            tvss_domain + '/api/v3/video-auth/url-signature-tokens',
-            mcp_id, query={'mcpids': mcp_id})['data'][0]
-        formats = []
-
-        rendition_url = data.get('renditionUrl')
-        if rendition_url:
-            formats = self._extract_m3u8_formats(
-                rendition_url, mcp_id, 'mp4',
-                'm3u8_native', m3u8_id='hls', fatal=False)
-
-        fallback_rendition_url = data.get('fallbackRenditionUrl')
-        if fallback_rendition_url:
-            formats.append({
-                'format_id': 'fallback',
-                'tbr': int_or_none(self._search_regex(
-                    r'_(\d+)\.mp4', fallback_rendition_url,
-                    'bitrate', default=None)),
-                'url': fallback_rendition_url,
-            })
-
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'thumbnail': try_get(iptc, lambda x: x['cloudinaryLink']['link'], compat_str),
-            'uploader': fmg.get('network'),
-            'duration': int_or_none(iptc.get('fileDuration')),
-            'formats': formats,
-            'description': try_get(iptc, lambda x: x['description']['en'], compat_str),
-            'timestamp': parse_iso8601(iptc.get('dateReleased')),
-        }