]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/limelight.py
Merge pull request #4 from debian-janitor/lintian-fixes
[youtubedl] / youtube_dl / extractor / limelight.py
index e635f3c4dc46c6407a166dec9ab2ef06981b6221..39f74d2822bc7296df8a5c16e5edfce3298e82ab 100644 (file)
@@ -4,38 +4,97 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_HTTPError
 from ..utils import (
     determine_ext,
     float_or_none,
     int_or_none,
 from ..utils import (
     determine_ext,
     float_or_none,
     int_or_none,
+    smuggle_url,
+    try_get,
+    unsmuggle_url,
+    ExtractorError,
 )
 
 
 class LimelightBaseIE(InfoExtractor):
     _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
 )
 
 
 class LimelightBaseIE(InfoExtractor):
     _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
-    _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json'
-
-    def _call_playlist_service(self, item_id, method, fatal=True):
-        return self._download_json(
-            self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
-            item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal)
-
-    def _call_api(self, organization_id, item_id, method):
-        return self._download_json(
-            self._API_URL % (organization_id, self._API_PATH, item_id, method),
-            item_id, 'Downloading API %s JSON' % method)
-
-    def _extract(self, item_id, pc_method, mobile_method, meta_method):
-        pc = self._call_playlist_service(item_id, pc_method)
-        metadata = self._call_api(pc['orgId'], item_id, meta_method)
-        mobile = self._call_playlist_service(item_id, mobile_method, fatal=False)
-        return pc, mobile, metadata
-
-    def _extract_info(self, streams, mobile_urls, properties):
-        video_id = properties['media_id']
+
+    @classmethod
+    def _extract_urls(cls, webpage, source_url):
+        lm = {
+            'Media': 'media',
+            'Channel': 'channel',
+            'ChannelList': 'channel_list',
+        }
+
+        def smuggle(url):
+            return smuggle_url(url, {'source_url': source_url})
+
+        entries = []
+        for kind, video_id in re.findall(
+                r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})',
+                webpage):
+            entries.append(cls.url_result(
+                smuggle('limelight:%s:%s' % (lm[kind], video_id)),
+                'Limelight%s' % kind, video_id))
+        for mobj in re.finditer(
+                # As per [1] class attribute should be exactly equal to
+                # LimelightEmbeddedPlayerFlash but numerous examples seen
+                # that don't exactly match it (e.g. [2]).
+                # 1. http://support.3playmedia.com/hc/en-us/articles/227732408-Limelight-Embedding-the-Captions-Plugin-with-the-Limelight-Player-on-Your-Webpage
+                # 2. http://www.sedona.com/FacilitatorTraining2017
+                r'''(?sx)
+                    <object[^>]+class=(["\'])(?:(?!\1).)*\bLimelightEmbeddedPlayerFlash\b(?:(?!\1).)*\1[^>]*>.*?
+                        <param[^>]+
+                            name=(["\'])flashVars\2[^>]+
+                            value=(["\'])(?:(?!\3).)*(?P<kind>media|channel(?:List)?)Id=(?P<id>[a-z0-9]{32})
+                ''', webpage):
+            kind, video_id = mobj.group('kind'), mobj.group('id')
+            entries.append(cls.url_result(
+                smuggle('limelight:%s:%s' % (kind, video_id)),
+                'Limelight%s' % kind.capitalize(), video_id))
+        # http://support.3playmedia.com/hc/en-us/articles/115009517327-Limelight-Embedding-the-Audio-Description-Plugin-with-the-Limelight-Player-on-Your-Web-Page)
+        for video_id in re.findall(
+                r'(?s)LimelightPlayerUtil\.embed\s*\(\s*{.*?\bmediaId["\']\s*:\s*["\'](?P<id>[a-z0-9]{32})',
+                webpage):
+            entries.append(cls.url_result(
+                smuggle('limelight:media:%s' % video_id),
+                LimelightMediaIE.ie_key(), video_id))
+        return entries
+
+    def _call_playlist_service(self, item_id, method, fatal=True, referer=None):
+        headers = {}
+        if referer:
+            headers['Referer'] = referer
+        try:
+            return self._download_json(
+                self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
+                item_id, 'Downloading PlaylistService %s JSON' % method,
+                fatal=fatal, headers=headers)
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                error = self._parse_json(e.cause.read().decode(), item_id)['detail']['contentAccessPermission']
+                if error == 'CountryDisabled':
+                    self.raise_geo_restricted()
+                raise ExtractorError(error, expected=True)
+            raise
+
+    def _extract(self, item_id, pc_method, mobile_method, referer=None):
+        pc = self._call_playlist_service(item_id, pc_method, referer=referer)
+        mobile = self._call_playlist_service(
+            item_id, mobile_method, fatal=False, referer=referer)
+        return pc, mobile
+
+    def _extract_info(self, pc, mobile, i, referer):
+        get_item = lambda x, y: try_get(x, lambda x: x[y][i], dict) or {}
+        pc_item = get_item(pc, 'playlistItems')
+        mobile_item = get_item(mobile, 'mediaList')
+        video_id = pc_item.get('mediaId') or mobile_item['mediaId']
+        title = pc_item.get('title') or mobile_item['title']
+
         formats = []
         urls = []
         formats = []
         urls = []
-        for stream in streams:
+        for stream in pc_item.get('streams', []):
             stream_url = stream.get('url')
             if not stream_url or stream.get('drmProtected') or stream_url in urls:
                 continue
             stream_url = stream.get('url')
             if not stream_url or stream.get('drmProtected') or stream_url in urls:
                 continue
@@ -48,13 +107,21 @@ class LimelightBaseIE(InfoExtractor):
                 fmt = {
                     'url': stream_url,
                     'abr': float_or_none(stream.get('audioBitRate')),
                 fmt = {
                     'url': stream_url,
                     'abr': float_or_none(stream.get('audioBitRate')),
-                    'vbr': float_or_none(stream.get('videoBitRate')),
                     'fps': float_or_none(stream.get('videoFrameRate')),
                     'fps': float_or_none(stream.get('videoFrameRate')),
-                    'width': int_or_none(stream.get('videoWidthInPixels')),
-                    'height': int_or_none(stream.get('videoHeightInPixels')),
                     'ext': ext,
                 }
                     'ext': ext,
                 }
-                rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url)
+                width = int_or_none(stream.get('videoWidthInPixels'))
+                height = int_or_none(stream.get('videoHeightInPixels'))
+                vbr = float_or_none(stream.get('videoBitRate'))
+                if width or height or vbr:
+                    fmt.update({
+                        'width': width,
+                        'height': height,
+                        'vbr': vbr,
+                    })
+                else:
+                    fmt['vcodec'] = 'none'
+                rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', stream_url)
                 if rtmp:
                     format_id = 'rtmp'
                     if stream.get('videoBitRate'):
                 if rtmp:
                     format_id = 'rtmp'
                     if stream.get('videoBitRate'):
@@ -88,7 +155,7 @@ class LimelightBaseIE(InfoExtractor):
                     })
                 formats.append(fmt)
 
                     })
                 formats.append(fmt)
 
-        for mobile_url in mobile_urls:
+        for mobile_url in mobile_item.get('mobileUrls', []):
             media_url = mobile_url.get('mobileUrl')
             format_id = mobile_url.get('targetMediaPlatform')
             if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls:
             media_url = mobile_url.get('mobileUrl')
             format_id = mobile_url.get('targetMediaPlatform')
             if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls:
@@ -112,45 +179,31 @@ class LimelightBaseIE(InfoExtractor):
 
         self._sort_formats(formats)
 
 
         self._sort_formats(formats)
 
-        title = properties['title']
-        description = properties.get('description')
-        timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
-        duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
-        filesize = int_or_none(properties.get('total_storage_in_bytes'))
-        categories = [properties.get('category')]
-        tags = properties.get('tags', [])
-        thumbnails = [{
-            'url': thumbnail['url'],
-            'width': int_or_none(thumbnail.get('width')),
-            'height': int_or_none(thumbnail.get('height')),
-        } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
-
         subtitles = {}
         subtitles = {}
-        for caption in properties.get('captions', []):
-            lang = caption.get('language_code')
-            subtitles_url = caption.get('url')
-            if lang and subtitles_url:
-                subtitles.setdefault(lang, []).append({
-                    'url': subtitles_url,
-                })
-        closed_captions_url = properties.get('closed_captions_url')
-        if closed_captions_url:
-            subtitles.setdefault('en', []).append({
-                'url': closed_captions_url,
-                'ext': 'ttml',
-            })
+        for flag in mobile_item.get('flags'):
+            if flag == 'ClosedCaptions':
+                closed_captions = self._call_playlist_service(
+                    video_id, 'getClosedCaptionsDetailsByMediaId',
+                    False, referer) or []
+                for cc in closed_captions:
+                    cc_url = cc.get('webvttFileUrl')
+                    if not cc_url:
+                        continue
+                    lang = cc.get('languageCode') or self._search_regex(r'/[a-z]{2}\.vtt', cc_url, 'lang', default='en')
+                    subtitles.setdefault(lang, []).append({
+                        'url': cc_url,
+                    })
+                break
+
+        get_meta = lambda x: pc_item.get(x) or mobile_item.get(x)
 
         return {
             'id': video_id,
             'title': title,
 
         return {
             'id': video_id,
             'title': title,
-            'description': description,
+            'description': get_meta('description'),
             'formats': formats,
             'formats': formats,
-            'timestamp': timestamp,
-            'duration': duration,
-            'filesize': filesize,
-            'categories': categories,
-            'tags': tags,
-            'thumbnails': thumbnails,
+            'duration': float_or_none(get_meta('durationInMilliseconds'), 1000),
+            'thumbnail': get_meta('previewImageUrl') or get_meta('thumbnailImageUrl'),
             'subtitles': subtitles,
         }
 
             'subtitles': subtitles,
         }
 
@@ -178,8 +231,6 @@ class LimelightMediaIE(LimelightBaseIE):
             'description': 'md5:8005b944181778e313d95c1237ddb640',
             'thumbnail': r're:^https?://.*\.jpeg$',
             'duration': 144.23,
             'description': 'md5:8005b944181778e313d95c1237ddb640',
             'thumbnail': r're:^https?://.*\.jpeg$',
             'duration': 144.23,
-            'timestamp': 1244136834,
-            'upload_date': '20090604',
         },
         'params': {
             # m3u8 download
         },
         'params': {
             # m3u8 download
@@ -195,27 +246,29 @@ class LimelightMediaIE(LimelightBaseIE):
             'title': '3Play Media Overview Video',
             'thumbnail': r're:^https?://.*\.jpeg$',
             'duration': 78.101,
             'title': '3Play Media Overview Video',
             'thumbnail': r're:^https?://.*\.jpeg$',
             'duration': 78.101,
-            'timestamp': 1338929955,
-            'upload_date': '20120605',
-            'subtitles': 'mincount:9',
+            # TODO: extract all languages that were accessible via API
+            # 'subtitles': 'mincount:9',
+            'subtitles': 'mincount:1',
         },
     }, {
         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
         'only_matching': True,
     }]
     _PLAYLIST_SERVICE_PATH = 'media'
         },
     }, {
         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
         'only_matching': True,
     }]
     _PLAYLIST_SERVICE_PATH = 'media'
-    _API_PATH = 'media'
 
     def _real_extract(self, url):
 
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
         video_id = self._match_id(url)
         video_id = self._match_id(url)
+        source_url = smuggled_data.get('source_url')
+        self._initialize_geo_bypass({
+            'countries': smuggled_data.get('geo_countries'),
+        })
 
 
-        pc, mobile, metadata = self._extract(
-            video_id, 'getPlaylistByMediaId', 'getMobilePlaylistByMediaId', 'properties')
+        pc, mobile = self._extract(
+            video_id, 'getPlaylistByMediaId',
+            'getMobilePlaylistByMediaId', source_url)
 
 
-        return self._extract_info(
-            pc['playlistItems'][0].get('streams', []),
-            mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
-            metadata)
+        return self._extract_info(pc, mobile, 0, source_url)
 
 
 class LimelightChannelIE(LimelightBaseIE):
 
 
 class LimelightChannelIE(LimelightBaseIE):
@@ -237,6 +290,7 @@ class LimelightChannelIE(LimelightBaseIE):
         'info_dict': {
             'id': 'ab6a524c379342f9b23642917020c082',
             'title': 'Javascript Sample Code',
         'info_dict': {
             'id': 'ab6a524c379342f9b23642917020c082',
             'title': 'Javascript Sample Code',
+            'description': 'Javascript Sample Code - http://www.delvenetworks.com/sample-code/playerCode-demo.html',
         },
         'playlist_mincount': 3,
     }, {
         },
         'playlist_mincount': 3,
     }, {
@@ -244,23 +298,23 @@ class LimelightChannelIE(LimelightBaseIE):
         'only_matching': True,
     }]
     _PLAYLIST_SERVICE_PATH = 'channel'
         'only_matching': True,
     }]
     _PLAYLIST_SERVICE_PATH = 'channel'
-    _API_PATH = 'channels'
 
     def _real_extract(self, url):
 
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
         channel_id = self._match_id(url)
         channel_id = self._match_id(url)
+        source_url = smuggled_data.get('source_url')
 
 
-        pc, mobile, medias = self._extract(
+        pc, mobile = self._extract(
             channel_id, 'getPlaylistByChannelId',
             channel_id, 'getPlaylistByChannelId',
-            'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1', 'media')
+            'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
+            source_url)
 
         entries = [
 
         entries = [
-            self._extract_info(
-                pc['playlistItems'][i].get('streams', []),
-                mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
-                medias['media_list'][i])
-            for i in range(len(medias['media_list']))]
+            self._extract_info(pc, mobile, i, source_url)
+            for i in range(len(pc['playlistItems']))]
 
 
-        return self.playlist_result(entries, channel_id, pc['title'])
+        return self.playlist_result(
+            entries, channel_id, pc.get('title'), mobile.get('description'))
 
 
 class LimelightChannelListIE(LimelightBaseIE):
 
 
 class LimelightChannelListIE(LimelightBaseIE):
@@ -293,10 +347,12 @@ class LimelightChannelListIE(LimelightBaseIE):
     def _real_extract(self, url):
         channel_list_id = self._match_id(url)
 
     def _real_extract(self, url):
         channel_list_id = self._match_id(url)
 
-        channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
+        channel_list = self._call_playlist_service(
+            channel_list_id, 'getMobileChannelListById')
 
         entries = [
             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
             for channel in channel_list['channelList']]
 
 
         entries = [
             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
             for channel in channel_list['channelList']]
 
-        return self.playlist_result(entries, channel_list_id, channel_list['title'])
+        return self.playlist_result(
+            entries, channel_list_id, channel_list['title'])