[amcnetworks] Fix free content extraction(closes #20354)

youtube_dl/extractor/amcnetworks.py

@@ -1,6 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .theplatform import ThePlatformIE
 from ..utils import (
     int_or_none,
@@ -11,25 +13,22 @@ from ..utils import (
 
 
 class AMCNetworksIE(ThePlatformIE):
-    _VALID_URL = r'https?://(?:www\.)?(?:amc|bbcamerica|ifc|(?:we|sundance)tv)\.com/(?:movies|shows(?:/[^/]+)+)/(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://(?:www\.)?(?P<site>amc|bbcamerica|ifc|(?:we|sundance)tv)\.com/(?P<id>(?:movies|shows(?:/[^/]+)+)/[^/?#&]+)'
     _TESTS = [{
-        'url': 'http://www.ifc.com/shows/maron/season-04/episode-01/step-1',
+        'url': 'https://www.bbcamerica.com/shows/the-graham-norton-show/videos/tina-feys-adorable-airline-themed-family-dinner--51631',
-        'md5': '',
         'info_dict': {
-            'id': 's3MX01Nl4vPH',
+            'id': '4Lq1dzOnZGt0',
             'ext': 'mp4',
-            'title': 'Maron - Season 4 - Step 1',
+            'title': "The Graham Norton Show - Season 28 - Tina Fey's Adorable Airline-Themed Family Dinner",
-            'description': 'In denial about his current situation, Marc is reluctantly convinced by his friends to enter rehab. Starring Marc Maron and Constance Zimmer.',
+            'description': "It turns out child stewardesses are very generous with the wine! All-new episodes of 'The Graham Norton Show' premiere Fridays at 11/10c on BBC America.",
-            'age_limit': 17,
+            'upload_date': '20201120',
-            'upload_date': '20160505',
+            'timestamp': 1605904350,
-            'timestamp': 1462468831,
             'uploader': 'AMCN',
         },
         'params': {
             # m3u8 download
             'skip_download': True,
         },
-        'skip': 'Requires TV provider accounts',
     }, {
         'url': 'http://www.bbcamerica.com/shows/the-hunt/full-episodes/season-1/episode-01-the-hardest-challenge',
         'only_matching': True,
@@ -55,32 +54,33 @@ class AMCNetworksIE(ThePlatformIE):
         'url': 'https://www.sundancetv.com/shows/riviera/full-episodes/season-1/episode-01-episode-1',
         'only_matching': True,
     }]
+    _REQUESTOR_ID_MAP = {
+        'amc': 'AMC',
+        'bbcamerica': 'BBCA',
+        'ifc': 'IFC',
+        'sundancetv': 'SUNDANCE',
+        'wetv': 'WETV',
+    }
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
+        site, display_id = re.match(self._VALID_URL, url).groups()
-        webpage = self._download_webpage(url, display_id)
+        requestor_id = self._REQUESTOR_ID_MAP[site]
+        properties = self._download_json(
+            'https://content-delivery-gw.svc.ds.amcn.com/api/v2/content/amcn/%s/url/%s' % (requestor_id.lower(), display_id),
+            display_id)['data']['properties']
         query = {
             'mbr': 'true',
             'manifest': 'm3u',
         }
-        media_url = self._search_regex(
+        tp_path = 'M_UwQC/media/' + properties['videoPid']
-            r'window\.platformLinkURL\s*=\s*[\'"]([^\'"]+)',
+        media_url = 'https://link.theplatform.com/s/' + tp_path
-            webpage, 'media url')
+        theplatform_metadata = self._download_theplatform_metadata(tp_path, display_id)
-        theplatform_metadata = self._download_theplatform_metadata(self._search_regex(
-            r'link\.theplatform\.com/s/([^?]+)',
-            media_url, 'theplatform_path'), display_id)
         info = self._parse_theplatform_metadata(theplatform_metadata)
         video_id = theplatform_metadata['pid']
         title = theplatform_metadata['title']
         rating = try_get(
             theplatform_metadata, lambda x: x['ratings'][0]['rating'])
-        auth_required = self._search_regex(
+        if properties.get('videoCategory') == 'TVE-Auth':
-            r'window\.authRequired\s*=\s*(true|false);',
-            webpage, 'auth required')
-        if auth_required == 'true':
-            requestor_id = self._search_regex(
-                r'window\.requestor_id\s*=\s*[\'"]([^\'"]+)',
-                webpage, 'requestor id')
             resource = self._get_mvpd_resource(
                 requestor_id, title, video_id, rating)
             query['auth'] = self._extract_mvpd_auth(

youtube_dl/extractor/extractors.py

@@ -534,7 +534,10 @@ from .laola1tv import (
     EHFTVIE,
     ITTFIE,
 )
-from .lbry import LBRYIE
+from .lbry import (
+    LBRYIE,
+    LBRYChannelIE,
+)
 from .lci import LCIIE
 from .lcp import (
     LcpPlayIE,

youtube_dl/extractor/lbry.py

@@ -1,6 +1,7 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import functools
 import json
 
 from .common import InfoExtractor
@@ -10,13 +11,73 @@ from ..utils import (
     ExtractorError,
     int_or_none,
     mimetype2ext,
+    OnDemandPagedList,
     try_get,
+    urljoin,
 )
 
 
-class LBRYIE(InfoExtractor):
+class LBRYBaseIE(InfoExtractor):
-    IE_NAME = 'lbry.tv'
+    _BASE_URL_REGEX = r'https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/'
-    _VALID_URL = r'https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/(?P<id>@[^:]+:[0-9a-z]+/[^:]+:[0-9a-z])'
+    _CLAIM_ID_REGEX = r'[0-9a-f]{1,40}'
+    _OPT_CLAIM_ID = '[^:/?#&]+(?::%s)?' % _CLAIM_ID_REGEX
+    _SUPPORTED_STREAM_TYPES = ['video', 'audio']
+
+    def _call_api_proxy(self, method, display_id, params, resource):
+        return self._download_json(
+            'https://api.lbry.tv/api/v1/proxy',
+            display_id, 'Downloading %s JSON metadata' % resource,
+            headers={'Content-Type': 'application/json-rpc'},
+            data=json.dumps({
+                'method': method,
+                'params': params,
+            }).encode())['result']
+
+    def _resolve_url(self, url, display_id, resource):
+        return self._call_api_proxy(
+            'resolve', display_id, {'urls': url}, resource)[url]
+
+    def _permanent_url(self, url, claim_name, claim_id):
+        return urljoin(url, '/%s:%s' % (claim_name, claim_id))
+
+    def _parse_stream(self, stream, url):
+        stream_value = stream.get('value') or {}
+        stream_type = stream_value.get('stream_type')
+        source = stream_value.get('source') or {}
+        media = stream_value.get(stream_type) or {}
+        signing_channel = stream.get('signing_channel') or {}
+        channel_name = signing_channel.get('name')
+        channel_claim_id = signing_channel.get('claim_id')
+        channel_url = None
+        if channel_name and channel_claim_id:
+            channel_url = self._permanent_url(url, channel_name, channel_claim_id)
+
+        info = {
+            'thumbnail': try_get(stream_value, lambda x: x['thumbnail']['url'], compat_str),
+            'description': stream_value.get('description'),
+            'license': stream_value.get('license'),
+            'timestamp': int_or_none(stream.get('timestamp')),
+            'tags': stream_value.get('tags'),
+            'duration': int_or_none(media.get('duration')),
+            'channel': try_get(signing_channel, lambda x: x['value']['title']),
+            'channel_id': channel_claim_id,
+            'channel_url': channel_url,
+            'ext': determine_ext(source.get('name')) or mimetype2ext(source.get('media_type')),
+            'filesize': int_or_none(source.get('size')),
+        }
+        if stream_type == 'audio':
+            info['vcodec'] = 'none'
+        else:
+            info.update({
+                'width': int_or_none(media.get('width')),
+                'height': int_or_none(media.get('height')),
+            })
+        return info
+
+
+class LBRYIE(LBRYBaseIE):
+    IE_NAME = 'lbry'
+    _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>\$/[^/]+/[^/]+/{1}|@{0}/{0}|(?!@){0})'.format(LBRYBaseIE._OPT_CLAIM_ID, LBRYBaseIE._CLAIM_ID_REGEX)
     _TESTS = [{
         # Video
         'url': 'https://lbry.tv/@Mantega:1/First-day-LBRY:1',
@@ -28,6 +89,8 @@ class LBRYIE(InfoExtractor):
             'description': 'md5:f6cb5c704b332d37f5119313c2c98f51',
             'timestamp': 1595694354,
             'upload_date': '20200725',
+            'width': 1280,
+            'height': 720,
         }
     }, {
         # Audio
@@ -40,6 +103,12 @@ class LBRYIE(InfoExtractor):
             'description': 'md5:661ac4f1db09f31728931d7b88807a61',
             'timestamp': 1591312601,
             'upload_date': '20200604',
+            'tags': list,
+            'duration': 2570,
+            'channel': 'The LBRY Foundation',
+            'channel_id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
+            'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212',
+            'vcodec': 'none',
         }
     }, {
         'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e',
@@ -47,45 +116,99 @@ class LBRYIE(InfoExtractor):
     }, {
         'url': "https://odysee.com/@ScammerRevolts:b0/I-SYSKEY'D-THE-SAME-SCAMMERS-3-TIMES!:b",
         'only_matching': True,
+    }, {
+        'url': 'https://lbry.tv/Episode-1:e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
+        'only_matching': True,
+    }, {
+        'url': 'https://lbry.tv/$/embed/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
+        'only_matching': True,
+    }, {
+        'url': 'https://lbry.tv/Episode-1:e7',
+        'only_matching': True,
+    }, {
+        'url': 'https://lbry.tv/@LBRYFoundation/Episode-1',
+        'only_matching': True,
+    }, {
+        'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
+        'only_matching': True,
     }]
 
-    def _call_api_proxy(self, method, display_id, params):
+    def _real_extract(self, url):
-        return self._download_json(
+        display_id = self._match_id(url)
-            'https://api.lbry.tv/api/v1/proxy', display_id,
+        if display_id.startswith('$/'):
-            headers={'Content-Type': 'application/json-rpc'},
+            display_id = display_id.split('/', 2)[-1].replace('/', ':')
-            data=json.dumps({
+        else:
-                'method': method,
+            display_id = display_id.replace(':', '#')
-                'params': params,
+        uri = 'lbry://' + display_id
-            }).encode())['result']
+        result = self._resolve_url(uri, display_id, 'stream')
+        result_value = result['value']
+        if result_value.get('stream_type') not in self._SUPPORTED_STREAM_TYPES:
+            raise ExtractorError('Unsupported URL', expected=True)
+        claim_id = result['claim_id']
+        title = result_value['title']
+        streaming_url = self._call_api_proxy(
+            'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url']
+        info = self._parse_stream(result, url)
+        info.update({
+            'id': claim_id,
+            'title': title,
+            'url': streaming_url,
+        })
+        return info
+
+
+class LBRYChannelIE(LBRYBaseIE):
+    IE_NAME = 'lbry:channel'
+    _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?#&]|$)' % LBRYBaseIE._OPT_CLAIM_ID
+    _TESTS = [{
+        'url': 'https://lbry.tv/@LBRYFoundation:0',
+        'info_dict': {
+            'id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
+            'title': 'The LBRY Foundation',
+            'description': 'Channel for the LBRY Foundation. Follow for updates and news.',
+        },
+        'playlist_count': 29,
+    }, {
+        'url': 'https://lbry.tv/@LBRYFoundation',
+        'only_matching': True,
+    }]
+    _PAGE_SIZE = 50
+
+    def _fetch_page(self, claim_id, url, page):
+        page += 1
+        result = self._call_api_proxy(
+            'claim_search', claim_id, {
+                'channel_ids': [claim_id],
+                'claim_type': 'stream',
+                'no_totals': True,
+                'page': page,
+                'page_size': self._PAGE_SIZE,
+                'stream_types': self._SUPPORTED_STREAM_TYPES,
+            }, 'page %d' % page)
+        for item in (result.get('items') or []):
+            stream_claim_name = item.get('name')
+            stream_claim_id = item.get('claim_id')
+            if not (stream_claim_name and stream_claim_id):
+                continue
+
+            info = self._parse_stream(item, url)
+            info.update({
+                '_type': 'url',
+                'id': stream_claim_id,
+                'title': try_get(item, lambda x: x['value']['title']),
+                'url': self._permanent_url(url, stream_claim_name, stream_claim_id),
+            })
+            yield info
 
     def _real_extract(self, url):
         display_id = self._match_id(url).replace(':', '#')
-        uri = 'lbry://' + display_id
+        result = self._resolve_url(
-        result = self._call_api_proxy(
+            'lbry://' + display_id, display_id, 'channel')
-            'resolve', display_id, {'urls': [uri]})[uri]
+        claim_id = result['claim_id']
-        result_value = result['value']
+        entries = OnDemandPagedList(
-        if result_value.get('stream_type') not in ('video', 'audio'):
+            functools.partial(self._fetch_page, claim_id, url),
-            raise ExtractorError('Unsupported URL', expected=True)
+            self._PAGE_SIZE)
-        streaming_url = self._call_api_proxy(
+        result_value = result.get('value') or {}
-            'get', display_id, {'uri': uri})['streaming_url']
+        return self.playlist_result(
-        source = result_value.get('source') or {}
+            entries, claim_id, result_value.get('title'),
-        media = result_value.get('video') or result_value.get('audio') or {}
+            result_value.get('description'))
-        signing_channel = result_value.get('signing_channel') or {}
-
-        return {
-            'id': result['claim_id'],
-            'title': result_value['title'],
-            'thumbnail': try_get(result_value, lambda x: x['thumbnail']['url'], compat_str),
-            'description': result_value.get('description'),
-            'license': result_value.get('license'),
-            'timestamp': int_or_none(result.get('timestamp')),
-            'tags': result_value.get('tags'),
-            'width': int_or_none(media.get('width')),
-            'height': int_or_none(media.get('height')),
-            'duration': int_or_none(media.get('duration')),
-            'channel': signing_channel.get('name'),
-            'channel_id': signing_channel.get('claim_id'),
-            'ext': determine_ext(source.get('name')) or mimetype2ext(source.get('media_type')),
-            'filesize': int_or_none(source.get('size')),
-            'url': streaming_url,
-        }