d/rules: Run offline tests.

[youtubedl] / youtube_dl / extractor / teamcoco.py

diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py
index b2a4b1fc05430558ad9b33a9aa3ce834107dc6e3..7640cf00a1d87bc2582e355899fca3e017babc69 100644 (file)
--- a/youtube_dl/extractor/teamcoco.py
+++ b/youtube_dl/extractor/teamcoco.py
@@ -1,33 +1,34 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8

 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

-import base64
-import binascii
-import re

 import json
 
 import json
 

-from .common import InfoExtractor
+from .turner import TurnerBaseIE

 from ..utils import (
 from ..utils import (

+    determine_ext,

     ExtractorError,
     ExtractorError,

+    int_or_none,
+    mimetype2ext,
+    parse_duration,
+    parse_iso8601,

     qualities,
     qualities,

-    determine_ext,

 )
 )

-from ..compat import compat_ord

 
 
 
 

-class TeamcocoIE(InfoExtractor):
-    _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
+class TeamcocoIE(TurnerBaseIE):
+    _VALID_URL = r'https?://(?:\w+\.)?teamcoco\.com/(?P<id>([^/]+/)*[^/?#]+)'

     _TESTS = [
         {
     _TESTS = [
         {

-            'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
-            'md5': '3f7746aa0dc86de18df7539903d399ea',
+            'url': 'http://teamcoco.com/video/mary-kay-remote',
+            'md5': '55d532f81992f5c92046ad02fec34d7d',

             'info_dict': {
                 'id': '80187',
                 'ext': 'mp4',
                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
                 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.',
             'info_dict': {
                 'id': '80187',
                 'ext': 'mp4',
                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
                 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.',

-                'duration': 504,
-                'age_limit': 0,
+                'duration': 495.0,
+                'upload_date': '20140402',
+                'timestamp': 1396407600,

             }
         }, {
             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
             }
         }, {
             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',


@@ -38,7 +39,8 @@ class TeamcocoIE(InfoExtractor):
                 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
                 'title': 'Louis C.K. Interview Pt. 1 11/3/11',
                 'duration': 288,
                 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
                 'title': 'Louis C.K. Interview Pt. 1 11/3/11',
                 'duration': 288,

-                'age_limit': 0,
+                'upload_date': '20111104',
+                'timestamp': 1320405840,

             }
         }, {
             'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',
             }
         }, {
             'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',


@@ -47,99 +49,151 @@ class TeamcocoIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Timothy Olyphant Raises A Toast To “Justified”',
                 'description': 'md5:15501f23f020e793aeca761205e42c24',
                 'ext': 'mp4',
                 'title': 'Timothy Olyphant Raises A Toast To “Justified”',
                 'description': 'md5:15501f23f020e793aeca761205e42c24',

+                'upload_date': '20150415',
+                'timestamp': 1429088400,

             },
             'params': {
                 'skip_download': True,  # m3u8 downloads
             }
             },
             'params': {
                 'skip_download': True,  # m3u8 downloads
             }

+        }, {
+            'url': 'http://teamcoco.com/video/full-episode-mon-6-1-joel-mchale-jake-tapper-and-musical-guest-courtney-barnett?playlist=x;eyJ0eXBlIjoidGFnIiwiaWQiOjl9',
+            'info_dict': {
+                'id': '89341',
+                'ext': 'mp4',
+                'title': 'Full Episode - Mon. 6/1 - Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett',
+                'description': 'Guests: Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett',
+            },
+            'params': {
+                'skip_download': True,  # m3u8 downloads
+            },
+            'skip': 'This video is no longer available.',
+        }, {
+            'url': 'http://teamcoco.com/video/the-conan-audiencey-awards-for-04/25/18',
+            'only_matching': True,
+        }, {
+            'url': 'http://teamcoco.com/italy/conan-jordan-schlansky-hit-the-streets-of-florence',
+            'only_matching': True,
+        }, {
+            'url': 'http://teamcoco.com/haiti/conan-s-haitian-history-lesson',
+            'only_matching': True,
+        }, {
+            'url': 'http://teamcoco.com/israel/conan-hits-the-streets-beaches-of-tel-aviv',
+            'only_matching': True,
+        }, {
+            'url': 'https://conan25.teamcoco.com/video/ice-cube-kevin-hart-conan-share-lyft',
+            'only_matching': True,

         }
     ]
         }
     ]

-    _VIDEO_ID_REGEXES = (
-        r'"eVar42"\s*:\s*(\d+)',
-        r'Ginger\.TeamCoco\.openInApp\("video",\s*"([^"]+)"',
-        r'"id_not"\s*:\s*(\d+)'
-    )
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-
-        display_id = mobj.group('display_id')
-        webpage, urlh = self._download_webpage_handle(url, display_id)
-        if 'src=expired' in urlh.geturl():
-            raise ExtractorError('This video is expired.', expected=True)
-
-        video_id = mobj.group('video_id')
-        if not video_id:
-            video_id = self._html_search_regex(
-                self._VIDEO_ID_REGEXES, webpage, 'video id')

 
 

-        data = None
+    def _graphql_call(self, query_template, object_type, object_id):
+        find_object = 'find' + object_type
+        return self._download_json(
+            'https://teamcoco.com/graphql', object_id, data=json.dumps({
+                'query': query_template % (find_object, object_id)
+            }).encode(), headers={
+                'Content-Type': 'application/json',
+            })['data'][find_object]

 
 

-        preload_codes = self._html_search_regex(
-            r'(function.+)setTimeout\(function\(\)\{playlist',
-            webpage, 'preload codes')
-        base64_fragments = re.findall(r'"([a-zA-z0-9+/=]+)"', preload_codes)
-        base64_fragments.remove('init')
-
-        def _check_sequence(cur_fragments):
-            if not cur_fragments:
-                return
-            for i in range(len(cur_fragments)):
-                cur_sequence = (''.join(cur_fragments[i:] + cur_fragments[:i])).encode('ascii')
-                try:
-                    raw_data = base64.b64decode(cur_sequence)
-                    if compat_ord(raw_data[0]) == compat_ord('{'):
-                        return json.loads(raw_data.decode('utf-8'))
-                except (TypeError, binascii.Error, UnicodeDecodeError, ValueError):
-                    continue
-
-        def _check_data():
-            for i in range(len(base64_fragments) + 1):
-                for j in range(i, len(base64_fragments) + 1):
-                    data = _check_sequence(base64_fragments[:i] + base64_fragments[j:])
-                    if data:
-                        return data
-
-        self.to_screen('Try to compute possible data sequence. This may take some time.')
-        data = _check_data()
-
-        if not data:
-            raise ExtractorError(
-                'Preload information could not be extracted', expected=True)
-
-        formats = []
-        get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
-        for filed in data['files']:
-            if determine_ext(filed['url']) == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    filed['url'], video_id, ext='mp4'))
-            else:
-                m_format = re.search(r'(\d+(k|p))\.mp4', filed['url'])
-                if m_format is not None:
-                    format_id = m_format.group(1)
-                else:
-                    format_id = filed['bitrate']
-                tbr = (
-                    int(filed['bitrate'])
-                    if filed['bitrate'].isdigit()
-                    else None)
-
-                formats.append({
-                    'url': filed['url'],
-                    'ext': 'mp4',
-                    'tbr': tbr,
-                    'format_id': format_id,
-                    'quality': get_quality(format_id),
-                })
-
-        self._sort_formats(formats)
-
-        return {
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        response = self._graphql_call('''{
+  %s(slug: "%s") {
+    ... on RecordSlug {
+      record {
+        id
+        title
+        teaser
+        publishOn
+        thumb {
+          preview
+        }
+        file {
+          url
+        }
+        tags {
+          name
+        }
+        duration
+        turnerMediaId
+        turnerMediaAuthToken
+      }
+    }
+    ... on NotFoundSlug {
+      status
+    }
+  }
+}''', 'Slug', display_id)
+        if response.get('status'):
+            raise ExtractorError('This video is no longer available.', expected=True)
+
+        record = response['record']
+        video_id = record['id']
+
+        info = {

             'id': video_id,
             'display_id': display_id,
             'id': video_id,
             'display_id': display_id,

-            'formats': formats,
-            'title': data['title'],
-            'thumbnail': data.get('thumb', {}).get('href'),
-            'description': data.get('teaser'),
-            'duration': data.get('duration'),
-            'age_limit': self._family_friendly_search(webpage),
+            'title': record['title'],
+            'thumbnail': record.get('thumb', {}).get('preview'),
+            'description': record.get('teaser'),
+            'duration': parse_duration(record.get('duration')),
+            'timestamp': parse_iso8601(record.get('publishOn')),

         }
         }

+
+        media_id = record.get('turnerMediaId')
+        if media_id:
+            self._initialize_geo_bypass({
+                'countries': ['US'],
+            })
+            info.update(self._extract_ngtv_info(media_id, {
+                'accessToken': record['turnerMediaAuthToken'],
+                'accessTokenType': 'jws',
+            }))
+        else:
+            d = self._download_json(
+                'https://teamcoco.com/_truman/d/' + video_id,
+                video_id, fatal=False) or {}
+            video_sources = d.get('meta') or {}
+            if not video_sources:
+                video_sources = self._graphql_call('''{
+  %s(id: "%s") {
+    src
+  }
+}''', 'RecordVideoSource', video_id) or {}
+
+            formats = []
+            get_quality = qualities(['low', 'sd', 'hd', 'uhd'])
+            for format_id, src in video_sources.get('src', {}).items():
+                if not isinstance(src, dict):
+                    continue
+                src_url = src.get('src')
+                if not src_url:
+                    continue
+                ext = determine_ext(src_url, mimetype2ext(src.get('type')))
+                if format_id == 'hls' or ext == 'm3u8':
+                    # compat_urllib_parse.urljoin does not work here
+                    if src_url.startswith('/'):
+                        src_url = 'http://ht.cdn.turner.com/tbs/big/teamcoco' + src_url
+                    formats.extend(self._extract_m3u8_formats(
+                        src_url, video_id, 'mp4', m3u8_id=format_id, fatal=False))
+                else:
+                    if src_url.startswith('/mp4:protected/'):
+                        # TODO Correct extraction for these files
+                        continue
+                    tbr = int_or_none(self._search_regex(
+                        r'(\d+)k\.mp4', src_url, 'tbr', default=None))
+
+                    formats.append({
+                        'url': src_url,
+                        'ext': ext,
+                        'tbr': tbr,
+                        'format_id': format_id,
+                        'quality': get_quality(format_id),
+                    })
+            if not formats:
+                formats = self._extract_m3u8_formats(
+                    record['file']['url'], video_id, 'mp4', fatal=False)
+            self._sort_formats(formats)
+            info['formats'] = formats
+
+        return info