[vgtv] Add support for generic bt.no URLs (#5620)

[youtube-dl] / youtube_dl / extractor / teamcoco.py

diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py
index 41677503c7a3f5d1530d4a4d8b8e2fc8631e28a2..95d58ddd0429e84abeef9f6dc3dea74f011318f9 100644 (file)
--- a/youtube_dl/extractor/teamcoco.py
+++ b/youtube_dl/extractor/teamcoco.py
@@ -1,6 +1,8 @@
+# -*- coding: utf-8 -*-
 from __future__ import unicode_literals
 
 import base64
+import binascii
 import re
 
 from .common import InfoExtractor
@@ -8,6 +10,7 @@ from ..utils import (
     ExtractorError,
     qualities,
 )
+from ..compat import compat_ord
 
 
 class TeamcocoIE(InfoExtractor):
@@ -21,6 +24,7 @@ class TeamcocoIE(InfoExtractor):
                 '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,
             }
         }, {
@@ -31,8 +35,20 @@ class TeamcocoIE(InfoExtractor):
                 'ext': 'mp4',
                 '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,
             }
+        }, {
+            'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',
+            'info_dict': {
+                'id': '88748',
+                'ext': 'mp4',
+                'title': 'Timothy Olyphant Raises A Toast To “Justified”',
+                'description': 'md5:15501f23f020e793aeca761205e42c24',
+            },
+            'params': {
+                'skip_download': True,  # m3u8 downloads
+            }
         }
     ]
     _VIDEO_ID_REGEXES = (
@@ -52,12 +68,41 @@ class TeamcocoIE(InfoExtractor):
             video_id = self._html_search_regex(
                 self._VIDEO_ID_REGEXES, webpage, 'video id')
 
+        data = preload = None
         preloads = re.findall(r'"preload":\s*"([^"]+)"', webpage)
-        if not preloads:
-            raise ExtractorError('Preload information could not be extracted')
-        preload = max([(len(p), p) for p in preloads])[1]
-        data = self._parse_json(
-            base64.b64decode(preload.encode('ascii')).decode('utf-8'), video_id)
+        if preloads:
+            preload = max([(len(p), p) for p in preloads])[1]
+
+        if not preload:
+            preload = ''.join(re.findall(r'this\.push\("([^"]+)"\);', webpage))
+
+        if not preload:
+            preload = self._html_search_regex([
+                r'player,\[?"([^"]+)"\]?', r'player.init\(\[?"([^"]+)"\]?\)'
+            ], webpage.replace('","', ''), 'preload data', default=None)
+
+        if not preload:
+            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')
+            for i in range(len(base64_fragments)):
+                cur_sequence = (''.join(base64_fragments[i:] + base64_fragments[:i])).encode('ascii')
+                try:
+                    raw_data = base64.b64decode(cur_sequence)
+                except (TypeError, binascii.Error):
+                    continue
+                if compat_ord(raw_data[0]) == compat_ord('{'):
+                    data = self._parse_json(raw_data.decode('utf-8'), video_id, fatal=False)
+
+        if not preload and not data:
+            raise ExtractorError(
+                'Preload information could not be extracted', expected=True)
+
+        if not data:
+            data = self._parse_json(
+                base64.b64decode(preload.encode('ascii')).decode('utf-8'), video_id)
 
         formats = []
         get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
@@ -93,5 +138,6 @@ class TeamcocoIE(InfoExtractor):
             'title': data['title'],
             'thumbnail': data.get('thumb', {}).get('href'),
             'description': data.get('teaser'),
+            'duration': data.get('duration'),
             'age_limit': self._family_friendly_search(webpage),
         }