]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/teamcoco.py
New upstream version 2018.04.25
[youtubedl] / youtube_dl / extractor / teamcoco.py
index 56be526383b590d9b00759400a269d5164beef2f..9056c8cbc29a4431a364be6b3731f5aef2f06529 100644 (file)
@@ -1,21 +1,24 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
-import base64
 import binascii
 import re
 import json
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_b64decode,
+    compat_ord,
+)
 from ..utils import (
     ExtractorError,
     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>.*)'
+    _VALID_URL = r'https?://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
     _TESTS = [
         {
             'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
@@ -50,6 +53,17 @@ class TeamcocoIE(InfoExtractor):
             '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
+            }
         }
     ]
     _VIDEO_ID_REGEXES = (
@@ -76,7 +90,7 @@ class TeamcocoIE(InfoExtractor):
         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 = re.findall(r'"([a-zA-Z0-9+/=]+)"', preload_codes)
         base64_fragments.remove('init')
 
         def _check_sequence(cur_fragments):
@@ -85,7 +99,7 @@ class TeamcocoIE(InfoExtractor):
             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)
+                    raw_data = compat_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):
@@ -108,10 +122,24 @@ class TeamcocoIE(InfoExtractor):
         formats = []
         get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
         for filed in data['files']:
-            if filed['type'] == 'hls':
-                formats.extend(self._extract_m3u8_formats(
-                    filed['url'], video_id, ext='mp4'))
+            if determine_ext(filed['url']) == 'm3u8':
+                # compat_urllib_parse.urljoin does not work here
+                if filed['url'].startswith('/'):
+                    m3u8_url = 'http://ht.cdn.turner.com/tbs/big/teamcoco' + filed['url']
+                else:
+                    m3u8_url = filed['url']
+                m3u8_formats = self._extract_m3u8_formats(
+                    m3u8_url, video_id, ext='mp4')
+                for m3u8_format in m3u8_formats:
+                    if m3u8_format not in formats:
+                        formats.append(m3u8_format)
+            elif determine_ext(filed['url']) == 'f4m':
+                # TODO Correct f4m extraction
+                continue
             else:
+                if filed['url'].startswith('/mp4:protected/'):
+                    # TODO Correct extraction for these files
+                    continue
                 m_format = re.search(r'(\d+(k|p))\.mp4', filed['url'])
                 if m_format is not None:
                     format_id = m_format.group(1)