youtube_dl/extractor/teamcoco.py

   1 # -*- coding: utf-8 -*-
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import binascii
   6 import re
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     ExtractorError,
  11     qualities,
  12 )
  13 from ..compat import compat_ord
  14
  15
  16 class TeamcocoIE(InfoExtractor):
  17     _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
  18     _TESTS = [
  19         {
  20             'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
  21             'md5': '3f7746aa0dc86de18df7539903d399ea',
  22             'info_dict': {
  23                 'id': '80187',
  24                 'ext': 'mp4',
  25                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
  26                 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.',
  27                 'duration': 504,
  28                 'age_limit': 0,
  29             }
  30         }, {
  31             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
  32             'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
  33             'info_dict': {
  34                 'id': '19705',
  35                 'ext': 'mp4',
  36                 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
  37                 'title': 'Louis C.K. Interview Pt. 1 11/3/11',
  38                 'duration': 288,
  39                 'age_limit': 0,
  40             }
  41         }, {
  42             'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',
  43             'info_dict': {
  44                 'id': '88748',
  45                 'ext': 'mp4',
  46                 'title': 'Timothy Olyphant Raises A Toast To “Justified”',
  47                 'description': 'md5:15501f23f020e793aeca761205e42c24',
  48             },
  49             'params': {
  50                 'skip_download': True,  # m3u8 downloads
  51             }
  52         }
  53     ]
  54     _VIDEO_ID_REGEXES = (
  55         r'"eVar42"\s*:\s*(\d+)',
  56         r'Ginger\.TeamCoco\.openInApp\("video",\s*"([^"]+)"',
  57         r'"id_not"\s*:\s*(\d+)'
  58     )
  59
  60     def _real_extract(self, url):
  61         mobj = re.match(self._VALID_URL, url)
  62
  63         display_id = mobj.group('display_id')
  64         webpage = self._download_webpage(url, display_id)
  65
  66         video_id = mobj.group('video_id')
  67         if not video_id:
  68             video_id = self._html_search_regex(
  69                 self._VIDEO_ID_REGEXES, webpage, 'video id')
  70
  71         data = preload = None
  72         preloads = re.findall(r'"preload":\s*"([^"]+)"', webpage)
  73         if preloads:
  74             preload = max([(len(p), p) for p in preloads])[1]
  75
  76         if not preload:
  77             preload = ''.join(re.findall(r'this\.push\("([^"]+)"\);', webpage))
  78
  79         if not preload:
  80             preload = self._html_search_regex([
  81                 r'player,\[?"([^"]+)"\]?', r'player.init\(\[?"([^"]+)"\]?\)'
  82             ], webpage.replace('","', ''), 'preload data', default=None)
  83
  84         if not preload:
  85             preload_codes = self._html_search_regex(
  86                 r'(function.+)setTimeout\(function\(\)\{playlist',
  87                 webpage, 'preload codes')
  88             base64_fragments = re.findall(r'"([a-zA-z0-9+/=]+)"', preload_codes)
  89             base64_fragments.remove('init')
  90             for i in range(len(base64_fragments)):
  91                 cur_sequence = (''.join(base64_fragments[i:] + base64_fragments[:i])).encode('ascii')
  92                 try:
  93                     raw_data = base64.b64decode(cur_sequence)
  94                 except (TypeError, binascii.Error):
  95                     continue
  96                 if compat_ord(raw_data[0]) == compat_ord('{'):
  97                     data = self._parse_json(raw_data.decode('utf-8'), video_id, fatal=False)
  98
  99         if not preload and not data:
 100             raise ExtractorError(
 101                 'Preload information could not be extracted', expected=True)
 102
 103         if not data:
 104             data = self._parse_json(
 105                 base64.b64decode(preload.encode('ascii')).decode('utf-8'), video_id)
 106
 107         formats = []
 108         get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
 109         for filed in data['files']:
 110             if filed['type'] == 'hls':
 111                 formats.extend(self._extract_m3u8_formats(
 112                     filed['url'], video_id, ext='mp4'))
 113             else:
 114                 m_format = re.search(r'(\d+(k|p))\.mp4', filed['url'])
 115                 if m_format is not None:
 116                     format_id = m_format.group(1)
 117                 else:
 118                     format_id = filed['bitrate']
 119                 tbr = (
 120                     int(filed['bitrate'])
 121                     if filed['bitrate'].isdigit()
 122                     else None)
 123
 124                 formats.append({
 125                     'url': filed['url'],
 126                     'ext': 'mp4',
 127                     'tbr': tbr,
 128                     'format_id': format_id,
 129                     'quality': get_quality(format_id),
 130                 })
 131
 132         self._sort_formats(formats)
 133
 134         return {
 135             'id': video_id,
 136             'display_id': display_id,
 137             'formats': formats,
 138             'title': data['title'],
 139             'thumbnail': data.get('thumb', {}).get('href'),
 140             'description': data.get('teaser'),
 141             'duration': data.get('duration'),
 142             'age_limit': self._family_friendly_search(webpage),
 143         }