_ Git - youtube-dl/blob - youtube_dl/extractor/teamcoco.py

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