Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bliptv.py

   1 from __future__ import unicode_literals
   2
   3 import datetime
   4 import json
   5 import re
   6 import socket
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     compat_http_client,
  11     compat_str,
  12     compat_urllib_error,
  13     compat_urllib_request,
  14
  15     ExtractorError,
  16     unescapeHTML,
  17 )
  18
  19
  20 class BlipTVIE(InfoExtractor):
  21     """Information extractor for blip.tv"""
  22
  23     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
  24
  25     _TEST = {
  26         'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
  27         'file': '5779306.mov',
  28         'md5': 'c6934ad0b6acf2bd920720ec888eb812',
  29         'info_dict': {
  30             'upload_date': '20111205',
  31             'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
  32             'uploader': 'Comic Book Resources - CBR TV',
  33             'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
  34         }
  35     }
  36
  37     def report_direct_download(self, title):
  38         """Report information extraction."""
  39         self.to_screen('%s: Direct download detected' % title)
  40
  41     def _real_extract(self, url):
  42         mobj = re.match(self._VALID_URL, url)
  43         if mobj is None:
  44             raise ExtractorError('Invalid URL: %s' % url)
  45
  46         # See https://github.com/rg3/youtube-dl/issues/857
  47         embed_mobj = re.search(r'^(?:https?://)?(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
  48         if embed_mobj:
  49             info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
  50             info_page = self._download_webpage(info_url, embed_mobj.group(1))
  51             video_id = self._search_regex(r'data-episode-id="(\d+)', info_page,  'video_id')
  52             return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
  53
  54         if '?' in url:
  55             cchar = '&'
  56         else:
  57             cchar = '?'
  58         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
  59         request = compat_urllib_request.Request(json_url)
  60         request.add_header('User-Agent', 'iTunes/10.6.1')
  61         self.report_extraction(mobj.group(1))
  62         urlh = self._request_webpage(request, None, False,
  63             'unable to download video info webpage')
  64
  65         try:
  66             json_code_bytes = urlh.read()
  67             json_code = json_code_bytes.decode('utf-8')
  68         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  69             raise ExtractorError('Unable to read video info webpage: %s' % compat_str(err))
  70
  71         try:
  72             json_data = json.loads(json_code)
  73             if 'Post' in json_data:
  74                 data = json_data['Post']
  75             else:
  76                 data = json_data
  77
  78             upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
  79             formats = []
  80             if 'additionalMedia' in data:
  81                 for f in sorted(data['additionalMedia'], key=lambda f: int(f['media_height'])):
  82                     if not int(f['media_width']): # filter m3u8
  83                         continue
  84                     formats.append({
  85                         'url': f['url'],
  86                         'format_id': f['role'],
  87                         'width': int(f['media_width']),
  88                         'height': int(f['media_height']),
  89                     })
  90             else:
  91                 formats.append({
  92                     'url': data['media']['url'],
  93                     'width': int(data['media']['width']),
  94                     'height': int(data['media']['height']),
  95                 })
  96
  97             self._sort_formats(formats)
  98
  99             return {
 100                 'id': compat_str(data['item_id']),
 101                 'uploader': data['display_name'],
 102                 'upload_date': upload_date,
 103                 'title': data['title'],
 104                 'thumbnail': data['thumbnailUrl'],
 105                 'description': data['description'],
 106                 'user_agent': 'iTunes/10.6.1',
 107                 'formats': formats,
 108             }
 109         except (ValueError, KeyError) as err:
 110             raise ExtractorError('Unable to parse video information: %s' % repr(err))
 111
 112
 113 class BlipTVUserIE(InfoExtractor):
 114     """Information Extractor for blip.tv users."""
 115
 116     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
 117     _PAGE_SIZE = 12
 118     IE_NAME = 'blip.tv:user'
 119
 120     def _real_extract(self, url):
 121         # Extract username
 122         mobj = re.match(self._VALID_URL, url)
 123         if mobj is None:
 124             raise ExtractorError('Invalid URL: %s' % url)
 125
 126         username = mobj.group(1)
 127
 128         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 129
 130         page = self._download_webpage(url, username, 'Downloading user page')
 131         mobj = re.search(r'data-users-id="([^"]+)"', page)
 132         page_base = page_base % mobj.group(1)
 133
 134
 135         # Download video ids using BlipTV Ajax calls. Result size per
 136         # query is limited (currently to 12 videos) so we need to query
 137         # page by page until there are no video ids - it means we got
 138         # all of them.
 139
 140         video_ids = []
 141         pagenum = 1
 142
 143         while True:
 144             url = page_base + "&page=" + str(pagenum)
 145             page = self._download_webpage(url, username,
 146                                           'Downloading video ids from page %d' % pagenum)
 147
 148             # Extract video identifiers
 149             ids_in_page = []
 150
 151             for mobj in re.finditer(r'href="/([^"]+)"', page):
 152                 if mobj.group(1) not in ids_in_page:
 153                     ids_in_page.append(unescapeHTML(mobj.group(1)))
 154
 155             video_ids.extend(ids_in_page)
 156
 157             # A little optimization - if current page is not
 158             # "full", ie. does not contain PAGE_SIZE video ids then
 159             # we can assume that this page is the last one - there
 160             # are no more ids on further pages - no need to query
 161             # again.
 162
 163             if len(ids_in_page) < self._PAGE_SIZE:
 164                 break
 165
 166             pagenum += 1
 167
 168         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
 169         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
 170         return [self.playlist_result(url_entries, playlist_title = username)]