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

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     parse_iso8601,
  10 )
  11
  12
  13 class DRBonanzaIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?dr\.dk/bonanza/(?:[^/]+/)+(?:[^/])+?(?:assetId=(?P<id>\d+))?(?:[#&]|$)'
  15
  16     _TESTS = [{
  17         'url': 'http://www.dr.dk/bonanza/serie/portraetter/Talkshowet.htm?assetId=65517',
  18         'info_dict': {
  19             'id': '65517',
  20             'ext': 'mp4',
  21             'title': 'Talkshowet - Leonard Cohen',
  22             'description': 'md5:8f34194fb30cd8c8a30ad8b27b70c0ca',
  23             'thumbnail': 're:^https?://.*\.(?:gif|jpg)$',
  24             'timestamp': 1295537932,
  25             'upload_date': '20110120',
  26             'duration': 3664,
  27         },
  28         'params': {
  29             'skip_download': True,  # requires rtmp
  30         },
  31     }, {
  32         'url': 'http://www.dr.dk/bonanza/radio/serie/sport/fodbold.htm?assetId=59410',
  33         'md5': '6dfe039417e76795fb783c52da3de11d',
  34         'info_dict': {
  35             'id': '59410',
  36             'ext': 'mp3',
  37             'title': 'EM fodbold 1992 Danmark - Tyskland finale Transmission',
  38             'description': 'md5:501e5a195749480552e214fbbed16c4e',
  39             'thumbnail': 're:^https?://.*\.(?:gif|jpg)$',
  40             'timestamp': 1223274900,
  41             'upload_date': '20081006',
  42             'duration': 7369,
  43         },
  44     }]
  45
  46     def _real_extract(self, url):
  47         url_id = self._match_id(url)
  48         webpage = self._download_webpage(url, url_id)
  49
  50         if url_id:
  51             info = json.loads(self._html_search_regex(r'({.*?%s.*})' % url_id, webpage, 'json'))
  52         else:
  53             # Just fetch the first video on that page
  54             info = json.loads(self._html_search_regex(r'bonanzaFunctions.newPlaylist\(({.*})\)', webpage, 'json'))
  55
  56         asset_id = str(info['AssetId'])
  57         title = info['Title'].rstrip(' \'\"-,.:;!?')
  58         duration = int_or_none(info.get('Duration'), scale=1000)
  59         # First published online. "FirstPublished" contains the date for original airing.
  60         timestamp = parse_iso8601(
  61             re.sub(r'\.\d+$', '', info['Created']))
  62
  63         def parse_filename_info(url):
  64             match = re.search(r'/\d+_(?P<width>\d+)x(?P<height>\d+)x(?P<bitrate>\d+)K\.(?P<ext>\w+)$', url)
  65             if match:
  66                 return {
  67                     'width': int(match.group('width')),
  68                     'height': int(match.group('height')),
  69                     'vbr': int(match.group('bitrate')),
  70                     'ext': match.group('ext')
  71                 }
  72             match = re.search(r'/\d+_(?P<bitrate>\d+)K\.(?P<ext>\w+)$', url)
  73             if match:
  74                 return {
  75                     'vbr': int(match.group('bitrate')),
  76                     'ext': match.group(2)
  77                 }
  78             return {}
  79
  80         video_types = ['VideoHigh', 'VideoMid', 'VideoLow']
  81         preferencemap = {
  82             'VideoHigh': -1,
  83             'VideoMid': -2,
  84             'VideoLow': -3,
  85             'Audio': -4,
  86         }
  87
  88         formats = []
  89         for file in info['Files']:
  90             if info['Type'] == "Video":
  91                 if file['Type'] in video_types:
  92                     format = parse_filename_info(file['Location'])
  93                     format.update({
  94                         'url': file['Location'],
  95                         'format_id': file['Type'].replace('Video', ''),
  96                         'preference': preferencemap.get(file['Type'], -10),
  97                     })
  98                     if format['url'].startswith('rtmp'):
  99                         rtmp_url = format['url']
 100                         format['rtmp_live'] = True  # --resume does not work
 101                         if '/bonanza/' in rtmp_url:
 102                             format['play_path'] = rtmp_url.split('/bonanza/')[1]
 103                     formats.append(format)
 104                 elif file['Type'] == "Thumb":
 105                     thumbnail = file['Location']
 106             elif info['Type'] == "Audio":
 107                 if file['Type'] == "Audio":
 108                     format = parse_filename_info(file['Location'])
 109                     format.update({
 110                         'url': file['Location'],
 111                         'format_id': file['Type'],
 112                         'vcodec': 'none',
 113                     })
 114                     formats.append(format)
 115                 elif file['Type'] == "Thumb":
 116                     thumbnail = file['Location']
 117
 118         description = '%s\n%s\n%s\n' % (
 119             info['Description'], info['Actors'], info['Colophon'])
 120
 121         self._sort_formats(formats)
 122
 123         display_id = re.sub(r'[^\w\d-]', '', re.sub(r' ', '-', title.lower())) + '-' + asset_id
 124         display_id = re.sub(r'-+', '-', display_id)
 125
 126         return {
 127             'id': asset_id,
 128             'display_id': display_id,
 129             'title': title,
 130             'formats': formats,
 131             'description': description,
 132             'thumbnail': thumbnail,
 133             'timestamp': timestamp,
 134             'duration': duration,
 135         }