yt_dlp/extractor/infoq.py

   1 from .bokecc import BokeCCBaseIE
   2 from ..compat import (
   3     compat_b64decode,
   4     compat_urllib_parse_unquote,
   5     compat_urlparse,
   6 )
   7 from ..utils import (
   8     ExtractorError,
   9     determine_ext,
  10     traverse_obj,
  11     update_url_query,
  12 )
  13
  14
  15 class InfoQIE(BokeCCBaseIE):
  16     _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
  17
  18     _TESTS = [{
  19         'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
  20         'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2',
  21         'info_dict': {
  22             'id': 'A-Few-of-My-Favorite-Python-Things',
  23             'ext': 'mp4',
  24             'description': 'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
  25             'title': 'A Few of My Favorite [Python] Things',
  26         },
  27     }, {
  28         'url': 'http://www.infoq.com/fr/presentations/changez-avis-sur-javascript',
  29         'only_matching': True,
  30     }, {
  31         'url': 'http://www.infoq.com/cn/presentations/openstack-continued-delivery',
  32         'md5': '4918d0cca1497f2244572caf626687ef',
  33         'info_dict': {
  34             'id': 'openstack-continued-delivery',
  35             'title': 'OpenStack持续交付之路',
  36             'ext': 'flv',
  37             'description': 'md5:308d981fb28fa42f49f9568322c683ff',
  38         },
  39         'skip': 'Sorry, the page you visited does not exist',
  40     }, {
  41         'url': 'https://www.infoq.com/presentations/Simple-Made-Easy',
  42         'md5': '0e34642d4d9ef44bf86f66f6399672db',
  43         'info_dict': {
  44             'id': 'Simple-Made-Easy',
  45             'title': 'Simple Made Easy',
  46             'ext': 'mp3',
  47             'description': 'md5:3e0e213a8bbd074796ef89ea35ada25b',
  48         },
  49         'params': {
  50             'format': 'bestaudio',
  51         },
  52     }]
  53
  54     def _extract_rtmp_video(self, webpage):
  55         # The server URL is hardcoded
  56         video_url = 'rtmpe://videof.infoq.com/cfx/st/'
  57
  58         # Extract video URL
  59         encoded_id = self._search_regex(
  60             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None)
  61
  62         real_id = compat_urllib_parse_unquote(compat_b64decode(encoded_id).decode('utf-8'))
  63         playpath = 'mp4:' + real_id
  64
  65         return [{
  66             'format_id': 'rtmp_video',
  67             'url': video_url,
  68             'ext': determine_ext(playpath),
  69             'play_path': playpath,
  70         }]
  71
  72     def _extract_cf_auth(self, webpage):
  73         policy = self._search_regex(r'InfoQConstants\.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
  74         signature = self._search_regex(r'InfoQConstants\.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
  75         key_pair_id = self._search_regex(r'InfoQConstants\.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
  76         return {
  77             'Policy': policy,
  78             'Signature': signature,
  79             'Key-Pair-Id': key_pair_id,
  80         }
  81
  82     def _extract_http_video(self, webpage):
  83         http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
  84         http_video_url = update_url_query(http_video_url, self._extract_cf_auth(webpage))
  85         return [{
  86             'format_id': 'http_video',
  87             'url': http_video_url,
  88             'http_headers': {'Referer': 'https://www.infoq.com/'},
  89         }]
  90
  91     def _extract_http_audio(self, webpage, video_id):
  92         try:
  93             http_audio_url = traverse_obj(self._form_hidden_inputs('mp3Form', webpage), 'filename')
  94         except ExtractorError:
  95             http_audio_url = None
  96         if not http_audio_url:
  97             return []
  98
  99         # base URL is found in the Location header in the response returned by
 100         # GET https://www.infoq.com/mp3download.action?filename=... when logged in.
 101         http_audio_url = compat_urlparse.urljoin('http://ress.infoq.com/downloads/mp3downloads/', http_audio_url)
 102         http_audio_url = update_url_query(http_audio_url, self._extract_cf_auth(webpage))
 103
 104         # audio file seem to be missing some times even if there is a download link
 105         # so probe URL to make sure
 106         if not self._is_valid_url(http_audio_url, video_id):
 107             return []
 108
 109         return [{
 110             'format_id': 'http_audio',
 111             'url': http_audio_url,
 112             'vcodec': 'none',
 113         }]
 114
 115     def _real_extract(self, url):
 116         video_id = self._match_id(url)
 117         webpage = self._download_webpage(url, video_id)
 118
 119         video_title = self._html_extract_title(webpage)
 120         video_description = self._html_search_meta('description', webpage, 'description')
 121
 122         if '/cn/' in url:
 123             # for China videos, HTTP video URL exists but always fails with 403
 124             formats = self._extract_bokecc_formats(webpage, video_id)
 125         else:
 126             formats = (
 127                 self._extract_rtmp_video(webpage)
 128                 + self._extract_http_video(webpage)
 129                 + self._extract_http_audio(webpage, video_id))
 130
 131         return {
 132             'id': video_id,
 133             'title': video_title,
 134             'description': video_description,
 135             'formats': formats,
 136         }