yt_dlp/extractor/platzi.py

   1 from .common import InfoExtractor
   2 from ..compat import (
   3     compat_b64decode,
   4     compat_str,
   5 )
   6 from ..utils import (
   7     ExtractorError,
   8     clean_html,
   9     int_or_none,
  10     str_or_none,
  11     try_get,
  12     url_or_none,
  13     urlencode_postdata,
  14     urljoin,
  15 )
  16
  17
  18 class PlatziBaseIE(InfoExtractor):
  19     _LOGIN_URL = 'https://platzi.com/login/'
  20     _NETRC_MACHINE = 'platzi'
  21
  22     def _perform_login(self, username, password):
  23         login_page = self._download_webpage(
  24             self._LOGIN_URL, None, 'Downloading login page')
  25
  26         login_form = self._hidden_inputs(login_page)
  27
  28         login_form.update({
  29             'email': username,
  30             'password': password,
  31         })
  32
  33         urlh = self._request_webpage(
  34             self._LOGIN_URL, None, 'Logging in',
  35             data=urlencode_postdata(login_form),
  36             headers={'Referer': self._LOGIN_URL})
  37
  38         # login succeeded
  39         if 'platzi.com/login' not in urlh.url:
  40             return
  41
  42         login_error = self._webpage_read_content(
  43             urlh, self._LOGIN_URL, None, 'Downloading login error page')
  44
  45         login = self._parse_json(
  46             self._search_regex(
  47                 r'login\s*=\s*({.+?})(?:\s*;|\s*</script)', login_error, 'login'),
  48             None)
  49
  50         for kind in ('error', 'password', 'nonFields'):
  51             error = str_or_none(login.get('%sError' % kind))
  52             if error:
  53                 raise ExtractorError(
  54                     'Unable to login: %s' % error, expected=True)
  55         raise ExtractorError('Unable to log in')
  56
  57
  58 class PlatziIE(PlatziBaseIE):
  59     _VALID_URL = r'''(?x)
  60                     https?://
  61                         (?:
  62                             platzi\.com/clases|           # es version
  63                             courses\.platzi\.com/classes  # en version
  64                         )/[^/]+/(?P<id>\d+)-[^/?\#&]+
  65                     '''
  66
  67     _TESTS = [{
  68         'url': 'https://platzi.com/clases/1311-next-js/12074-creando-nuestra-primera-pagina/',
  69         'md5': '8f56448241005b561c10f11a595b37e3',
  70         'info_dict': {
  71             'id': '12074',
  72             'ext': 'mp4',
  73             'title': 'Creando nuestra primera página',
  74             'description': 'md5:4c866e45034fc76412fbf6e60ae008bc',
  75             'duration': 420,
  76         },
  77         'skip': 'Requires platzi account credentials',
  78     }, {
  79         'url': 'https://courses.platzi.com/classes/1367-communication-codestream/13430-background/',
  80         'info_dict': {
  81             'id': '13430',
  82             'ext': 'mp4',
  83             'title': 'Background',
  84             'description': 'md5:49c83c09404b15e6e71defaf87f6b305',
  85             'duration': 360,
  86         },
  87         'skip': 'Requires platzi account credentials',
  88         'params': {
  89             'skip_download': True,
  90         },
  91     }]
  92
  93     def _real_extract(self, url):
  94         lecture_id = self._match_id(url)
  95
  96         webpage = self._download_webpage(url, lecture_id)
  97
  98         data = self._parse_json(
  99             self._search_regex(
 100                 # client_data may contain "};" so that we have to try more
 101                 # strict regex first
 102                 (r'client_data\s*=\s*({.+?})\s*;\s*\n',
 103                  r'client_data\s*=\s*({.+?})\s*;'),
 104                 webpage, 'client data'),
 105             lecture_id)
 106
 107         material = data['initialState']['material']
 108         desc = material['description']
 109         title = desc['title']
 110
 111         formats = []
 112         for server_id, server in material['videos'].items():
 113             if not isinstance(server, dict):
 114                 continue
 115             for format_id in ('hls', 'dash'):
 116                 format_url = url_or_none(server.get(format_id))
 117                 if not format_url:
 118                     continue
 119                 if format_id == 'hls':
 120                     formats.extend(self._extract_m3u8_formats(
 121                         format_url, lecture_id, 'mp4',
 122                         entry_protocol='m3u8_native', m3u8_id=format_id,
 123                         note='Downloading %s m3u8 information' % server_id,
 124                         fatal=False))
 125                 elif format_id == 'dash':
 126                     formats.extend(self._extract_mpd_formats(
 127                         format_url, lecture_id, mpd_id=format_id,
 128                         note='Downloading %s MPD manifest' % server_id,
 129                         fatal=False))
 130
 131         content = str_or_none(desc.get('content'))
 132         description = (clean_html(compat_b64decode(content).decode('utf-8'))
 133                        if content else None)
 134         duration = int_or_none(material.get('duration'), invscale=60)
 135
 136         return {
 137             'id': lecture_id,
 138             'title': title,
 139             'description': description,
 140             'duration': duration,
 141             'formats': formats,
 142         }
 143
 144
 145 class PlatziCourseIE(PlatziBaseIE):
 146     _VALID_URL = r'''(?x)
 147                     https?://
 148                         (?:
 149                             platzi\.com/clases|           # es version
 150                             courses\.platzi\.com/classes  # en version
 151                         )/(?P<id>[^/?\#&]+)
 152                     '''
 153     _TESTS = [{
 154         'url': 'https://platzi.com/clases/next-js/',
 155         'info_dict': {
 156             'id': '1311',
 157             'title': 'Curso de Next.js',
 158         },
 159         'playlist_count': 22,
 160     }, {
 161         'url': 'https://courses.platzi.com/classes/communication-codestream/',
 162         'info_dict': {
 163             'id': '1367',
 164             'title': 'Codestream Course',
 165         },
 166         'playlist_count': 14,
 167     }]
 168
 169     @classmethod
 170     def suitable(cls, url):
 171         return False if PlatziIE.suitable(url) else super(PlatziCourseIE, cls).suitable(url)
 172
 173     def _real_extract(self, url):
 174         course_name = self._match_id(url)
 175
 176         webpage = self._download_webpage(url, course_name)
 177
 178         props = self._parse_json(
 179             self._search_regex(r'data\s*=\s*({.+?})\s*;', webpage, 'data'),
 180             course_name)['initialProps']
 181
 182         entries = []
 183         for chapter_num, chapter in enumerate(props['concepts'], 1):
 184             if not isinstance(chapter, dict):
 185                 continue
 186             materials = chapter.get('materials')
 187             if not materials or not isinstance(materials, list):
 188                 continue
 189             chapter_title = chapter.get('title')
 190             chapter_id = str_or_none(chapter.get('id'))
 191             for material in materials:
 192                 if not isinstance(material, dict):
 193                     continue
 194                 if material.get('material_type') != 'video':
 195                     continue
 196                 video_url = urljoin(url, material.get('url'))
 197                 if not video_url:
 198                     continue
 199                 entries.append({
 200                     '_type': 'url_transparent',
 201                     'url': video_url,
 202                     'title': str_or_none(material.get('name')),
 203                     'id': str_or_none(material.get('id')),
 204                     'ie_key': PlatziIE.ie_key(),
 205                     'chapter': chapter_title,
 206                     'chapter_number': chapter_num,
 207                     'chapter_id': chapter_id,
 208                 })
 209
 210         course_id = compat_str(try_get(props, lambda x: x['course']['id']))
 211         course_title = try_get(props, lambda x: x['course']['name'], compat_str)
 212
 213         return self.playlist_result(entries, course_id, course_title)