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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import (
   6     compat_b64decode,
   7     compat_str,
   8 )
   9 from ..utils import (
  10     clean_html,
  11     ExtractorError,
  12     int_or_none,
  13     str_or_none,
  14     try_get,
  15     url_or_none,
  16     urlencode_postdata,
  17     urljoin,
  18 )
  19
  20
  21 class PlatziBaseIE(InfoExtractor):
  22     _LOGIN_URL = 'https://platzi.com/login/'
  23     _NETRC_MACHINE = 'platzi'
  24
  25     def _real_initialize(self):
  26         self._login()
  27
  28     def _login(self):
  29         username, password = self._get_login_info()
  30         if username is None:
  31             return
  32
  33         login_page = self._download_webpage(
  34             self._LOGIN_URL, None, 'Downloading login page')
  35
  36         login_form = self._hidden_inputs(login_page)
  37
  38         login_form.update({
  39             'email': username,
  40             'password': password,
  41         })
  42
  43         urlh = self._request_webpage(
  44             self._LOGIN_URL, None, 'Logging in',
  45             data=urlencode_postdata(login_form),
  46             headers={'Referer': self._LOGIN_URL})
  47
  48         # login succeeded
  49         if 'platzi.com/login' not in urlh.geturl():
  50             return
  51
  52         login_error = self._webpage_read_content(
  53             urlh, self._LOGIN_URL, None, 'Downloading login error page')
  54
  55         login = self._parse_json(
  56             self._search_regex(
  57                 r'login\s*=\s*({.+?})(?:\s*;|\s*</script)', login_error, 'login'),
  58             None)
  59
  60         for kind in ('error', 'password', 'nonFields'):
  61             error = str_or_none(login.get('%sError' % kind))
  62             if error:
  63                 raise ExtractorError(
  64                     'Unable to login: %s' % error, expected=True)
  65         raise ExtractorError('Unable to log in')
  66
  67
  68 class PlatziIE(PlatziBaseIE):
  69     _VALID_URL = r'''(?x)
  70                     https?://
  71                         (?:
  72                             platzi\.com/clases|           # es version
  73                             courses\.platzi\.com/classes  # en version
  74                         )/[^/]+/(?P<id>\d+)-[^/?\#&]+
  75                     '''
  76
  77     _TESTS = [{
  78         'url': 'https://platzi.com/clases/1311-next-js/12074-creando-nuestra-primera-pagina/',
  79         'md5': '8f56448241005b561c10f11a595b37e3',
  80         'info_dict': {
  81             'id': '12074',
  82             'ext': 'mp4',
  83             'title': 'Creando nuestra primera página',
  84             'description': 'md5:4c866e45034fc76412fbf6e60ae008bc',
  85             'duration': 420,
  86         },
  87         'skip': 'Requires platzi account credentials',
  88     }, {
  89         'url': 'https://courses.platzi.com/classes/1367-communication-codestream/13430-background/',
  90         'info_dict': {
  91             'id': '13430',
  92             'ext': 'mp4',
  93             'title': 'Background',
  94             'description': 'md5:49c83c09404b15e6e71defaf87f6b305',
  95             'duration': 360,
  96         },
  97         'skip': 'Requires platzi account credentials',
  98         'params': {
  99             'skip_download': True,
 100         },
 101     }]
 102
 103     def _real_extract(self, url):
 104         lecture_id = self._match_id(url)
 105
 106         webpage = self._download_webpage(url, lecture_id)
 107
 108         data = self._parse_json(
 109             self._search_regex(
 110                 # client_data may contain "};" so that we have to try more
 111                 # strict regex first
 112                 (r'client_data\s*=\s*({.+?})\s*;\s*\n',
 113                  r'client_data\s*=\s*({.+?})\s*;'),
 114                 webpage, 'client data'),
 115             lecture_id)
 116
 117         material = data['initialState']['material']
 118         desc = material['description']
 119         title = desc['title']
 120
 121         formats = []
 122         for server_id, server in material['videos'].items():
 123             if not isinstance(server, dict):
 124                 continue
 125             for format_id in ('hls', 'dash'):
 126                 format_url = url_or_none(server.get(format_id))
 127                 if not format_url:
 128                     continue
 129                 if format_id == 'hls':
 130                     formats.extend(self._extract_m3u8_formats(
 131                         format_url, lecture_id, 'mp4',
 132                         entry_protocol='m3u8_native', m3u8_id=format_id,
 133                         note='Downloading %s m3u8 information' % server_id,
 134                         fatal=False))
 135                 elif format_id == 'dash':
 136                     formats.extend(self._extract_mpd_formats(
 137                         format_url, lecture_id, mpd_id=format_id,
 138                         note='Downloading %s MPD manifest' % server_id,
 139                         fatal=False))
 140         self._sort_formats(formats)
 141
 142         content = str_or_none(desc.get('content'))
 143         description = (clean_html(compat_b64decode(content).decode('utf-8'))
 144                        if content else None)
 145         duration = int_or_none(material.get('duration'), invscale=60)
 146
 147         return {
 148             'id': lecture_id,
 149             'title': title,
 150             'description': description,
 151             'duration': duration,
 152             'formats': formats,
 153         }
 154
 155
 156 class PlatziCourseIE(PlatziBaseIE):
 157     _VALID_URL = r'''(?x)
 158                     https?://
 159                         (?:
 160                             platzi\.com/clases|           # es version
 161                             courses\.platzi\.com/classes  # en version
 162                         )/(?P<id>[^/?\#&]+)
 163                     '''
 164     _TESTS = [{
 165         'url': 'https://platzi.com/clases/next-js/',
 166         'info_dict': {
 167             'id': '1311',
 168             'title': 'Curso de Next.js',
 169         },
 170         'playlist_count': 22,
 171     }, {
 172         'url': 'https://courses.platzi.com/classes/communication-codestream/',
 173         'info_dict': {
 174             'id': '1367',
 175             'title': 'Codestream Course',
 176         },
 177         'playlist_count': 14,
 178     }]
 179
 180     @classmethod
 181     def suitable(cls, url):
 182         return False if PlatziIE.suitable(url) else super(PlatziCourseIE, cls).suitable(url)
 183
 184     def _real_extract(self, url):
 185         course_name = self._match_id(url)
 186
 187         webpage = self._download_webpage(url, course_name)
 188
 189         props = self._parse_json(
 190             self._search_regex(r'data\s*=\s*({.+?})\s*;', webpage, 'data'),
 191             course_name)['initialProps']
 192
 193         entries = []
 194         for chapter_num, chapter in enumerate(props['concepts'], 1):
 195             if not isinstance(chapter, dict):
 196                 continue
 197             materials = chapter.get('materials')
 198             if not materials or not isinstance(materials, list):
 199                 continue
 200             chapter_title = chapter.get('title')
 201             chapter_id = str_or_none(chapter.get('id'))
 202             for material in materials:
 203                 if not isinstance(material, dict):
 204                     continue
 205                 if material.get('material_type') != 'video':
 206                     continue
 207                 video_url = urljoin(url, material.get('url'))
 208                 if not video_url:
 209                     continue
 210                 entries.append({
 211                     '_type': 'url_transparent',
 212                     'url': video_url,
 213                     'title': str_or_none(material.get('name')),
 214                     'id': str_or_none(material.get('id')),
 215                     'ie_key': PlatziIE.ie_key(),
 216                     'chapter': chapter_title,
 217                     'chapter_number': chapter_num,
 218                     'chapter_id': chapter_id,
 219                 })
 220
 221         course_id = compat_str(try_get(props, lambda x: x['course']['id']))
 222         course_title = try_get(props, lambda x: x['course']['name'], compat_str)
 223
 224         return self.playlist_result(entries, course_id, course_title)