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