youtube_dlc/extractor/alura.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8 from ..compat import (
   9     compat_urlparse,
  10 )
  11
  12 from ..utils import (
  13     urlencode_postdata,
  14     urljoin,
  15     int_or_none,
  16     clean_html,
  17     ExtractorError
  18 )
  19
  20
  21 class AluraIE(InfoExtractor):
  22     _VALID_URL = r'https?://(?:cursos\.)?alura\.com\.br/course/(?P<course_name>[^/]+)/task/(?P<id>\d+)'
  23     _LOGIN_URL = 'https://cursos.alura.com.br/loginForm?urlAfterLogin=/loginForm'
  24     _VIDEO_URL = 'https://cursos.alura.com.br/course/%s/task/%s/video'
  25     _NETRC_MACHINE = 'alura'
  26     _TESTS = [{
  27         'url': 'https://cursos.alura.com.br/course/clojure-mutabilidade-com-atoms-e-refs/task/60095',
  28         'info_dict': {
  29             'id': '60095',
  30             'ext': 'mp4',
  31             'title': 'Referências, ref-set e alter'
  32         },
  33         'skip': 'Requires alura account credentials'},
  34         {
  35             # URL without video
  36             'url': 'https://cursos.alura.com.br/course/clojure-mutabilidade-com-atoms-e-refs/task/60098',
  37             'only_matching': True},
  38         {
  39             'url': 'https://cursos.alura.com.br/course/fundamentos-market-digital/task/55219',
  40             'only_matching': True}
  41     ]
  42
  43     def _real_extract(self, url):
  44
  45         video_id = self._match_id(url)
  46         course = self._search_regex(self._VALID_URL, url, 'post url', group='course_name')
  47         video_url = self._VIDEO_URL % (course, video_id)
  48
  49         video_dict = self._download_json(video_url, video_id, 'Searching for videos')
  50
  51         if video_dict:
  52             webpage = self._download_webpage(url, video_id)
  53             video_title = clean_html(self._search_regex(
  54                 r'<span[^>]+class=(["\'])task-body-header-title-text\1[^>]*>(?P<title>[^<]+)',
  55                 webpage, 'title', group='title'))
  56
  57             formats = []
  58             for video_obj in video_dict:
  59                 video_url_m3u8 = video_obj.get('link')
  60                 video_format = self._extract_m3u8_formats(
  61                     video_url_m3u8, None, 'mp4', entry_protocol='m3u8_native',
  62                     m3u8_id='hls', fatal=False)
  63                 for f in video_format:
  64                     m = re.search(r'^[\w \W]*-(?P<res>\w*).mp4[\W \w]*', f['url'])
  65                     if m:
  66                         if not f.get('height'):
  67                             f['height'] = int('720' if m.group('res') == 'hd' else '480')
  68                 formats.extend(video_format)
  69
  70             self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))
  71
  72             return {
  73                 'id': video_id,
  74                 'title': video_title,
  75                 "formats": formats
  76             }
  77
  78     def _real_initialize(self):
  79         self._login()
  80
  81     def _login(self):
  82         username, password = self._get_login_info()
  83         if username is None:
  84             return
  85         pass
  86
  87         login_page = self._download_webpage(
  88             self._LOGIN_URL, None, 'Downloading login popup')
  89
  90         def is_logged(webpage):
  91             return any(re.search(p, webpage) for p in (
  92                 r'href=[\"|\']?/signout[\"|\']',
  93                 r'>Logout<'))
  94
  95         # already logged in
  96         if is_logged(login_page):
  97             return
  98
  99         login_form = self._hidden_inputs(login_page)
 100
 101         login_form.update({
 102             'username': username,
 103             'password': password,
 104         })
 105
 106         post_url = self._search_regex(
 107             r'<form[^>]+class=["|\']signin-form["|\'] action=["|\'](?P<url>.+?)["|\']', login_page,
 108             'post url', default=self._LOGIN_URL, group='url')
 109
 110         if not post_url.startswith('http'):
 111             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
 112
 113         response = self._download_webpage(
 114             post_url, None, 'Logging in',
 115             data=urlencode_postdata(login_form),
 116             headers={'Content-Type': 'application/x-www-form-urlencoded'})
 117
 118         if not is_logged(response):
 119             error = self._html_search_regex(
 120                 r'(?s)<p[^>]+class="alert-message[^"]*">(.+?)</p>',
 121                 response, 'error message', default=None)
 122             if error:
 123                 raise ExtractorError('Unable to login: %s' % error, expected=True)
 124             raise ExtractorError('Unable to log in')
 125
 126
 127 class AluraCourseIE(AluraIE):
 128
 129     _VALID_URL = r'https?://(?:cursos\.)?alura\.com\.br/course/(?P<id>[^/]+)'
 130     _LOGIN_URL = 'https://cursos.alura.com.br/loginForm?urlAfterLogin=/loginForm'
 131     _NETRC_MACHINE = 'aluracourse'
 132     _TESTS = [{
 133         'url': 'https://cursos.alura.com.br/course/clojure-mutabilidade-com-atoms-e-refs',
 134         'only_matching': True,
 135     }]
 136
 137     @classmethod
 138     def suitable(cls, url):
 139         return False if AluraIE.suitable(url) else super(AluraCourseIE, cls).suitable(url)
 140
 141     def _real_extract(self, url):
 142
 143         course_path = self._match_id(url)
 144         webpage = self._download_webpage(url, course_path)
 145
 146         course_title = self._search_regex(
 147             r'<h1.*?>(.*?)<strong>(?P<course_title>.*?)</strong></h[0-9]>', webpage,
 148             'course title', default=course_path, group='course_title')
 149
 150         entries = []
 151         if webpage:
 152             for path in re.findall(r'<a\b(?=[^>]* class="[^"]*(?<=[" ])courseSectionList-section[" ])(?=[^>]* href="([^"]*))', webpage):
 153                 page_url = urljoin(url, path)
 154                 section_path = self._download_webpage(page_url, course_path)
 155                 for path_video in re.findall(r'<a\b(?=[^>]* class="[^"]*(?<=[" ])task-menu-nav-item-link-VIDEO[" ])(?=[^>]* href="([^"]*))', section_path):
 156                     chapter = clean_html(
 157                         self._search_regex(
 158                             r'<h3[^>]+class=(["\'])task-menu-section-title-text\1[^>]*>(?P<chapter>[^<]+)',
 159                             section_path,
 160                             'chapter',
 161                             group='chapter'))
 162
 163                     chapter_number = int_or_none(
 164                         self._search_regex(
 165                             r'<span[^>]+class=(["\'])task-menu-section-title-number[^>]*>(.*?)<strong>(?P<chapter_number>[^<]+)</strong>',
 166                             section_path,
 167                             'chapter number',
 168                             group='chapter_number'))
 169                     video_url = urljoin(url, path_video)
 170
 171                     entry = {
 172                         '_type': 'url_transparent',
 173                         'id': self._match_id(video_url),
 174                         'url': video_url,
 175                         'id_key': self.ie_key(),
 176                         'chapter': chapter,
 177                         'chapter_number': chapter_number
 178                     }
 179                     entries.append(entry)
 180         return self.playlist_result(entries, course_path, course_title)