Fix "invalid escape sequences" error on Python 3.6
[youtube-dl] / youtube_dl / extractor / lynda.py
index c2678652e7b6dd0feaa68945238030a93073c244..da94eab561b91d6b70675911e432b5750d5d5b04 100644 (file)
@@ -71,6 +71,11 @@ class LyndaBaseIE(InfoExtractor):
         signin_page = self._download_webpage(
             self._SIGNIN_URL, None, 'Downloading signin page')
 
+        # Already logged in
+        if any(re.search(p, signin_page) for p in (
+                r'isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')):
+            return
+
         # Step 2: submit email
         signin_form = self._search_regex(
             r'(?s)(<form[^>]+data-form-name=["\']signin["\'][^>]*>.+?</form>)',
@@ -85,27 +90,17 @@ class LyndaBaseIE(InfoExtractor):
             password_form, self._USER_URL, {'email': username, 'password': password},
             'Submitting password', signin_url)
 
-    def _logout(self):
-        username, _ = self._get_login_info()
-        if username is None:
-            return
-
-        self._download_webpage(
-            'http://www.lynda.com/ajax/logout.aspx', None,
-            'Logging out', 'Unable to log out', fatal=False)
-
 
 class LyndaIE(LyndaBaseIE):
     IE_NAME = 'lynda'
     IE_DESC = 'lynda.com videos'
-    _VALID_URL = r'https?://www\.lynda\.com/(?:[^/]+/[^/]+/\d+|player/embed)/(?P<id>\d+)'
-    _NETRC_MACHINE = 'lynda'
+    _VALID_URL = r'https?://(?:www\.)?lynda\.com/(?:[^/]+/[^/]+/(?P<course_id>\d+)|player/embed)/(?P<id>\d+)'
 
     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
 
     _TESTS = [{
-        'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
-        'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
+        'url': 'https://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
+        # md5 is unstable
         'info_dict': {
             'id': '114408',
             'ext': 'mp4',
@@ -117,19 +112,71 @@ class LyndaIE(LyndaBaseIE):
         'only_matching': True,
     }]
 
+    def _raise_unavailable(self, video_id):
+        self.raise_login_required(
+            'Video %s is only available for members' % video_id)
+
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        course_id = mobj.group('course_id')
+
+        query = {
+            'videoId': video_id,
+            'type': 'video',
+        }
 
         video = self._download_json(
-            'http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
-            video_id, 'Downloading video JSON')
+            'https://www.lynda.com/ajax/player', video_id,
+            'Downloading video JSON', fatal=False, query=query)
+
+        # Fallback scenario
+        if not video:
+            query['courseId'] = course_id
+
+            play = self._download_json(
+                'https://www.lynda.com/ajax/course/%s/%s/play'
+                % (course_id, video_id), video_id, 'Downloading play JSON')
+
+            if not play:
+                self._raise_unavailable(video_id)
+
+            formats = []
+            for formats_dict in play:
+                urls = formats_dict.get('urls')
+                if not isinstance(urls, dict):
+                    continue
+                cdn = formats_dict.get('name')
+                for format_id, format_url in urls.items():
+                    if not format_url:
+                        continue
+                    formats.append({
+                        'url': format_url,
+                        'format_id': '%s-%s' % (cdn, format_id) if cdn else format_id,
+                        'height': int_or_none(format_id),
+                    })
+            self._sort_formats(formats)
+
+            conviva = self._download_json(
+                'https://www.lynda.com/ajax/player/conviva', video_id,
+                'Downloading conviva JSON', query=query)
+
+            return {
+                'id': video_id,
+                'title': conviva['VideoTitle'],
+                'description': conviva.get('VideoDescription'),
+                'release_year': int_or_none(conviva.get('ReleaseYear')),
+                'duration': int_or_none(conviva.get('Duration')),
+                'creator': conviva.get('Author'),
+                'formats': formats,
+            }
 
         if 'Status' in video:
             raise ExtractorError(
                 'lynda returned error: %s' % video['Message'], expected=True)
 
         if video.get('HasAccess') is False:
-            self.raise_login_required('Video %s is only available for members' % video_id)
+            self._raise_unavailable(video_id)
 
         video_id = compat_str(video.get('ID') or video_id)
         duration = int_or_none(video.get('DurationInSeconds'))
@@ -153,7 +200,7 @@ class LyndaIE(LyndaBaseIE):
             for prioritized_stream_id, prioritized_stream in prioritized_streams.items():
                 formats.extend([{
                     'url': video_url,
-                    'width': int_or_none(format_id),
+                    'height': int_or_none(format_id),
                     'format_id': '%s-%s' % (prioritized_stream_id, format_id),
                 } for format_id, video_url in prioritized_stream.items()])
 
@@ -192,7 +239,7 @@ class LyndaIE(LyndaBaseIE):
             return srt
 
     def _get_subtitles(self, video_id):
-        url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
+        url = 'https://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
         subs = self._download_json(url, None, False)
         if subs:
             return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]}
@@ -214,11 +261,9 @@ class LyndaCourseIE(LyndaBaseIE):
         course_id = mobj.group('courseid')
 
         course = self._download_json(
-            'http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
+            'https://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
             course_id, 'Downloading course JSON')
 
-        self._logout()
-
         if course.get('Status') == 'NotFound':
             raise ExtractorError(
                 'Course %s does not exist' % course_id, expected=True)
@@ -238,7 +283,7 @@ class LyndaCourseIE(LyndaBaseIE):
                 if video_id:
                     entries.append({
                         '_type': 'url_transparent',
-                        'url': 'http://www.lynda.com/%s/%s-4.html' % (course_path, video_id),
+                        'url': 'https://www.lynda.com/%s/%s-4.html' % (course_path, video_id),
                         'ie_key': LyndaIE.ie_key(),
                         'chapter': chapter.get('Title'),
                         'chapter_number': int_or_none(chapter.get('ChapterIndex')),
@@ -251,5 +296,6 @@ class LyndaCourseIE(LyndaBaseIE):
                 % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)
 
         course_title = course.get('Title')
+        course_description = course.get('Description')
 
-        return self.playlist_result(entries, course_id, course_title)
+        return self.playlist_result(entries, course_id, course_title, course_description)