Merge changes from buster branch for the NMU while I was away.

[youtubedl] / youtube_dl / extractor / lynda.py

diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py
index da94eab561b91d6b70675911e432b5750d5d5b04..3084c6dffc9cd3b32ede52906669299807a9ad20 100644 (file)
--- a/youtube_dl/extractor/lynda.py
+++ b/youtube_dl/extractor/lynda.py
@@ -4,7 +4,6 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_HTTPError,
     compat_str,
     compat_urlparse,
 )
@@ -16,7 +15,7 @@ from ..utils import (
 
 
 class LyndaBaseIE(InfoExtractor):
-    _SIGNIN_URL = 'https://www.lynda.com/signin'
+    _SIGNIN_URL = 'https://www.lynda.com/signin/lynda'
     _PASSWORD_URL = 'https://www.lynda.com/signin/password'
     _USER_URL = 'https://www.lynda.com/signin/user'
     _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
@@ -44,21 +43,15 @@ class LyndaBaseIE(InfoExtractor):
         form_data = self._hidden_inputs(form_html)
         form_data.update(extra_form_data)
 
-        try:
-            response = self._download_json(
-                action_url, None, note,
-                data=urlencode_postdata(form_data),
-                headers={
-                    'Referer': referrer_url,
-                    'X-Requested-With': 'XMLHttpRequest',
-                })
-        except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
-                response = self._parse_json(e.cause.read().decode('utf-8'), None)
-                self._check_error(response, ('email', 'password'))
-            raise
-
-        self._check_error(response, 'ErrorMessage')
+        response = self._download_json(
+            action_url, None, note,
+            data=urlencode_postdata(form_data),
+            headers={
+                'Referer': referrer_url,
+                'X-Requested-With': 'XMLHttpRequest',
+            }, expected_status=(418, 500, ))
+
+        self._check_error(response, ('email', 'password', 'ErrorMessage'))
 
         return response, action_url
 
@@ -94,7 +87,15 @@ class LyndaBaseIE(InfoExtractor):
 class LyndaIE(LyndaBaseIE):
     IE_NAME = 'lynda'
     IE_DESC = 'lynda.com videos'
-    _VALID_URL = r'https?://(?:www\.)?lynda\.com/(?:[^/]+/[^/]+/(?P<course_id>\d+)|player/embed)/(?P<id>\d+)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?(?:lynda\.com|educourse\.ga)/
+                        (?:
+                            (?:[^/]+/){2,3}(?P<course_id>\d+)|
+                            player/embed
+                        )/
+                        (?P<id>\d+)
+                    '''
 
     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
 
@@ -110,6 +111,12 @@ class LyndaIE(LyndaBaseIE):
     }, {
         'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0',
         'only_matching': True,
+    }, {
+        'url': 'https://educourse.ga/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Willkommen-Grundlagen-guten-Gestaltung/393570/393572-4.html',
+        'only_matching': True,
     }]
 
     def _raise_unavailable(self, video_id):
@@ -241,8 +248,9 @@ class LyndaIE(LyndaBaseIE):
     def _get_subtitles(self, 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)}]}
+        fixed_subs = self._fix_subtitles(subs)
+        if fixed_subs:
+            return {'en': [{'ext': 'srt', 'data': fixed_subs}]}
         else:
             return {}
 
@@ -253,16 +261,39 @@ class LyndaCourseIE(LyndaBaseIE):
 
     # Course link equals to welcome/introduction video link of same course
     # We will recognize it as course link
-    _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
+    _VALID_URL = r'https?://(?:www|m)\.(?:lynda\.com|educourse\.ga)/(?P<coursepath>(?:[^/]+/){2,3}(?P<courseid>\d+))-2\.html'
+
+    _TESTS = [{
+        'url': 'https://www.lynda.com/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         course_path = mobj.group('coursepath')
         course_id = mobj.group('courseid')
 
+        item_template = 'https://www.lynda.com/%s/%%s-4.html' % course_path
+
         course = self._download_json(
             'https://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
-            course_id, 'Downloading course JSON')
+            course_id, 'Downloading course JSON', fatal=False)
+
+        if not course:
+            webpage = self._download_webpage(url, course_id)
+            entries = [
+                self.url_result(
+                    item_template % video_id, ie=LyndaIE.ie_key(),
+                    video_id=video_id)
+                for video_id in re.findall(
+                    r'data-video-id=["\'](\d+)', webpage)]
+            return self.playlist_result(
+                entries, course_id,
+                self._og_search_title(webpage, fatal=False),
+                self._og_search_description(webpage))
 
         if course.get('Status') == 'NotFound':
             raise ExtractorError(
@@ -283,7 +314,7 @@ class LyndaCourseIE(LyndaBaseIE):
                 if video_id:
                     entries.append({
                         '_type': 'url_transparent',
-                        'url': 'https://www.lynda.com/%s/%s-4.html' % (course_path, video_id),
+                        'url': item_template % video_id,
                         'ie_key': LyndaIE.ie_key(),
                         'chapter': chapter.get('Title'),
                         'chapter_number': int_or_none(chapter.get('ChapterIndex')),