New upstream version 2017.09.24

[youtubedl] / youtube_dl / extractor / udemy.py

diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py
index dae1aa3c6ba6fd211233abbd316fb78409c33faf..207c4a6a7ee8131c3e2e5d5823aefb336ad47c47 100644 (file)
--- a/youtube_dl/extractor/udemy.py
+++ b/youtube_dl/extractor/udemy.py
@@ -15,6 +15,7 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     int_or_none,
+    js_to_json,
     sanitized_Request,
     unescapeHTML,
     urlencode_postdata,
@@ -52,6 +53,10 @@ class UdemyIE(InfoExtractor):
         # new URL schema
         'url': 'https://www.udemy.com/electric-bass-right-from-the-start/learn/v4/t/lecture/4580906',
         'only_matching': True,
+    }, {
+        # no url in outputs format entry
+        'url': 'https://www.udemy.com/learn-web-development-complete-step-by-step-guide-to-success/learn/v4/t/lecture/4125812',
+        'only_matching': True,
     }]
 
     def _extract_course_info(self, webpage, video_id):
@@ -69,7 +74,7 @@ class UdemyIE(InfoExtractor):
             return compat_urlparse.urljoin(base_url, url) if not url.startswith('http') else url
 
         checkout_url = unescapeHTML(self._search_regex(
-            r'href=(["\'])(?P<url>(?:https?://(?:www\.)?udemy\.com)?/payment/checkout/.+?)\1',
+            r'href=(["\'])(?P<url>(?:https?://(?:www\.)?udemy\.com)?/(?:payment|cart)/checkout/.+?)\1',
             webpage, 'checkout url', group='url', default=None))
         if checkout_url:
             raise ExtractorError(
@@ -219,7 +224,7 @@ class UdemyIE(InfoExtractor):
 
         def extract_output_format(src, f_id):
             return {
-                'url': src['url'],
+                'url': src.get('url'),
                 'format_id': '%sp' % (src.get('height') or f_id),
                 'width': int_or_none(src.get('width')),
                 'height': int_or_none(src.get('height')),
@@ -264,6 +269,25 @@ class UdemyIE(InfoExtractor):
                     f = add_output_format_meta(f, format_id)
                 formats.append(f)
 
+        def extract_subtitles(track_list):
+            if not isinstance(track_list, list):
+                return
+            for track in track_list:
+                if not isinstance(track, dict):
+                    continue
+                if track.get('kind') != 'captions':
+                    continue
+                src = track.get('src')
+                if not src or not isinstance(src, compat_str):
+                    continue
+                lang = track.get('language') or track.get(
+                    'srclang') or track.get('label')
+                sub_dict = automatic_captions if track.get(
+                    'autogenerated') is True else subtitles
+                sub_dict.setdefault(lang, []).append({
+                    'url': src,
+                })
+
         download_urls = asset.get('download_urls')
         if isinstance(download_urls, dict):
             extract_formats(download_urls.get('Video'))
@@ -311,23 +335,16 @@ class UdemyIE(InfoExtractor):
                 extract_formats(data.get('sources'))
                 if not duration:
                     duration = int_or_none(data.get('duration'))
-                tracks = data.get('tracks')
-                if isinstance(tracks, list):
-                    for track in tracks:
-                        if not isinstance(track, dict):
-                            continue
-                        if track.get('kind') != 'captions':
-                            continue
-                        src = track.get('src')
-                        if not src or not isinstance(src, compat_str):
-                            continue
-                        lang = track.get('language') or track.get(
-                            'srclang') or track.get('label')
-                        sub_dict = automatic_captions if track.get(
-                            'autogenerated') is True else subtitles
-                        sub_dict.setdefault(lang, []).append({
-                            'url': src,
-                        })
+                extract_subtitles(data.get('tracks'))
+
+            if not subtitles and not automatic_captions:
+                text_tracks = self._parse_json(
+                    self._search_regex(
+                        r'text-tracks=(["\'])(?P<data>\[.+?\])\1', view_html,
+                        'text tracks', default='{}', group='data'), video_id,
+                    transform_source=lambda s: js_to_json(unescapeHTML(s)),
+                    fatal=False)
+                extract_subtitles(text_tracks)
 
         self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))