New upstream version 2018.04.25

[youtubedl] / youtube_dl / extractor / udemy.py

diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py
index dae1aa3c6ba6fd211233abbd316fb78409c33faf..6d6c0a98fa64e9e2afc68ce2ad569f5a91d5c24b 100644 (file)
--- a/youtube_dl/extractor/udemy.py
+++ b/youtube_dl/extractor/udemy.py
@@ -5,6 +5,7 @@ import re
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
+    compat_kwargs,
     compat_str,
     compat_urllib_request,
     compat_urlparse,
@@ -15,6 +16,7 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     int_or_none,
+    js_to_json,
     sanitized_Request,
     unescapeHTML,
     urlencode_postdata,
@@ -52,16 +54,20 @@ 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):
         course = self._parse_json(
             unescapeHTML(self._search_regex(
-                r'ng-init=["\'].*\bcourse=({.+?});', webpage, 'course', default='{}')),
+                r'ng-init=["\'].*\bcourse=({.+?})[;"\']',
+                webpage, 'course', default='{}')),
             video_id, fatal=False) or {}
         course_id = course.get('id') or self._search_regex(
-            (r'"id"\s*:\s*(\d+)', r'data-course-id=["\'](\d+)'),
-            webpage, 'course id')
+            r'data-course-id=["\'](\d+)', webpage, 'course id')
         return course_id, course.get('title')
 
     def _enroll_course(self, base_url, webpage, course_id):
@@ -69,7 +75,7 @@ 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(
@@ -109,6 +115,11 @@ class UdemyIE(InfoExtractor):
                 error_str += ' - %s' % error_data.get('formErrors')
             raise ExtractorError(error_str, expected=True)
 
+    def _download_webpage(self, *args, **kwargs):
+        kwargs.setdefault('headers', {})['User-Agent'] = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_5) AppleWebKit/603.2.4 (KHTML, like Gecko) Version/10.1.1 Safari/603.2.4'
+        return super(UdemyIE, self)._download_webpage(
+            *args, **compat_kwargs(kwargs))
+
     def _download_json(self, url_or_request, *args, **kwargs):
         headers = {
             'X-Udemy-Snail-Case': 'true',
@@ -159,7 +170,7 @@ class UdemyIE(InfoExtractor):
         })
 
         response = self._download_webpage(
-            self._LOGIN_URL, None, 'Logging in as %s' % username,
+            self._LOGIN_URL, None, 'Logging in',
             data=urlencode_postdata(login_form),
             headers={
                 'Referer': self._ORIGIN_URL,
@@ -219,7 +230,7 @@ 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')),
@@ -252,6 +263,11 @@ class UdemyIE(InfoExtractor):
                 video_url = source.get('file') or source.get('src')
                 if not video_url or not isinstance(video_url, compat_str):
                     continue
+                if source.get('type') == 'application/x-mpegURL' or determine_ext(video_url) == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        video_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                        m3u8_id='hls', fatal=False))
+                    continue
                 format_id = source.get('label')
                 f = {
                     'url': video_url,
@@ -264,6 +280,25 @@ 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 +346,16 @@ 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'))