Update changelog.

[youtubedl] / youtube_dl / extractor / udemy.py

diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py
index dae1aa3c6ba6fd211233abbd316fb78409c33faf..79c45f80e0f827d07a2ef362597b6e7fbab122e7 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,
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,

+    compat_kwargs,

     compat_str,
     compat_urllib_request,
     compat_urlparse,
     compat_str,
     compat_urllib_request,
     compat_urlparse,


@@ -15,8 +16,11 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     int_or_none,
     ExtractorError,
     float_or_none,
     int_or_none,

+    js_to_json,

     sanitized_Request,
     sanitized_Request,

+    try_get,

     unescapeHTML,
     unescapeHTML,

+    url_or_none,

     urlencode_postdata,
 )
 
     urlencode_postdata,
 )
 


@@ -52,16 +56,24 @@ 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,
         # 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,
+    }, {
+        # only outputs rendition
+        'url': 'https://www.udemy.com/how-you-can-help-your-local-community-5-amazing-examples/learn/v4/t/lecture/3225750?start=0',
+        'only_matching': True,

     }]
 
     def _extract_course_info(self, webpage, video_id):
         course = self._parse_json(
             unescapeHTML(self._search_regex(
     }]
 
     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(
             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):
         return course_id, course.get('title')
 
     def _enroll_course(self, base_url, webpage, course_id):


@@ -69,7 +81,7 @@ class UdemyIE(InfoExtractor):
             return compat_urlparse.urljoin(base_url, url) if not url.startswith('http') else url
 
         checkout_url = unescapeHTML(self._search_regex(
             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(
             webpage, 'checkout url', group='url', default=None))
         if checkout_url:
             raise ExtractorError(


@@ -95,7 +107,7 @@ class UdemyIE(InfoExtractor):
             % (course_id, lecture_id),
             lecture_id, 'Downloading lecture JSON', query={
                 'fields[lecture]': 'title,description,view_html,asset',
             % (course_id, lecture_id),
             lecture_id, 'Downloading lecture JSON', query={
                 'fields[lecture]': 'title,description,view_html,asset',

-                'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data',
+                'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,stream_urls,captions,data',

             })
 
     def _handle_error(self, response):
             })
 
     def _handle_error(self, response):


@@ -109,6 +121,11 @@ class UdemyIE(InfoExtractor):
                 error_str += ' - %s' % error_data.get('formErrors')
             raise ExtractorError(error_str, expected=True)
 
                 error_str += ' - %s' % error_data.get('formErrors')
             raise ExtractorError(error_str, expected=True)
 

+    def _download_webpage_handle(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_handle(
+            *args, **compat_kwargs(kwargs))
+

     def _download_json(self, url_or_request, *args, **kwargs):
         headers = {
             'X-Udemy-Snail-Case': 'true',
     def _download_json(self, url_or_request, *args, **kwargs):
         headers = {
             'X-Udemy-Snail-Case': 'true',


@@ -135,7 +152,7 @@ class UdemyIE(InfoExtractor):
         self._login()
 
     def _login(self):
         self._login()
 
     def _login(self):

-        (username, password) = self._get_login_info()
+        username, password = self._get_login_info()

         if username is None:
             return
 
         if username is None:
             return
 


@@ -159,7 +176,7 @@ class UdemyIE(InfoExtractor):
         })
 
         response = self._download_webpage(
         })
 
         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,
             data=urlencode_postdata(login_form),
             headers={
                 'Referer': self._ORIGIN_URL,


@@ -219,7 +236,7 @@ class UdemyIE(InfoExtractor):
 
         def extract_output_format(src, f_id):
             return {
 
         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')),
                 'format_id': '%sp' % (src.get('height') or f_id),
                 'width': int_or_none(src.get('width')),
                 'height': int_or_none(src.get('height')),


@@ -249,8 +266,13 @@ class UdemyIE(InfoExtractor):
             if not isinstance(source_list, list):
                 return
             for source in source_list:
             if not isinstance(source_list, list):
                 return
             for source in source_list:

-                video_url = source.get('file') or source.get('src')
-                if not video_url or not isinstance(video_url, compat_str):
+                video_url = url_or_none(source.get('file') or source.get('src'))
+                if not video_url:
+                    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 = {
                     continue
                 format_id = source.get('label')
                 f = {


@@ -264,9 +286,44 @@ class UdemyIE(InfoExtractor):
                     f = add_output_format_meta(f, format_id)
                 formats.append(f)
 
                     f = add_output_format_meta(f, format_id)
                 formats.append(f)
 

-        download_urls = asset.get('download_urls')
-        if isinstance(download_urls, dict):
-            extract_formats(download_urls.get('Video'))
+        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 = url_or_none(track.get('src'))
+                if not src:
+                    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,
+                })
+
+        for url_kind in ('download', 'stream'):
+            urls = asset.get('%s_urls' % url_kind)
+            if isinstance(urls, dict):
+                extract_formats(urls.get('Video'))
+
+        captions = asset.get('captions')
+        if isinstance(captions, list):
+            for cc in captions:
+                if not isinstance(cc, dict):
+                    continue
+                cc_url = url_or_none(cc.get('url'))
+                if not cc_url:
+                    continue
+                lang = try_get(cc, lambda x: x['locale']['locale'], compat_str)
+                sub_dict = (automatic_captions if cc.get('source') == 'auto'
+                            else subtitles)
+                sub_dict.setdefault(lang or 'en', []).append({
+                    'url': cc_url,
+                })

 
         view_html = lecture.get('view_html')
         if view_html:
 
         view_html = lecture.get('view_html')
         if view_html:


@@ -311,23 +368,22 @@ class UdemyIE(InfoExtractor):
                 extract_formats(data.get('sources'))
                 if not duration:
                     duration = int_or_none(data.get('duration'))
                 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)
+
+        if not formats and outputs:
+            for format_id, output in outputs.items():
+                f = extract_output_format(output, format_id)
+                if f.get('url'):
+                    formats.append(f)

 
         self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))
 
 
         self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))