diff options
Diffstat (limited to 'youtube_dl/extractor/udemy.py')
| -rw-r--r-- | youtube_dl/extractor/udemy.py | 79 | 
1 files changed, 58 insertions, 21 deletions
diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py index cce29c6e0..dae1aa3c6 100644 --- a/youtube_dl/extractor/udemy.py +++ b/youtube_dl/extractor/udemy.py @@ -212,12 +212,15 @@ class UdemyIE(InfoExtractor):          thumbnail = asset.get('thumbnail_url') or asset.get('thumbnailUrl')          duration = float_or_none(asset.get('data', {}).get('duration')) +        subtitles = {} +        automatic_captions = {} +          formats = [] -        def extract_output_format(src): +        def extract_output_format(src, f_id):              return {                  'url': src['url'], -                'format_id': '%sp' % (src.get('height') or format_id), +                'format_id': '%sp' % (src.get('height') or f_id),                  'width': int_or_none(src.get('width')),                  'height': int_or_none(src.get('height')),                  'vbr': int_or_none(src.get('video_bitrate_in_kbps')), @@ -237,30 +240,33 @@ class UdemyIE(InfoExtractor):          def add_output_format_meta(f, key):              output = outputs.get(key)              if isinstance(output, dict): -                output_format = extract_output_format(output) +                output_format = extract_output_format(output, key)                  output_format.update(f)                  return output_format              return f +        def extract_formats(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): +                    continue +                format_id = source.get('label') +                f = { +                    'url': video_url, +                    'format_id': '%sp' % format_id, +                    'height': int_or_none(format_id), +                } +                if format_id: +                    # Some videos contain additional metadata (e.g. +                    # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208) +                    f = add_output_format_meta(f, format_id) +                formats.append(f) +          download_urls = asset.get('download_urls')          if isinstance(download_urls, dict): -            video = download_urls.get('Video') -            if isinstance(video, list): -                for format_ in video: -                    video_url = format_.get('file') -                    if not video_url: -                        continue -                    format_id = format_.get('label') -                    f = { -                        'url': format_['file'], -                        'format_id': '%sp' % format_id, -                        'height': int_or_none(format_id), -                    } -                    if format_id: -                        # Some videos contain additional metadata (e.g. -                        # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208) -                        f = add_output_format_meta(f, format_id) -                    formats.append(f) +            extract_formats(download_urls.get('Video'))          view_html = lecture.get('view_html')          if view_html: @@ -294,6 +300,35 @@ class UdemyIE(InfoExtractor):                          'height': height,                      }, res)) +            # react rendition since 2017.04.15 (see +            # https://github.com/rg3/youtube-dl/issues/12744) +            data = self._parse_json( +                self._search_regex( +                    r'videojs-setup-data=(["\'])(?P<data>{.+?})\1', view_html, +                    'setup data', default='{}', group='data'), video_id, +                transform_source=unescapeHTML, fatal=False) +            if data and isinstance(data, dict): +                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, +                        }) +          self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))          return { @@ -302,7 +337,9 @@ class UdemyIE(InfoExtractor):              'description': description,              'thumbnail': thumbnail,              'duration': duration, -            'formats': formats +            'formats': formats, +            'subtitles': subtitles, +            'automatic_captions': automatic_captions,          }  | 
