From 57a38a38c32ea2eb1ca54ee4ba3fcd31a9b7f328 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Fri, 4 Aug 2017 23:44:07 +0700 Subject: [PATCH] [udemy] Fix subtitles extraction (closes #13812) --- youtube_dl/extractor/udemy.py | 47 ++++++++++++++++++++++------------- 1 file changed, 30 insertions(+), 17 deletions(-) diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py index 160be1b1b..3b02f43e3 100644 --- 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, @@ -268,6 +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')) @@ -315,23 +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\[.+?\])\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')) -- 2.22.2