From dcc8522fdba4c9286ebc0548caf05b425bc68773 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 9 Apr 2020 02:11:19 +0700 Subject: [PATCH 001/156] [motherless] Fix extraction (closes #24699) --- youtube_dl/extractor/motherless.py | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/youtube_dl/extractor/motherless.py b/youtube_dl/extractor/motherless.py index 43fd70f11..b1615b4d8 100644 --- a/youtube_dl/extractor/motherless.py +++ b/youtube_dl/extractor/motherless.py @@ -26,7 +26,7 @@ class MotherlessIE(InfoExtractor): 'categories': ['Gaming', 'anal', 'reluctant', 'rough', 'Wife'], 'upload_date': '20100913', 'uploader_id': 'famouslyfuckedup', - 'thumbnail': r're:http://.*\.jpg', + 'thumbnail': r're:https?://.*\.jpg', 'age_limit': 18, } }, { @@ -40,7 +40,7 @@ class MotherlessIE(InfoExtractor): 'game', 'hairy'], 'upload_date': '20140622', 'uploader_id': 'Sulivana7x', - 'thumbnail': r're:http://.*\.jpg', + 'thumbnail': r're:https?://.*\.jpg', 'age_limit': 18, }, 'skip': '404', @@ -54,7 +54,7 @@ class MotherlessIE(InfoExtractor): 'categories': ['superheroine heroine superher'], 'upload_date': '20140827', 'uploader_id': 'shade0230', - 'thumbnail': r're:http://.*\.jpg', + 'thumbnail': r're:https?://.*\.jpg', 'age_limit': 18, } }, { @@ -76,7 +76,8 @@ class MotherlessIE(InfoExtractor): raise ExtractorError('Video %s is for friends only' % video_id, expected=True) title = self._html_search_regex( - r'id="view-upload-title">\s+([^<]+)<', webpage, 'title') + (r'(?s)]+\bclass=["\']media-meta-title[^>]+>(.+?)', + r'id="view-upload-title">\s+([^<]+)<'), webpage, 'title') video_url = (self._html_search_regex( (r'setup\(\{\s*["\']file["\']\s*:\s*(["\'])(?P(?:(?!\1).)+)\1', r'fileurl\s*=\s*(["\'])(?P(?:(?!\1).)+)\1'), @@ -84,14 +85,15 @@ class MotherlessIE(InfoExtractor): or 'http://cdn4.videos.motherlessmedia.com/videos/%s.mp4?fs=opencloud' % video_id) age_limit = self._rta_search(webpage) view_count = str_to_int(self._html_search_regex( - r'Views\s+([^<]+)<', + (r'>(\d+)\s+Views<', r'Views\s+([^<]+)<'), webpage, 'view count', fatal=False)) like_count = str_to_int(self._html_search_regex( - r'Favorited\s+([^<]+)<', + (r'>(\d+)\s+Favorites<', r'Favorited\s+([^<]+)<'), webpage, 'like count', fatal=False)) upload_date = self._html_search_regex( - r'Uploaded\s+([^<]+)<', webpage, 'upload date') + (r'class=["\']count[^>]+>(\d+\s+[a-zA-Z]{3}\s+\d{4})<', + r'Uploaded\s+([^<]+)<'), webpage, 'upload date') if 'Ago' in upload_date: days = int(re.search(r'([0-9]+)', upload_date).group(1)) upload_date = (datetime.datetime.now() - datetime.timedelta(days=days)).strftime('%Y%m%d') From 5caf88ccb4bfe3d1b53885b78b2bc509ba333f15 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 9 Apr 2020 03:52:29 +0700 Subject: [PATCH 002/156] [nova:embed] Fix extraction (closes #24700) --- youtube_dl/extractor/nova.py | 106 +++++++++++++++++++++++------------ 1 file changed, 71 insertions(+), 35 deletions(-) diff --git a/youtube_dl/extractor/nova.py b/youtube_dl/extractor/nova.py index 2850af5db..47b9748f0 100644 --- a/youtube_dl/extractor/nova.py +++ b/youtube_dl/extractor/nova.py @@ -6,6 +6,7 @@ import re from .common import InfoExtractor from ..utils import ( clean_html, + determine_ext, int_or_none, js_to_json, qualities, @@ -33,42 +34,76 @@ class NovaEmbedIE(InfoExtractor): webpage = self._download_webpage(url, video_id) - bitrates = self._parse_json( - self._search_regex( - r'(?s)(?:src|bitrates)\s*=\s*({.+?})\s*;', webpage, 'formats'), - video_id, transform_source=js_to_json) - - QUALITIES = ('lq', 'mq', 'hq', 'hd') - quality_key = qualities(QUALITIES) - + duration = None formats = [] - for format_id, format_list in bitrates.items(): - if not isinstance(format_list, list): - format_list = [format_list] - for format_url in format_list: - format_url = url_or_none(format_url) - if not format_url: - continue - if format_id == 'hls': - formats.extend(self._extract_m3u8_formats( - format_url, video_id, ext='mp4', - entry_protocol='m3u8_native', m3u8_id='hls', - fatal=False)) - continue - f = { - 'url': format_url, - } - f_id = format_id - for quality in QUALITIES: - if '%s.mp4' % quality in format_url: - f_id += '-%s' % quality - f.update({ - 'quality': quality_key(quality), - 'format_note': quality.upper(), + + player = self._parse_json( + self._search_regex( + r'Player\.init\s*\([^,]+,\s*({.+?})\s*,\s*{.+?}\s*\)\s*;', + webpage, 'player', default='{}'), video_id, fatal=False) + if player: + for format_id, format_list in player['tracks'].items(): + if not isinstance(format_list, list): + format_list = [format_list] + for format_dict in format_list: + if not isinstance(format_dict, dict): + continue + format_url = url_or_none(format_dict.get('src')) + format_type = format_dict.get('type') + ext = determine_ext(format_url) + if (format_type == 'application/x-mpegURL' + or format_id == 'HLS' or ext == 'm3u8'): + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', + entry_protocol='m3u8_native', m3u8_id='hls', + fatal=False)) + elif (format_type == 'application/dash+xml' + or format_id == 'DASH' or ext == 'mpd'): + formats.extend(self._extract_mpd_formats( + format_url, video_id, mpd_id='dash', fatal=False)) + else: + formats.append({ + 'url': format_url, }) - break - f['format_id'] = f_id - formats.append(f) + duration = int_or_none(player.get('duration')) + else: + # Old path, not actual as of 08.04.2020 + bitrates = self._parse_json( + self._search_regex( + r'(?s)(?:src|bitrates)\s*=\s*({.+?})\s*;', webpage, 'formats'), + video_id, transform_source=js_to_json) + + QUALITIES = ('lq', 'mq', 'hq', 'hd') + quality_key = qualities(QUALITIES) + + for format_id, format_list in bitrates.items(): + if not isinstance(format_list, list): + format_list = [format_list] + for format_url in format_list: + format_url = url_or_none(format_url) + if not format_url: + continue + if format_id == 'hls': + formats.extend(self._extract_m3u8_formats( + format_url, video_id, ext='mp4', + entry_protocol='m3u8_native', m3u8_id='hls', + fatal=False)) + continue + f = { + 'url': format_url, + } + f_id = format_id + for quality in QUALITIES: + if '%s.mp4' % quality in format_url: + f_id += '-%s' % quality + f.update({ + 'quality': quality_key(quality), + 'format_note': quality.upper(), + }) + break + f['format_id'] = f_id + formats.append(f) + self._sort_formats(formats) title = self._og_search_title( @@ -81,7 +116,8 @@ class NovaEmbedIE(InfoExtractor): r'poster\s*:\s*(["\'])(?P(?:(?!\1).)+)\1', webpage, 'thumbnail', fatal=False, group='value') duration = int_or_none(self._search_regex( - r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False)) + r'videoDuration\s*:\s*(\d+)', webpage, 'duration', + default=duration)) return { 'id': video_id, From 6b09401b0ba95da5669d249c8930b3adb873d96e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 9 Apr 2020 22:42:43 +0700 Subject: [PATCH 003/156] [youtube] Skip broken multifeed videos (closes #24711) --- youtube_dl/extractor/youtube.py | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 908defecd..633b839e0 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -1840,15 +1840,26 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # fields may contain comma as well (see # https://github.com/ytdl-org/youtube-dl/issues/8536) feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed)) + + def feed_entry(name): + return try_get(feed_data, lambda x: x[name][0], compat_str) + + feed_id = feed_entry('id') + if not feed_id: + continue + feed_title = feed_entry('title') + title = video_title + if feed_title: + title += ' (%s)' % feed_title entries.append({ '_type': 'url_transparent', 'ie_key': 'Youtube', 'url': smuggle_url( '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]), {'force_singlefeed': True}), - 'title': '%s (%s)' % (video_title, feed_data['title'][0]), + 'title': title, }) - feed_ids.append(feed_data['id'][0]) + feed_ids.append(feed_id) self.to_screen( 'Downloading multifeed video (%s) - add --no-playlist to just download video %s' % (', '.join(feed_ids), video_id)) From b9e5f872916a7d753ae237459b10622c1c2c3471 Mon Sep 17 00:00:00 2001 From: tom Date: Thu, 9 Apr 2020 21:50:45 +1000 Subject: [PATCH 004/156] [soundcloud] Extract AAC format --- youtube_dl/extractor/soundcloud.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index ff6be0b54..02d56184d 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -350,6 +350,8 @@ class SoundcloudIE(InfoExtractor): format_id_list = [] if protocol: format_id_list.append(protocol) + if f.get('ext') == 'aac': + f['abr'] = '256' for k in ('ext', 'abr'): v = f.get(k) if v: From 75294a5ed03f4443970478f3f4eac572239cec45 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Fri, 10 Apr 2020 17:24:21 +0700 Subject: [PATCH 005/156] [soundcloud] Improve AAC format extraction (closes #19173, closes #24708) --- youtube_dl/extractor/soundcloud.py | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index 02d56184d..422ce1626 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -246,7 +246,12 @@ class SoundcloudIE(InfoExtractor): 'comment_count': int, 'repost_count': int, }, - } + }, + { + # with AAC HQ format available via OAuth token + 'url': 'https://soundcloud.com/wandw/the-chainsmokers-ft-daya-dont-let-me-down-ww-remix-1', + 'only_matching': True, + }, ] _API_V2_BASE = 'https://api-v2.soundcloud.com/' @@ -350,7 +355,8 @@ class SoundcloudIE(InfoExtractor): format_id_list = [] if protocol: format_id_list.append(protocol) - if f.get('ext') == 'aac': + ext = f.get('ext') + if ext == 'aac': f['abr'] = '256' for k in ('ext', 'abr'): v = f.get(k) @@ -362,9 +368,13 @@ class SoundcloudIE(InfoExtractor): abr = f.get('abr') if abr: f['abr'] = int(abr) + if protocol == 'hls': + protocol = 'm3u8' if ext == 'aac' else 'm3u8_native' + else: + protocol = 'http' f.update({ 'format_id': '_'.join(format_id_list), - 'protocol': 'm3u8_native' if protocol == 'hls' else 'http', + 'protocol': protocol, 'preference': -10 if preview else None, }) formats.append(f) From 533f3e3557af85e28afd72d291cb51a769c7dd7a Mon Sep 17 00:00:00 2001 From: AndrewMBL <62922222+AndrewMBL@users.noreply.github.com> Date: Tue, 31 Mar 2020 15:25:04 +1100 Subject: [PATCH 006/156] [thisoldhouse] Fix video id extraction (closes #24548) Added support for: with of without "www." and either ".chorus.build" or ".com" It now validated correctly on older URL's ```