From 1666c2709f575a20c7ae1d0c76935b855e872dc9 Mon Sep 17 00:00:00 2001 From: Dan Walker Date: Sun, 25 Oct 2020 06:08:57 -0700 Subject: [PATCH 1/4] Fixed tiktok.py The JSON formatting of the response changed, corrected. #TESTING Successfully downloaded and played a tiktok video --- youtube_dlc/extractor/tiktok.py | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/youtube_dlc/extractor/tiktok.py b/youtube_dlc/extractor/tiktok.py index 52e5f4f1fa..f27aa42507 100644 --- a/youtube_dlc/extractor/tiktok.py +++ b/youtube_dlc/extractor/tiktok.py @@ -14,17 +14,17 @@ class TikTokBaseIE(InfoExtractor): def _extract_aweme(self, video_data, webpage, url): video_info = try_get( - video_data, lambda x: x['videoData']['itemInfos'], dict) + video_data, lambda x: x['itemInfo']['itemStruct'], dict) author_info = try_get( - video_data, lambda x: x['videoData']['authorInfos'], dict) - share_info = try_get(video_data, lambda x: x['shareMeta'], dict) + video_data, lambda x: x['itemInfo']['itemStruct']['author'], dict) + share_info = try_get(video_data, lambda x: x['itemInfo']['shareMeta'], dict) unique_id = str_or_none(author_info.get('uniqueId')) timestamp = try_get(video_info, lambda x: int(x['createTime']), int) date = datetime.fromtimestamp(timestamp).strftime('%Y%m%d') - height = try_get(video_info, lambda x: x['video']['videoMeta']['height'], int) - width = try_get(video_info, lambda x: x['video']['videoMeta']['width'], int) + height = try_get(video_info, lambda x: x['video']['height'], int) + width = try_get(video_info, lambda x: x['video']['width'], int) thumbnails = [] thumbnails.append({ 'url': video_info.get('thumbnail') or self._og_search_thumbnail(webpage), @@ -34,7 +34,7 @@ def _extract_aweme(self, video_data, webpage, url): formats = [] formats.append({ - 'url': try_get(video_info, lambda x: x['video']['urls'][0]), + 'url': try_get(video_info, lambda x: x['video']['playAddr']), 'ext': 'mp4', 'height': height, 'width': width From 4f5a0ad82214456c3d4832bbf4e15e416e1a16cb Mon Sep 17 00:00:00 2001 From: Unknown Date: Mon, 26 Oct 2020 02:15:46 +0100 Subject: [PATCH 2/4] [tiktok] fallback url + cookie --- youtube_dlc/extractor/tiktok.py | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/youtube_dlc/extractor/tiktok.py b/youtube_dlc/extractor/tiktok.py index f27aa42507..0cfd2168a6 100644 --- a/youtube_dlc/extractor/tiktok.py +++ b/youtube_dlc/extractor/tiktok.py @@ -12,7 +12,8 @@ class TikTokBaseIE(InfoExtractor): - def _extract_aweme(self, video_data, webpage, url): + def _extract_aweme(self, props_data, webpage, url): + video_data = try_get(props_data, lambda x: x['pageProps'], expected_type=dict) video_info = try_get( video_data, lambda x: x['itemInfo']['itemStruct'], dict) author_info = try_get( @@ -32,14 +33,20 @@ def _extract_aweme(self, video_data, webpage, url): 'height': height }) + url = '' + if not url: + url = try_get(video_info, lambda x: x['video']['playAddr']) + if not url: + url = try_get(video_info, lambda x: x['video']['downloadAddr']) formats = [] formats.append({ - 'url': try_get(video_info, lambda x: x['video']['playAddr']), + 'url': url, 'ext': 'mp4', 'height': height, 'width': width }) + tracker = try_get(props_data, lambda x: x['initialProps']['$wid']) return { 'comment_count': int_or_none(video_info.get('commentCount')), 'duration': try_get(video_info, lambda x: x['video']['videoMeta']['duration'], int), @@ -63,6 +70,7 @@ def _extract_aweme(self, video_data, webpage, url): 'formats': formats, 'http_headers': { 'Referer': url, + 'Cookie': 'tt_webid=%s; tt_webid_v2=%s' % (tracker, tracker), } } @@ -130,10 +138,10 @@ def _real_extract(self, url): r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P[^<]+)', webpage, 'json_string', group='json_string_ld') json_data = self._parse_json(json_string, video_id) - video_data = try_get(json_data, lambda x: x['props']['pageProps'], expected_type=dict) + props_data = try_get(json_data, lambda x: x['props'], expected_type=dict) # Chech statusCode for success - if video_data.get('statusCode') == 0: - return self._extract_aweme(video_data, webpage, url) + if props_data.get('pageProps').get('statusCode') == 0: + return self._extract_aweme(props_data, webpage, url) raise ExtractorError('Video not available', video_id=video_id) From fb2c9277bc16bbfb22ccc61bc913d9259f3e6ba9 Mon Sep 17 00:00:00 2001 From: Unknown Date: Mon, 26 Oct 2020 04:32:37 +0100 Subject: [PATCH 3/4] [youtube] ultimate age gate fix --- youtube_dlc/extractor/youtube.py | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) diff --git a/youtube_dlc/extractor/youtube.py b/youtube_dlc/extractor/youtube.py index b5b229ccbc..4fb49b864c 100644 --- a/youtube_dlc/extractor/youtube.py +++ b/youtube_dlc/extractor/youtube.py @@ -1742,6 +1742,13 @@ def extract_player_response(player_response, video_id): add_dash_mpd_pr(pl_response) return pl_response + def extract_embedded_config(embed_webpage, video_id): + embedded_config = self._search_regex( + r'setConfig\(({.*})\);', + embed_webpage, 'ytInitialData', default=None) + if embedded_config: + return embedded_config + player_response = {} # Get video info @@ -1755,8 +1762,17 @@ def extract_player_response(player_response, video_id): # this can be viewed without login into Youtube url = proto + '://www.youtube.com/embed/%s' % video_id embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage') - # check if video is only playable on youtube - if so it requires auth (cookies) - if re.search(r'player-unavailable">', embed_webpage) is not None: + ext = extract_embedded_config(embed_webpage, video_id) + # playabilityStatus = re.search(r'{\\\"status\\\":\\\"(?P[^\"]+)\\\"', ext) + playable_in_embed = re.search(r'{\\\"playableInEmbed\\\":(?P[^\,]+)', ext) + if not playable_in_embed: + self.to_screen('Could not determine whether playabale in embed for video %s' % video_id) + playable_in_embed = '' + else: + playable_in_embed = playable_in_embed.group('playableinEmbed') + # check if video is only playable on youtube in other words not playable in embed - if so it requires auth (cookies) + # if re.search(r'player-unavailable">', embed_webpage) is not None: + if playable_in_embed == 'false': ''' # TODO apply this patch when Support for Python 2.6(!) and above drops if ({'VISITOR_INFO1_LIVE', 'HSID', 'SSID', 'SID'} <= cookie_keys From 8045ef7760329db0e69cce0f9af3e1e9efbe1820 Mon Sep 17 00:00:00 2001 From: Unknown Date: Mon, 26 Oct 2020 04:34:25 +0100 Subject: [PATCH 4/4] next release: 2020.10.26 --- .github/ISSUE_TEMPLATE/1_broken_site.md | 6 +++--- .github/ISSUE_TEMPLATE/2_site_support_request.md | 4 ++-- .github/ISSUE_TEMPLATE/3_site_feature_request.md | 4 ++-- .github/ISSUE_TEMPLATE/4_bug_report.md | 6 +++--- .github/ISSUE_TEMPLATE/5_feature_request.md | 4 ++-- youtube_dlc/version.py | 2 +- 6 files changed, 13 insertions(+), 13 deletions(-) diff --git a/.github/ISSUE_TEMPLATE/1_broken_site.md b/.github/ISSUE_TEMPLATE/1_broken_site.md index 92db313a44..bf42510040 100644 --- a/.github/ISSUE_TEMPLATE/1_broken_site.md +++ b/.github/ISSUE_TEMPLATE/1_broken_site.md @@ -21,7 +21,7 @@ ## Checklist - [ ] I'm reporting a broken site support -- [ ] I've verified that I'm running youtube-dlc version **2020.10.24** +- [ ] I've verified that I'm running youtube-dlc version **2020.10.26** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped - [ ] I've searched the bugtracker for similar issues including closed ones @@ -44,7 +44,7 @@ ## Verbose log [debug] User config: [] [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj'] [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251 - [debug] youtube-dlc version 2020.10.24 + [debug] youtube-dlc version 2020.10.26 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4 [debug] Proxy map: {} diff --git a/.github/ISSUE_TEMPLATE/2_site_support_request.md b/.github/ISSUE_TEMPLATE/2_site_support_request.md index e70c853e1d..889005097b 100644 --- a/.github/ISSUE_TEMPLATE/2_site_support_request.md +++ b/.github/ISSUE_TEMPLATE/2_site_support_request.md @@ -21,7 +21,7 @@ ## Checklist - [ ] I'm reporting a new site support request -- [ ] I've verified that I'm running youtube-dlcc version **2020.10.24** +- [ ] I've verified that I'm running youtube-dlcc version **2020.10.26** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that none of provided URLs violate any copyrights - [ ] I've searched the bugtracker for similar site support requests including closed ones diff --git a/.github/ISSUE_TEMPLATE/3_site_feature_request.md b/.github/ISSUE_TEMPLATE/3_site_feature_request.md index f91c70835c..e5d7143887 100644 --- a/.github/ISSUE_TEMPLATE/3_site_feature_request.md +++ b/.github/ISSUE_TEMPLATE/3_site_feature_request.md @@ -21,13 +21,13 @@ ## Checklist - [ ] I'm reporting a site feature request -- [ ] I've verified that I'm running youtube-dlc version **2020.10.24** +- [ ] I've verified that I'm running youtube-dlc version **2020.10.26** - [ ] I've searched the bugtracker for similar site feature requests including closed ones diff --git a/.github/ISSUE_TEMPLATE/4_bug_report.md b/.github/ISSUE_TEMPLATE/4_bug_report.md index b34dbe8777..9de52f98c9 100644 --- a/.github/ISSUE_TEMPLATE/4_bug_report.md +++ b/.github/ISSUE_TEMPLATE/4_bug_report.md @@ -21,7 +21,7 @@ ## Checklist - [ ] I'm reporting a broken site support issue -- [ ] I've verified that I'm running youtube-dlc version **2020.10.24** +- [ ] I've verified that I'm running youtube-dlc version **2020.10.26** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped - [ ] I've searched the bugtracker for similar bug reports including closed ones @@ -46,7 +46,7 @@ ## Verbose log [debug] User config: [] [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj'] [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251 - [debug] youtube-dlc version 2020.10.24 + [debug] youtube-dlc version 2020.10.26 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4 [debug] Proxy map: {} diff --git a/.github/ISSUE_TEMPLATE/5_feature_request.md b/.github/ISSUE_TEMPLATE/5_feature_request.md index 8646df1608..86fac96dde 100644 --- a/.github/ISSUE_TEMPLATE/5_feature_request.md +++ b/.github/ISSUE_TEMPLATE/5_feature_request.md @@ -21,13 +21,13 @@ ## Checklist - [ ] I'm reporting a feature request -- [ ] I've verified that I'm running youtube-dlc version **2020.10.24** +- [ ] I've verified that I'm running youtube-dlc version **2020.10.26** - [ ] I've searched the bugtracker for similar feature requests including closed ones diff --git a/youtube_dlc/version.py b/youtube_dlc/version.py index 5eb581675c..440d8e4882 100644 --- a/youtube_dlc/version.py +++ b/youtube_dlc/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2020.10.24-5' +__version__ = '2020.10.25'