From 8b4b400aef83b233502ece7321ee84f6ab9e213e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Tue, 19 Jun 2018 23:00:36 +0700 Subject: [PATCH] [peertube] Improve generic support (closes #16733) --- youtube_dl/extractor/generic.py | 2 +- youtube_dl/extractor/peertube.py | 47 +++++++++++++++++++++++--------- 2 files changed, 35 insertions(+), 14 deletions(-) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index dad951b75..6c0f772ac 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -3076,7 +3076,7 @@ def _real_extract(self, url): return self.playlist_from_matches( cloudflarestream_urls, video_id, video_title, ie=CloudflareStreamIE.ie_key()) - peertube_urls = PeerTubeIE._extract_urls(webpage) + peertube_urls = PeerTubeIE._extract_urls(webpage, url) if peertube_urls: return self.playlist_from_matches( peertube_urls, video_id, video_title, ie=PeerTubeIE.ie_key()) diff --git a/youtube_dl/extractor/peertube.py b/youtube_dl/extractor/peertube.py index a481b3151..d9849a2ba 100644 --- a/youtube_dl/extractor/peertube.py +++ b/youtube_dl/extractor/peertube.py @@ -116,12 +116,14 @@ class PeerTubeIE(InfoExtractor): videos\.tcit\.fr| peertube\.cpy\.re )''' + _UUID_RE = r'[\da-fA-F]{8}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{12}' _VALID_URL = r'''(?x) - https?:// - %s - /(?:videos/(?:watch|embed)|api/v\d/videos)/ - (?P[^/?\#&]+) - ''' % _INSTANCES_RE + (?: + peertube:(?P[^:]+):| + https?://(?P%s)/(?:videos/(?:watch|embed)|api/v\d/videos)/ + ) + (?P%s) + ''' % (_INSTANCES_RE, _UUID_RE) _TESTS = [{ 'url': 'https://peertube.moe/videos/watch/2790feb0-8120-4e63-9af3-c943c69f5e6c', 'md5': '80f24ff364cc9d333529506a263e7feb', @@ -157,21 +159,40 @@ class PeerTubeIE(InfoExtractor): }, { 'url': 'https://tube.openalgeria.org/api/v1/videos/c1875674-97d0-4c94-a058-3f7e64c962e8', 'only_matching': True, + }, { + 'url': 'peertube:video.blender.org:b37a5b9f-e6b5-415c-b700-04a5cd6ec205', + 'only_matching': True, }] @staticmethod - def _extract_urls(webpage): - return [ - mobj.group('url') - for mobj in re.finditer( - r'''(?x)]+\bsrc=(["\'])(?P(?:https?:)?//%s/videos/embed/[^/?\#&]+)\1''' - % PeerTubeIE._INSTANCES_RE, webpage)] + def _extract_peertube_url(webpage, source_url): + mobj = re.match( + r'https?://(?P[^/]+)/videos/watch/(?P%s)' + % PeerTubeIE._UUID_RE, source_url) + if mobj and any(p in webpage for p in ( + 'PeerTube<', + 'There will be other non JS-based clients to access PeerTube', + '>We are sorry but it seems that PeerTube is not compatible with your web browser.<')): + return 'peertube:%s:%s' % mobj.group('host', 'id') + + @staticmethod + def _extract_urls(webpage, source_url): + entries = re.findall( + r'''(?x)<iframe[^>]+\bsrc=["\'](?P<url>(?:https?:)?//%s/videos/embed/%s)''' + % (PeerTubeIE._INSTANCES_RE, PeerTubeIE._UUID_RE), webpage) + if not entries: + peertube_url = PeerTubeIE._extract_peertube_url(webpage, source_url) + if peertube_url: + entries = [peertube_url] + return entries def _real_extract(self, url): - video_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + host = mobj.group('host') or mobj.group('host_2') + video_id = mobj.group('id') video = self._download_json( - urljoin(url, '/api/v1/videos/%s' % video_id), video_id) + 'https://%s/api/v1/videos/%s' % (host, video_id), video_id) title = video['name']