From bf7fa94ec7202bde963a75ab903996ac575910db Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Fri, 23 Jan 2015 16:31:52 +0100 Subject: [PATCH 1/4] [downloader/f4m] build_fragments_list: Support videos with more than 1 segment --- youtube_dl/downloader/f4m.py | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py index c460c167a..c68b2c303 100644 --- a/youtube_dl/downloader/f4m.py +++ b/youtube_dl/downloader/f4m.py @@ -177,13 +177,12 @@ def build_fragments_list(boot_info): """ Return a list of (segment, fragment) for each fragment in the video """ res = [] segment_run_table = boot_info['segments'][0] - # I've only found videos with one segment - segment_run_entry = segment_run_table['segment_run'][0] - n_frags = segment_run_entry[1] fragment_run_entry_table = boot_info['fragments'][0]['fragments'] first_frag_number = fragment_run_entry_table[0]['first'] - for (i, frag_number) in zip(range(1, n_frags + 1), itertools.count(first_frag_number)): - res.append((1, frag_number)) + fragments_counter = itertools.count(first_frag_number) + for segment, fragments_count in segment_run_table['segment_run']: + for _ in range(fragments_count): + res.append((segment, next(fragments_counter))) return res From 0920e5830f890580ec16cdd10bfe8def73a1a09f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Fri, 23 Jan 2015 16:39:23 +0100 Subject: [PATCH 2/4] [atresplayer] Don't include f4m formats if they are protected by DRM (fixes #4705) --- youtube_dl/extractor/atresplayer.py | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/youtube_dl/extractor/atresplayer.py b/youtube_dl/extractor/atresplayer.py index 5db1941b3..37321ef1d 100644 --- a/youtube_dl/extractor/atresplayer.py +++ b/youtube_dl/extractor/atresplayer.py @@ -107,7 +107,14 @@ class AtresPlayerIE(InfoExtractor): for _, video_url in fmt_json['resultObject'].items(): if video_url.endswith('/Manifest'): - formats.extend(self._extract_f4m_formats(video_url[:-9] + '/manifest.f4m', video_id)) + if 'geodeswowsmpra3player' in video_url: + f4m_path = video_url.split('smil:', 1)[-1].split('free_', 1)[0] + f4m_url = 'http://drg.antena3.com/{0}hds/es/sd.f4m'.format(f4m_path) + # this videos are protected by DRM, the f4m downloader doesn't support them + continue + else: + f4m_url = video_url[:-9] + '/manifest.f4m' + formats.extend(self._extract_f4m_formats(f4m_url, video_id)) else: formats.append({ 'url': video_url, From c2e64f71d075bbc916f343916ff1e679f642a821 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Fri, 23 Jan 2015 21:58:40 +0600 Subject: [PATCH 3/4] [twitch] Add support for bookmarks --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/twitch.py | 41 ++++++++++++++++++++++++++++---- 2 files changed, 37 insertions(+), 5 deletions(-) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 9ab90ac62..9e1ce5db3 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -467,6 +467,7 @@ from .twitch import ( TwitchVodIE, TwitchProfileIE, TwitchPastBroadcastsIE, + TwitchBookmarksIE, TwitchStreamIE, ) from .ubu import UbuIE diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index 340cadcf5..741df7cbc 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -220,12 +220,18 @@ class TwitchPlaylistBaseIE(TwitchBaseIE): response = self._download_json( self._PLAYLIST_URL % (channel_id, offset, limit), channel_id, 'Downloading %s videos JSON page %d' % (self._PLAYLIST_TYPE, counter)) - videos = response['videos'] - if not videos: + page_entries = self._extract_playlist_page(response) + if not page_entries: break - entries.extend([self.url_result(video['url']) for video in videos]) + entries.extend(page_entries) offset += limit - return self.playlist_result(entries, channel_id, channel_name) + return self.playlist_result( + [self.url_result(entry) for entry in set(entries)], + channel_id, channel_name) + + def _extract_playlist_page(self, response): + videos = response.get('videos') + return [video['url'] for video in videos] if videos else [] def _real_extract(self, url): return self._extract_playlist(self._match_id(url)) @@ -262,6 +268,31 @@ class TwitchPastBroadcastsIE(TwitchPlaylistBaseIE): } +class TwitchBookmarksIE(TwitchPlaylistBaseIE): + IE_NAME = 'twitch:bookmarks' + _VALID_URL = r'%s/(?P[^/]+)/profile/bookmarks/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE + _PLAYLIST_URL = '%s/api/bookmark/?user=%%s&offset=%%d&limit=%%d' % TwitchBaseIE._API_BASE + _PLAYLIST_TYPE = 'bookmarks' + + _TEST = { + 'url': 'http://www.twitch.tv/ognos/profile/bookmarks', + 'info_dict': { + 'id': 'ognos', + 'title': 'Ognos', + }, + 'playlist_mincount': 3, + } + + def _extract_playlist_page(self, response): + entries = [] + for bookmark in response.get('bookmarks', []): + video = bookmark.get('video') + if not video: + continue + entries.append(video['url']) + return entries + + class TwitchStreamIE(TwitchBaseIE): IE_NAME = 'twitch:stream' _VALID_URL = r'%s/(?P[^/]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE @@ -348,4 +379,4 @@ class TwitchStreamIE(TwitchBaseIE): 'view_count': view_count, 'formats': formats, 'is_live': True, - } + } \ No newline at end of file From ebd46aed5119899826629cf751ba5abe7a65d50b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Fri, 23 Jan 2015 22:21:55 +0600 Subject: [PATCH 4/4] [atresplayer] Filter URLs and clarify android format ids --- youtube_dl/extractor/atresplayer.py | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/youtube_dl/extractor/atresplayer.py b/youtube_dl/extractor/atresplayer.py index 37321ef1d..8fd69b971 100644 --- a/youtube_dl/extractor/atresplayer.py +++ b/youtube_dl/extractor/atresplayer.py @@ -105,7 +105,9 @@ class AtresPlayerIE(InfoExtractor): raise ExtractorError( '%s returned error: %s' % (self.IE_NAME, result), expected=True) - for _, video_url in fmt_json['resultObject'].items(): + for format_id, video_url in fmt_json['resultObject'].items(): + if format_id == 'token' or not video_url.startswith('http'): + continue if video_url.endswith('/Manifest'): if 'geodeswowsmpra3player' in video_url: f4m_path = video_url.split('smil:', 1)[-1].split('free_', 1)[0] @@ -118,7 +120,7 @@ class AtresPlayerIE(InfoExtractor): else: formats.append({ 'url': video_url, - 'format_id': 'android', + 'format_id': 'android-%s' % format_id, 'preference': 1, }) self._sort_formats(formats)