From c9e12a618c9420c2bb21c09bf47b9469785f492e Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Sun, 27 May 2018 12:10:12 +0100 Subject: [PATCH] [9c9media] extract mpd formats and subtitles --- youtube_dl/extractor/ctvnews.py | 4 +- youtube_dl/extractor/extractors.py | 5 +- youtube_dl/extractor/ninecninemedia.py | 93 ++++++++++---------------- youtube_dl/extractor/rds.py | 2 +- 4 files changed, 41 insertions(+), 63 deletions(-) diff --git a/youtube_dl/extractor/ctvnews.py b/youtube_dl/extractor/ctvnews.py index 55a127b76..03f8cefb7 100644 --- a/youtube_dl/extractor/ctvnews.py +++ b/youtube_dl/extractor/ctvnews.py @@ -11,10 +11,10 @@ class CTVNewsIE(InfoExtractor): _VALID_URL = r'https?://(?:.+?\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P[0-9.]+)' _TESTS = [{ 'url': 'http://www.ctvnews.ca/video?clipId=901995', - 'md5': '10deb320dc0ccb8d01d34d12fc2ea672', + 'md5': '9b8624ba66351a23e0b6e1391971f9af', 'info_dict': { 'id': '901995', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Extended: \'That person cannot be me\' Johnson says', 'description': 'md5:958dd3b4f5bbbf0ed4d045c790d89285', 'timestamp': 1467286284, diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index d54e8df9f..2f485012f 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -718,10 +718,7 @@ NickRuIE, ) from .niconico import NiconicoIE, NiconicoPlaylistIE -from .ninecninemedia import ( - NineCNineMediaStackIE, - NineCNineMediaIE, -) +from .ninecninemedia import NineCNineMediaIE from .ninegag import NineGagIE from .ninenow import NineNowIE from .nintendo import NintendoIE diff --git a/youtube_dl/extractor/ninecninemedia.py b/youtube_dl/extractor/ninecninemedia.py index 8961309fd..875665d43 100644 --- a/youtube_dl/extractor/ninecninemedia.py +++ b/youtube_dl/extractor/ninecninemedia.py @@ -13,38 +13,11 @@ ) -class NineCNineMediaBaseIE(InfoExtractor): - _API_BASE_TEMPLATE = 'http://capi.9c9media.com/destinations/%s/platforms/desktop/contents/%s/' - - -class NineCNineMediaStackIE(NineCNineMediaBaseIE): - IE_NAME = '9c9media:stack' - _GEO_COUNTRIES = ['CA'] - _VALID_URL = r'9c9media:stack:(?P[^:]+):(?P\d+):(?P\d+):(?P\d+)' - - def _real_extract(self, url): - destination_code, content_id, package_id, stack_id = re.match(self._VALID_URL, url).groups() - stack_base_url_template = self._API_BASE_TEMPLATE + 'contentpackages/%s/stacks/%s/manifest.' - stack_base_url = stack_base_url_template % (destination_code, content_id, package_id, stack_id) - - formats = [] - formats.extend(self._extract_m3u8_formats( - stack_base_url + 'm3u8', stack_id, 'mp4', - 'm3u8_native', m3u8_id='hls', fatal=False)) - formats.extend(self._extract_f4m_formats( - stack_base_url + 'f4m', stack_id, - f4m_id='hds', fatal=False)) - self._sort_formats(formats) - - return { - 'id': stack_id, - 'formats': formats, - } - - -class NineCNineMediaIE(NineCNineMediaBaseIE): +class NineCNineMediaIE(InfoExtractor): IE_NAME = '9c9media' + _GEO_COUNTRIES = ['CA'] _VALID_URL = r'9c9media:(?P[^:]+):(?P\d+)' + _API_BASE_TEMPLATE = 'http://capi.9c9media.com/destinations/%s/platforms/desktop/contents/%s/' def _real_extract(self, url): destination_code, content_id = re.match(self._VALID_URL, url).groups() @@ -58,13 +31,26 @@ def _real_extract(self, url): content_package = content['ContentPackages'][0] package_id = content_package['Id'] content_package_url = api_base_url + 'contentpackages/%s/' % package_id - content_package = self._download_json(content_package_url, content_id) + content_package = self._download_json( + content_package_url, content_id, query={ + '$include': '[HasClosedCaptions]', + }) - if content_package.get('Constraints', {}).get('Security', {}).get('Type') == 'adobe-drm': + if content_package.get('Constraints', {}).get('Security', {}).get('Type'): raise ExtractorError('This video is DRM protected.', expected=True) - stacks = self._download_json(content_package_url + 'stacks/', package_id)['Items'] - multistacks = len(stacks) > 1 + manifest_base_url = content_package_url + 'manifest.' + formats = [] + formats.extend(self._extract_m3u8_formats( + manifest_base_url + 'm3u8', content_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) + formats.extend(self._extract_f4m_formats( + manifest_base_url + 'f4m', content_id, + f4m_id='hds', fatal=False)) + formats.extend(self._extract_mpd_formats( + manifest_base_url + 'mpd', content_id, + mpd_id='dash', fatal=False)) + self._sort_formats(formats) thumbnails = [] for image in content.get('Images', []): @@ -85,10 +71,12 @@ def _real_extract(self, url): continue container.append(e_name) - description = content.get('Desc') or content.get('ShortDesc') season = content.get('Season', {}) - base_info = { - 'description': description, + + info = { + 'id': content_id, + 'title': title, + 'description': content.get('Desc') or content.get('ShortDesc'), 'timestamp': parse_iso8601(content.get('BroadcastDateTime')), 'episode_number': int_or_none(content.get('Episode')), 'season': season.get('Name'), @@ -97,26 +85,19 @@ def _real_extract(self, url): 'series': content.get('Media', {}).get('Name'), 'tags': tags, 'categories': categories, + 'duration': float_or_none(content_package.get('Duration')), + 'formats': formats, } - entries = [] - for stack in stacks: - stack_id = compat_str(stack['Id']) - entry = { - '_type': 'url_transparent', - 'url': '9c9media:stack:%s:%s:%s:%s' % (destination_code, content_id, package_id, stack_id), - 'id': stack_id, - 'title': '%s_part%s' % (title, stack['Name']) if multistacks else title, - 'duration': float_or_none(stack.get('Duration')), - 'ie_key': 'NineCNineMediaStack', + if content_package.get('HasClosedCaptions'): + info['subtitles'] = { + 'en': [{ + 'url': manifest_base_url + 'vtt', + 'ext': 'vtt', + }, { + 'url': manifest_base_url + 'srt', + 'ext': 'srt', + }] } - entry.update(base_info) - entries.append(entry) - return { - '_type': 'multi_video', - 'id': content_id, - 'title': title, - 'description': description, - 'entries': entries, - } + return info diff --git a/youtube_dl/extractor/rds.py b/youtube_dl/extractor/rds.py index bf200ea4d..8c016a77d 100644 --- a/youtube_dl/extractor/rds.py +++ b/youtube_dl/extractor/rds.py @@ -19,7 +19,7 @@ class RDSIE(InfoExtractor): 'info_dict': { 'id': '604333', 'display_id': 'fowler-jr-prend-la-direction-de-jacksonville', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Fowler Jr. prend la direction de Jacksonville', 'description': 'Dante Fowler Jr. est le troisième choix du repêchage 2015 de la NFL. ', 'timestamp': 1430397346,