X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fnpo.py;h=557dffa46846ff8a4c94f6bb102a186fa1ce5eb8;hb=674fb0fcc54c72448f80a0573f7fd116f220827e;hp=6d20f755a89a244e41feb1da5e2416feeb351d7e;hpb=5c4a81d9345f0520a9d40acbb8212fb94bcc9e24;p=youtube-dl diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py index 6d20f755a..557dffa46 100644 --- a/youtube_dl/extractor/npo.py +++ b/youtube_dl/extractor/npo.py @@ -1,7 +1,5 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..utils import ( fix_xml_ampersands, @@ -13,9 +11,18 @@ from ..utils import ( ) -class NPOIE(InfoExtractor): +class NPOBaseIE(InfoExtractor): + def _get_token(self, video_id): + token_page = self._download_webpage( + 'http://ida.omroep.nl/npoplayer/i.js', + video_id, note='Downloading token') + return self._search_regex( + r'npoplayer\.token = "(.+?)"', token_page, 'token') + + +class NPOIE(NPOBaseIE): IE_NAME = 'npo.nl' - _VALID_URL = r'https?://www\.npo\.nl/[^/]+/[^/]+/(?P[^/?]+)' + _VALID_URL = r'https?://(?:www\.)?npo\.nl/(?!live|radio)[^/]+/[^/]+/(?P[^/?]+)' _TESTS = [ { @@ -80,8 +87,7 @@ class NPOIE(InfoExtractor): ] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) return self._get_info(video_id) def _get_info(self, video_id): @@ -91,12 +97,8 @@ class NPOIE(InfoExtractor): # We have to remove the javascript callback transform_source=strip_jsonp, ) - token_page = self._download_webpage( - 'http://ida.omroep.nl/npoplayer/i.js', - video_id, - note='Downloading token' - ) - token = self._search_regex(r'npoplayer\.token = "(.+?)"', token_page, 'token') + + token = self._get_token(video_id) formats = [] @@ -159,6 +161,13 @@ class NPOIE(InfoExtractor): self._sort_formats(formats) + subtitles = {} + if metadata.get('tt888') == 'ja': + subtitles['nl'] = [{ + 'ext': 'vtt', + 'url': 'http://e.omroep.nl/tt888/%s' % video_id, + }] + return { 'id': video_id, 'title': metadata['titel'], @@ -167,6 +176,164 @@ class NPOIE(InfoExtractor): 'upload_date': unified_strdate(metadata.get('gidsdatum')), 'duration': parse_duration(metadata.get('tijdsduur')), 'formats': formats, + 'subtitles': subtitles, + } + + +class NPOLiveIE(NPOBaseIE): + IE_NAME = 'npo.nl:live' + _VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P.+)' + + _TEST = { + 'url': 'http://www.npo.nl/live/npo-1', + 'info_dict': { + 'id': 'LI_NEDERLAND1_136692', + 'display_id': 'npo-1', + 'ext': 'mp4', + 'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'description': 'Livestream', + 'is_live': True, + }, + 'params': { + 'skip_download': True, + } + } + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + live_id = self._search_regex( + r'data-prid="([^"]+)"', webpage, 'live id') + + metadata = self._download_json( + 'http://e.omroep.nl/metadata/%s' % live_id, + display_id, transform_source=strip_jsonp) + + token = self._get_token(display_id) + + formats = [] + + streams = metadata.get('streams') + if streams: + for stream in streams: + stream_type = stream.get('type').lower() + # smooth streaming is not supported + if stream_type in ['ss', 'ms']: + continue + stream_info = self._download_json( + 'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp' + % (stream.get('url'), token), + display_id, 'Downloading %s JSON' % stream_type) + if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0): + continue + stream_url = self._download_json( + stream_info['stream'], display_id, + 'Downloading %s URL' % stream_type, + transform_source=strip_jsonp) + if stream_type == 'hds': + f4m_formats = self._extract_f4m_formats(stream_url, display_id) + # f4m downloader downloads only piece of live stream + for f4m_format in f4m_formats: + f4m_format['preference'] = -1 + formats.extend(f4m_formats) + elif stream_type == 'hls': + formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4')) + else: + formats.append({ + 'url': stream_url, + 'preference': -10, + }) + + self._sort_formats(formats) + + return { + 'id': live_id, + 'display_id': display_id, + 'title': self._live_title(metadata['titel']), + 'description': metadata['info'], + 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'], + 'formats': formats, + 'is_live': True, + } + + +class NPORadioIE(InfoExtractor): + IE_NAME = 'npo.nl:radio' + _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P[^/]+)/?$' + + _TEST = { + 'url': 'http://www.npo.nl/radio/radio-1', + 'info_dict': { + 'id': 'radio-1', + 'ext': 'mp3', + 'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'is_live': True, + }, + 'params': { + 'skip_download': True, + } + } + + @staticmethod + def _html_get_attribute_regex(attribute): + return r'{0}\s*=\s*\'([^\']+)\''.format(attribute) + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + title = self._html_search_regex( + self._html_get_attribute_regex('data-channel'), webpage, 'title') + + stream = self._parse_json( + self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'), + video_id) + + codec = stream.get('codec') + + return { + 'id': video_id, + 'url': stream['url'], + 'title': self._live_title(title), + 'acodec': codec, + 'ext': codec, + 'is_live': True, + } + + +class NPORadioFragmentIE(InfoExtractor): + IE_NAME = 'npo.nl:radio:fragment' + _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P\d+)' + + _TEST = { + 'url': 'http://www.npo.nl/radio/radio-5/fragment/174356', + 'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2', + 'info_dict': { + 'id': '174356', + 'ext': 'mp3', + 'title': 'Jubileumconcert Willeke Alberti', + }, + } + + def _real_extract(self, url): + audio_id = self._match_id(url) + + webpage = self._download_webpage(url, audio_id) + + title = self._html_search_regex( + r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id, + webpage, 'title') + + audio_url = self._search_regex( + r"data-streams='([^']+)'", webpage, 'audio url') + + return { + 'id': audio_id, + 'url': audio_url, + 'title': title, }