X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Feagleplatform.py;h=36fef07b72be4db2183f2ed5fdee63981781a688;hb=HEAD;hp=0f8c73fd7d330da33afd3c2a3d2cb732e4d3ff33;hpb=d6712378e73951bede475569c887a1ac73f660a9;p=youtube-dl diff --git a/youtube_dl/extractor/eagleplatform.py b/youtube_dl/extractor/eagleplatform.py index 0f8c73fd7..36fef07b7 100644 --- a/youtube_dl/extractor/eagleplatform.py +++ b/youtube_dl/extractor/eagleplatform.py @@ -8,7 +8,8 @@ from ..compat import compat_HTTPError from ..utils import ( ExtractorError, int_or_none, - url_basename, + unsmuggle_url, + url_or_none, ) @@ -23,13 +24,13 @@ class EaglePlatformIE(InfoExtractor): _TESTS = [{ # http://lenta.ru/news/2015/03/06/navalny/ 'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201', - 'md5': '881ee8460e1b7735a8be938e2ffb362b', + # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used 'info_dict': { 'id': '227304', 'ext': 'mp4', 'title': 'Навальный вышел на свободу', 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 87, 'view_count': int, 'age_limit': 0, @@ -43,22 +44,72 @@ class EaglePlatformIE(InfoExtractor): 'id': '12820', 'ext': 'mp4', 'title': "'O Sole Mio", - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 216, 'view_count': int, }, 'skip': 'Georestricted', + }, { + # referrer protected video (https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/) + 'url': 'eagleplatform:tvrainru.media.eagleplatform.com:582306', + 'only_matching': True, }] + @staticmethod + def _extract_url(webpage): + # Regular iframe embedding + mobj = re.search( + r']+src=(["\'])(?P(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1', + webpage) + if mobj is not None: + return mobj.group('url') + PLAYER_JS_RE = r''' + ]+ + src=(?P["\'])(?:https?:)?//(?P(?:(?!(?P=qjs)).)+\.media\.eagleplatform\.com)/player/player\.js(?P=qjs) + .+? + ''' + # "Basic usage" embedding (see http://dultonmedia.github.io/eplayer/) + mobj = re.search( + r'''(?xs) + %s + ]+ + class=(?P["\'])eagleplayer(?P=qclass)[^>]+ + data-id=["\'](?P\d+) + ''' % PLAYER_JS_RE, webpage) + if mobj is not None: + return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict() + # Generalization of "Javascript code usage", "Combined usage" and + # "Usage without attaching to DOM" embeddings (see + # http://dultonmedia.github.io/eplayer/) + mobj = re.search( + r'''(?xs) + %s + + ''' % PLAYER_JS_RE, webpage) + if mobj is not None: + return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict() + @staticmethod def _handle_error(response): status = int_or_none(response.get('status', 200)) if status != 200: raise ExtractorError(' '.join(response['errors']), expected=True) - def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'): + def _download_json(self, url_or_request, video_id, *args, **kwargs): try: - response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note) + response = super(EaglePlatformIE, self)._download_json( + url_or_request, video_id, *args, **kwargs) except ExtractorError as ee: if isinstance(ee.cause, compat_HTTPError): response = self._parse_json(ee.cause.read().decode('utf-8'), video_id) @@ -70,11 +121,24 @@ class EaglePlatformIE(InfoExtractor): return self._download_json(url_or_request, video_id, note)['data'][0] def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + mobj = re.match(self._VALID_URL, url) host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id') + headers = {} + query = { + 'id': video_id, + } + + referrer = smuggled_data.get('referrer') + if referrer: + headers['Referer'] = referrer + query['referrer'] = referrer + player_data = self._download_json( - 'http://%s/api/player_data?id=%s' % (host, video_id), video_id) + 'http://%s/api/player_data' % host, video_id, + headers=headers, query=query) media = player_data['data']['playlist']['viewports'][0]['medialist'][0] @@ -95,26 +159,38 @@ class EaglePlatformIE(InfoExtractor): m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON') m3u8_formats = self._extract_m3u8_formats( - m3u8_url, video_id, - 'mp4', entry_protocol='m3u8_native', m3u8_id='hls') + m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False) formats.extend(m3u8_formats) - mp4_url = self._get_video_url( + m3u8_formats_dict = {} + for f in m3u8_formats: + if f.get('height') is not None: + m3u8_formats_dict[f['height']] = f + + mp4_data = self._download_json( # Secure mp4 URL is constructed according to Player.prototype.mp4 from # http://lentaru.media.eagleplatform.com/player/player.js - re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4', secure_m3u8), - video_id, 'Downloading mp4 JSON') - mp4_url_basename = url_basename(mp4_url) - for m3u8_format in m3u8_formats: - mobj = re.search('/([^/]+)/index\.m3u8', m3u8_format['url']) - if mobj: - http_format = m3u8_format.copy() - http_format.update({ - 'url': mp4_url.replace(mp4_url_basename, mobj.group(1)), - 'format_id': m3u8_format['format_id'].replace('hls', 'http'), - 'protocol': 'http', - }) - formats.append(http_format) + re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4s', secure_m3u8), + video_id, 'Downloading mp4 JSON', fatal=False) + if mp4_data: + for format_id, format_url in mp4_data.get('data', {}).items(): + if not url_or_none(format_url): + continue + height = int_or_none(format_id) + if height is not None and m3u8_formats_dict.get(height): + f = m3u8_formats_dict[height].copy() + f.update({ + 'format_id': f['format_id'].replace('hls', 'http'), + 'protocol': 'http', + }) + else: + f = { + 'format_id': 'http-%s' % format_id, + 'height': int_or_none(format_id), + } + f['url'] = format_url + formats.append(f) self._sort_formats(formats)