X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/540fd68c40df72763aee5d75598675c45cfa9aba..83c8b40add31fb501d587b64fb17a936ffd5dbc5:/youtube_dl/extractor/rai.py?ds=sidebyside diff --git a/youtube_dl/extractor/rai.py b/youtube_dl/extractor/rai.py index 1631faf..dc640b1 100644 --- a/youtube_dl/extractor/rai.py +++ b/youtube_dl/extractor/rai.py @@ -1,23 +1,118 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, -) +from ..compat import compat_urlparse from ..utils import ( + determine_ext, + ExtractorError, + find_xpath_attr, + fix_xml_ampersands, + int_or_none, parse_duration, unified_strdate, + update_url_query, + xpath_text, ) -class RaiIE(InfoExtractor): - _VALID_URL = r'(?P(?Phttp://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it))/dl/.+?-(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html)' +class RaiBaseIE(InfoExtractor): + def _extract_relinker_formats(self, relinker_url, video_id): + formats = [] + + for platform in ('mon', 'flash', 'native'): + relinker = self._download_xml( + relinker_url, video_id, + note='Downloading XML metadata for platform %s' % platform, + transform_source=fix_xml_ampersands, + query={'output': 45, 'pl': platform}, + headers=self.geo_verification_headers()) + + media_url = find_xpath_attr(relinker, './url', 'type', 'content').text + if media_url == 'http://download.rai.it/video_no_available.mp4': + self.raise_geo_restricted() + + ext = determine_ext(media_url) + if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'): + continue + + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + media_url, video_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + elif ext == 'f4m': + manifest_url = update_url_query( + media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'), + {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'}) + formats.extend(self._extract_f4m_formats( + manifest_url, video_id, f4m_id='hds', fatal=False)) + else: + bitrate = int_or_none(xpath_text(relinker, 'bitrate')) + formats.append({ + 'url': media_url, + 'tbr': bitrate if bitrate > 0 else None, + 'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http', + }) + + return formats + + def _extract_from_content_id(self, content_id, base_url): + media = self._download_json( + 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id, + content_id, 'Downloading video JSON') + + thumbnails = [] + for image_type in ('image', 'image_medium', 'image_300'): + thumbnail_url = media.get(image_type) + if thumbnail_url: + thumbnails.append({ + 'url': compat_urlparse.urljoin(base_url, thumbnail_url), + }) + + formats = [] + media_type = media['type'] + if 'Audio' in media_type: + formats.append({ + 'format_id': media.get('formatoAudio'), + 'url': media['audioUrl'], + 'ext': media.get('formatoAudio'), + }) + elif 'Video' in media_type: + formats.extend(self._extract_relinker_formats(media['mediaUri'], content_id)) + self._sort_formats(formats) + else: + raise ExtractorError('not a media file') + + subtitles = {} + captions = media.get('subtitlesUrl') + if captions: + STL_EXT = '.stl' + SRT_EXT = '.srt' + if captions.endswith(STL_EXT): + captions = captions[:-len(STL_EXT)] + SRT_EXT + subtitles['it'] = [{ + 'ext': 'srt', + 'url': captions, + }] + + return { + 'id': content_id, + 'title': media['name'], + 'description': media.get('desc'), + 'thumbnails': thumbnails, + 'uploader': media.get('author'), + 'upload_date': unified_strdate(media.get('date')), + 'duration': parse_duration(media.get('length')), + 'formats': formats, + 'subtitles': subtitles, + } + + +class RaiTVIE(RaiBaseIE): + _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/(?:[^/]+/)+(?:media|ondemand)/.+?-(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html' _TESTS = [ { 'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html', - 'md5': 'c064c0b2d09c278fb293116ef5d0a32d', + 'md5': '8970abf8caf8aef4696e7b1f2adfc696', 'info_dict': { 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391', 'ext': 'mp4', @@ -25,20 +120,22 @@ class RaiIE(InfoExtractor): 'description': 'md5:f27c544694cacb46a078db84ec35d2d9', 'upload_date': '20140407', 'duration': 6160, + 'thumbnail': 're:^https?://.*\.jpg$', } }, { + # no m3u8 stream 'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html', - 'md5': '8bb9c151924ce241b74dd52ef29ceafa', + # HDS download, MD5 is unstable 'info_dict': { 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'TG PRIMO TEMPO', - 'description': '', 'upload_date': '20140612', 'duration': 1758, + 'thumbnail': 're:^https?://.*\.jpg$', }, - 'skip': 'Error 404', + 'skip': 'Geo-restricted to Italy', }, { 'url': 'http://www.rainews.it/dl/rainews/media/state-of-the-net-Antonella-La-Carpia-regole-virali-7aafdea9-0e5d-49d5-88a6-7e65da67ae13.html', @@ -54,110 +151,117 @@ class RaiIE(InfoExtractor): }, { 'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-b4a49761-e0cc-4b14-8736-2729f6f73132-tg2.html', - 'md5': '35694f062977fe6619943f08ed935730', 'info_dict': { 'id': 'b4a49761-e0cc-4b14-8736-2729f6f73132', 'ext': 'mp4', 'title': 'Alluvione in Sardegna e dissesto idrogeologico', 'description': 'Edizione delle ore 20:30 ', - } + }, + 'skip': 'invalid urls', }, { 'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html', - 'md5': '02b64456f7cc09f96ff14e7dd489017e', + 'md5': 'e57493e1cb8bc7c564663f363b171847', 'info_dict': { 'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Il Candidato - Primo episodio: "Le Primarie"', - 'description': 'Primo appuntamento con "Il candidato" con Filippo Timi, alias Piero Zucca presidente!', - 'uploader': 'RaiTre', + 'description': 'md5:364b604f7db50594678f483353164fb8', + 'upload_date': '20140923', + 'duration': 386, + 'thumbnail': 're:^https?://.*\.jpg$', } - } + }, ] - def _extract_relinker_url(self, webpage): - return self._proto_relative_url(self._search_regex( - [r'name="videourl" content="([^"]+)"', r'var\s+videoURL(?:_MP4)?\s*=\s*"([^"]+)"'], - webpage, 'relinker url', default=None)) - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - host = mobj.group('host') + video_id = self._match_id(url) + + return self._extract_from_content_id(video_id, url) - webpage = self._download_webpage(url, video_id) - relinker_url = self._extract_relinker_url(webpage) +class RaiIE(RaiBaseIE): + _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/.+?-(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html' + _TESTS = [ + { + 'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html', + 'md5': '2dd727e61114e1ee9c47f0da6914e178', + 'info_dict': { + 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af', + 'ext': 'mp4', + 'title': 'Il pacco', + 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a', + 'upload_date': '20141221', + }, + }, + { + # Direct relinker URL + 'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews', + # HDS live stream, MD5 is unstable + 'info_dict': { + 'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc', + 'ext': 'flv', + 'title': 'EuroNews', + }, + 'skip': 'Geo-restricted to Italy', + }, + { + # Embedded content item ID + 'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined', + 'md5': '84c1135ce960e8822ae63cec34441d63', + 'info_dict': { + 'id': '0960e765-62c8-474a-ac4b-7eb3e2be39c8', + 'ext': 'mp4', + 'title': 'TG1 ore 20:00 del 02/07/2016', + 'upload_date': '20160702', + }, + }, + { + 'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html', + # HDS live stream, MD5 is unstable + 'info_dict': { + 'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9', + 'ext': 'flv', + 'title': 'La diretta di Rainews24', + }, + }, + ] - if not relinker_url: - iframe_path = self._search_regex( - r']+src="/?(dl/[^"]+\?iframe\b[^"]*)"', - webpage, 'iframe') - webpage = self._download_webpage( - '%s/%s' % (host, iframe_path), video_id) - relinker_url = self._extract_relinker_url(webpage) + @classmethod + def suitable(cls, url): + return False if RaiTVIE.suitable(url) else super(RaiIE, cls).suitable(url) - relinker = self._download_json( - '%s&output=47' % relinker_url, video_id) + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) - media_url = relinker['video'][0] - ct = relinker.get('ct') - if ct == 'f4m': - formats = self._extract_f4m_formats( - media_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id) - else: - formats = [{ - 'url': media_url, - 'format_id': ct, - }] + iframe_url = self._search_regex( + [r']+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"', + r'drawMediaRaiTV\(["\'](.+?)["\']'], + webpage, 'iframe', default=None) + if iframe_url: + if not iframe_url.startswith('http'): + iframe_url = compat_urlparse.urljoin(url, iframe_url) + return self.url_result(iframe_url) - json_link = self._html_search_meta( - 'jsonlink', webpage, 'JSON link', default=None) - if json_link: - media = self._download_json( - host + json_link, video_id, 'Downloading video JSON') - title = media.get('name') - description = media.get('desc') - thumbnail = media.get('image_300') or media.get('image_medium') or media.get('image') - duration = parse_duration(media.get('length')) - uploader = media.get('author') - upload_date = unified_strdate(media.get('date')) - else: - title = (self._search_regex( - r'var\s+videoTitolo\s*=\s*"(.+?)";', - webpage, 'title', default=None) or self._og_search_title(webpage)).replace('\\"', '"') - description = self._og_search_description(webpage) - thumbnail = self._og_search_thumbnail(webpage) - duration = None - uploader = self._html_search_meta('Editore', webpage, 'uploader') - upload_date = unified_strdate(self._html_search_meta( - 'item-date', webpage, 'upload date', default=None)) - - subtitles = self.extract_subtitles(video_id, webpage) + content_item_id = self._search_regex( + r'initEdizione\((?P[\'"])ContentItem-(?P[^\'"]+)(?P=q1)', + webpage, 'content item ID', group='content_id', default=None) + if content_item_id: + return self._extract_from_content_id(content_item_id, url) + + relinker_url = compat_urlparse.urljoin(url, self._search_regex( + r'(?:var\s+videoURL|mediaInfo\.mediaUri)\s*=\s*(?P[\'"])(?P(https?:)?//mediapolis\.rai\.it/relinker/relinkerServlet\.htm\?cont=\d+)(?P=q1)', + webpage, 'relinker URL', group='url')) + formats = self._extract_relinker_formats(relinker_url, video_id) + self._sort_formats(formats) + + title = self._search_regex( + r'var\s+videoTitolo\s*=\s*([\'"])(?P[^\'"]+)\1', + webpage, 'title', group='title', default=None) or self._og_search_title(webpage) return { 'id': video_id, 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'uploader': uploader, - 'upload_date': upload_date, - 'duration': duration, 'formats': formats, - 'subtitles': subtitles, } - - def _get_subtitles(self, video_id, webpage): - subtitles = {} - m = re.search(r'<meta name="closedcaption" content="(?P<captions>[^"]+)"', webpage) - if m: - captions = m.group('captions') - STL_EXT = '.stl' - SRT_EXT = '.srt' - if captions.endswith(STL_EXT): - captions = captions[:-len(STL_EXT)] + SRT_EXT - subtitles['it'] = [{ - 'ext': 'srt', - 'url': 'http://www.rai.tv%s' % compat_urllib_parse.quote(captions), - }] - return subtitles