From c6d3f81a4077aaf9cffc6aa2d0dec92f38e74bb0 Mon Sep 17 00:00:00 2001 From: nixxo Date: Mon, 29 May 2023 06:20:03 +0200 Subject: [PATCH] [extractor/rai] Rewrite extractors (#5940) Authored by: nixxo, danog Closes #5672, closes #6341 Co-authored-by: Daniil Gentili --- yt_dlp/extractor/_extractors.py | 3 +- yt_dlp/extractor/rai.py | 575 +++++++++++++++----------------- 2 files changed, 271 insertions(+), 307 deletions(-) diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py index bf041ae61..d9028a831 100644 --- a/yt_dlp/extractor/_extractors.py +++ b/yt_dlp/extractor/_extractors.py @@ -1548,6 +1548,8 @@ RadLiveSeasonIE, ) from .rai import ( + RaiIE, + RaiCulturaIE, RaiPlayIE, RaiPlayLiveIE, RaiPlayPlaylistIE, @@ -1556,7 +1558,6 @@ RaiPlaySoundPlaylistIE, RaiNewsIE, RaiSudtirolIE, - RaiIE, ) from .raywenderlich import ( RayWenderlichIE, diff --git a/yt_dlp/extractor/rai.py b/yt_dlp/extractor/rai.py index cab12cc21..df4102a40 100644 --- a/yt_dlp/extractor/rai.py +++ b/yt_dlp/extractor/rai.py @@ -1,19 +1,12 @@ import re from .common import InfoExtractor -from ..compat import ( - compat_str, - compat_urlparse, -) from ..utils import ( clean_html, determine_ext, ExtractorError, filter_dict, - find_xpath_attr, - fix_xml_ampersands, GeoRestrictedError, - HEADRequest, int_or_none, join_nonempty, parse_duration, @@ -35,82 +28,70 @@ class RaiBaseIE(InfoExtractor): _GEO_BYPASS = False def _extract_relinker_info(self, relinker_url, video_id, audio_only=False): + def fix_cdata(s): + # remove \r\n\t before and after to avoid + # polluted text with xpath_text + s = re.sub(r'(\]\]>)[\r\n\t]+()[\r\n\t]+( 0 else None, + 'format_id': join_nonempty('https', bitrate, delim='-'), + }) + else: + raise ExtractorError('Unrecognized media file found') - if xpath_text(relinker, './license_url', default='{}') != '{}': - self.report_drm(video_id) - - if not geoprotection: - geoprotection = xpath_text( - relinker, './geoprotection', default=None) == 'Y' - - if not is_live: - is_live = xpath_text( - relinker, './is_live', default=None) == 'Y' - if not duration: - duration = parse_duration(xpath_text( - relinker, './duration', default=None)) - - url_elem = find_xpath_attr(relinker, './url', 'type', 'content') - if url_elem is None: - continue - - media_url = url_elem.text - - # This does not imply geo restriction (e.g. - # http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html) - if '/video_no_available.mp4' in media_url: - continue - - ext = determine_ext(media_url) - if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'): - continue - - if ext == 'mp3': - formats.append({ - 'url': media_url, - 'vcodec': 'none', - 'acodec': 'mp3', - 'format_id': 'http-mp3', - }) - break - elif ext == 'm3u8' or 'format=m3u8' in media_url or platform == 'mon': - formats.extend(self._extract_m3u8_formats( - media_url, video_id, 'mp4', 'm3u8_native', - m3u8_id='hls', fatal=False)) - elif ext == 'f4m' or platform == 'flash': - 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': f'http-{bitrate if bitrate > 0 else "http"}', - }) - - if not formats and geoprotection is True: + if (not formats and geoprotection is True) or '/video_no_available.mp4' in media_url: self.raise_geo_restricted(countries=self._GEO_COUNTRIES, metadata_available=True) - if not audio_only: - formats.extend(self._create_http_urls(relinker_url, formats)) + if not audio_only and not is_live: + formats.extend(self._create_http_urls(media_url, relinker_url, formats)) return filter_dict({ 'is_live': is_live, @@ -118,38 +99,31 @@ def _extract_relinker_info(self, relinker_url, video_id, audio_only=False): 'formats': formats, }) - def _create_http_urls(self, relinker_url, fmts): - _RELINKER_REG = r'https?://(?P[^/]+?)/(?:i/)?(?P[^/]+?)/(?P.+?)/(?P\w+)(?:_(?P[\d\,]+))?(?:\.mp4|/playlist\.m3u8).+?' + def _create_http_urls(self, manifest_url, relinker_url, fmts): + _MANIFEST_REG = r'/(?P\w+)(?:_(?P[\d\,]+))?(?:\.mp4)?(?:\.csmil)?/playlist\.m3u8' _MP4_TMPL = '%s&overrideUserAgentRule=mp4-%s' _QUALITY = { # tbr: w, h - '250': [352, 198], - '400': [512, 288], - '700': [512, 288], - '800': [700, 394], - '1200': [736, 414], - '1800': [1024, 576], - '2400': [1280, 720], - '3200': [1440, 810], - '3600': [1440, 810], - '5000': [1920, 1080], - '10000': [1920, 1080], + 250: [352, 198], + 400: [512, 288], + 600: [512, 288], + 700: [512, 288], + 800: [700, 394], + 1200: [736, 414], + 1500: [920, 518], + 1800: [1024, 576], + 2400: [1280, 720], + 3200: [1440, 810], + 3600: [1440, 810], + 5000: [1920, 1080], + 10000: [1920, 1080], } - def test_url(url): - resp = self._request_webpage( - HEADRequest(url), None, headers={'User-Agent': 'Rai'}, - fatal=False, errnote=False, note=False) - - if resp is False: + def percentage(number, target, pc=20, roof=125): + '''check if the target is in the range of number +/- percent''' + if not number or number < 0: return False - - if resp.code == 200: - return False if resp.url == url else resp.url - return None - - # filter out audio-only formats - fmts = [f for f in fmts if not f.get('vcodec') == 'none'] + return abs(target - number) < min(float(number) * float(pc) / 100.0, roof) def get_format_info(tbr): import math @@ -157,67 +131,78 @@ def get_format_info(tbr): if len(fmts) == 1 and not br: br = fmts[0].get('tbr') if br and br > 300: - tbr = compat_str(math.floor(br / 100) * 100) + tbr = math.floor(br / 100) * 100 else: - tbr = '250' + tbr = 250 # try extracting info from available m3u8 formats - format_copy = None + format_copy = [None, None] for f in fmts: if f.get('tbr'): - br_limit = math.floor(br / 100) - if br_limit - 1 <= math.floor(f['tbr'] / 100) <= br_limit + 1: - format_copy = f.copy() + if percentage(tbr, f['tbr']): + format_copy[0] = f.copy() + if [f.get('width'), f.get('height')] == _QUALITY.get(tbr): + format_copy[1] = f.copy() + format_copy[1]['tbr'] = tbr + + # prefer format with similar bitrate because there might be + # multiple video with the same resolution but different bitrate + format_copy = format_copy[0] or format_copy[1] or {} return { + 'format_id': f'https-{tbr}', 'width': format_copy.get('width'), 'height': format_copy.get('height'), 'tbr': format_copy.get('tbr'), 'vcodec': format_copy.get('vcodec'), 'acodec': format_copy.get('acodec'), 'fps': format_copy.get('fps'), - 'format_id': f'https-{tbr}', } if format_copy else { + 'format_id': f'https-{tbr}', 'width': _QUALITY[tbr][0], 'height': _QUALITY[tbr][1], - 'format_id': f'https-{tbr}', - 'tbr': int(tbr), + 'tbr': tbr, + 'vcodec': 'avc1', + 'acodec': 'mp4a', + 'fps': 25, } - loc = test_url(_MP4_TMPL % (relinker_url, '*')) - if not isinstance(loc, compat_str): - return [] + # filter out single-stream formats + fmts = [f for f in fmts + if not f.get('vcodec') == 'none' and not f.get('acodec') == 'none'] - mobj = re.match( - _RELINKER_REG, - test_url(relinker_url) or '') + mobj = re.search(_MANIFEST_REG, manifest_url) if not mobj: return [] - available_qualities = mobj.group('quality').split(',') if mobj.group('quality') else ['*'] - available_qualities = [i for i in available_qualities if i] formats = [] - for q in available_qualities: - fmt = { + for q in filter(None, available_qualities): + self.write_debug(f'Creating https format for quality {q}') + formats.append({ 'url': _MP4_TMPL % (relinker_url, q), 'protocol': 'https', 'ext': 'mp4', **get_format_info(q) - } - formats.append(fmt) + }) return formats + @staticmethod + def _get_thumbnails_list(thumbs, url): + return [{ + 'url': urljoin(url, thumb_url), + } for thumb_url in (thumbs or {}).values() if thumb_url] + @staticmethod def _extract_subtitles(url, video_data): STL_EXT = 'stl' SRT_EXT = 'srt' subtitles = {} - subtitles_array = video_data.get('subtitlesArray') or [] + subtitles_array = video_data.get('subtitlesArray') or video_data.get('subtitleList') or [] for k in ('subtitles', 'subtitlesUrl'): subtitles_array.append({'url': video_data.get(k)}) for subtitle in subtitles_array: sub_url = subtitle.get('url') - if sub_url and isinstance(sub_url, compat_str): + if sub_url and isinstance(sub_url, str): sub_lang = subtitle.get('language') or 'it' sub_url = urljoin(url, sub_url) sub_ext = determine_ext(sub_url, SRT_EXT) @@ -236,7 +221,7 @@ def _extract_subtitles(url, video_data): class RaiPlayIE(RaiBaseIE): _VALID_URL = rf'(?Phttps?://(?:www\.)?raiplay\.it/.+?-(?P{RaiBaseIE._UUID_RE}))\.(?:html|json)' _TESTS = [{ - 'url': 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html', + 'url': 'https://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html', 'md5': '8970abf8caf8aef4696e7b1f2adfc696', 'info_dict': { 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391', @@ -244,22 +229,20 @@ class RaiPlayIE(RaiBaseIE): 'title': 'Report del 07/04/2014', 'alt_title': 'St 2013/14 - Report - Espresso nel caffè - 07/04/2014', 'description': 'md5:d730c168a58f4bb35600fc2f881ec04e', - 'thumbnail': r're:^https?://.*\.jpg$', - 'uploader': 'Rai Gulp', + 'thumbnail': r're:^https?://www\.raiplay\.it/.+\.jpg', + 'uploader': 'Rai 3', + 'creator': 'Rai 3', 'duration': 6160, 'series': 'Report', 'season': '2013/14', - 'subtitles': { - 'it': 'count:4', - }, + 'subtitles': {'it': 'count:4'}, 'release_year': 2022, 'episode': 'Espresso nel caffè - 07/04/2014', 'timestamp': 1396919880, 'upload_date': '20140408', + 'formats': 'count:4', }, - 'params': { - 'skip_download': True, - }, + 'params': {'skip_download': True}, }, { # 1080p direct mp4 url 'url': 'https://www.raiplay.it/video/2021/11/Blanca-S1E1-Senza-occhi-b1255a4a-8e72-4a2f-b9f3-fc1308e00736.html', @@ -270,8 +253,9 @@ class RaiPlayIE(RaiBaseIE): 'title': 'Blanca - S1E1 - Senza occhi', 'alt_title': 'St 1 Ep 1 - Blanca - Senza occhi', 'description': 'md5:75f95d5c030ec8bac263b1212322e28c', - 'thumbnail': r're:^https?://.*\.jpg$', - 'uploader': 'Rai 1', + 'thumbnail': r're:^https://www\.raiplay\.it/dl/img/.+\.jpg', + 'uploader': 'Rai Premium', + 'creator': 'Rai Fiction', 'duration': 6493, 'series': 'Blanca', 'season': 'Season 1', @@ -281,6 +265,30 @@ class RaiPlayIE(RaiBaseIE): 'episode': 'Senza occhi', 'timestamp': 1637318940, 'upload_date': '20211119', + 'formats': 'count:12', + }, + 'params': {'skip_download': True}, + 'expected_warnings': ['Video not available. Likely due to geo-restriction.'] + }, { + # 1500 quality + 'url': 'https://www.raiplay.it/video/2012/09/S1E11---Tutto-cio-che-luccica-0cab3323-732e-45d6-8e86-7704acab6598.html', + 'md5': 'a634d20e8ab2d43724c273563f6bf87a', + 'info_dict': { + 'id': '0cab3323-732e-45d6-8e86-7704acab6598', + 'ext': 'mp4', + 'title': 'Mia and Me - S1E11 - Tutto ciò che luccica', + 'alt_title': 'St 1 Ep 11 - Mia and Me - Tutto ciò che luccica', + 'description': 'md5:4969e594184b1920c4c1f2b704da9dea', + 'thumbnail': r're:^https?://.*\.jpg$', + 'uploader': 'Rai Gulp', + 'series': 'Mia and Me', + 'season': 'Season 1', + 'episode_number': 11, + 'release_year': 2015, + 'season_number': 1, + 'episode': 'Tutto ciò che luccica', + 'timestamp': 1348495020, + 'upload_date': '20120924', }, }, { 'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?', @@ -299,57 +307,40 @@ def _real_extract(self, url): base, video_id = self._match_valid_url(url).groups() media = self._download_json( - base + '.json', video_id, 'Downloading video JSON') + f'{base}.json', video_id, 'Downloading video JSON') if not self.get_param('allow_unplayable_formats'): - if try_get( - media, - (lambda x: x['rights_management']['rights']['drm'], - lambda x: x['program_info']['rights_management']['rights']['drm']), - dict): + if traverse_obj(media, (('program_info', None), 'rights_management', 'rights', 'drm')): self.report_drm(video_id) - title = media['name'] video = media['video'] - relinker_info = self._extract_relinker_info(video['content_url'], video_id) - - thumbnails = [] - for _, value in media.get('images', {}).items(): - if value: - thumbnails.append({ - 'url': urljoin(url, value), - }) - - date_published = media.get('date_published') - time_published = media.get('time_published') - if date_published and time_published: - date_published += ' ' + time_published - - subtitles = self._extract_subtitles(url, video) - - program_info = media.get('program_info') or {} + date_published = join_nonempty( + media.get('date_published'), media.get('time_published'), delim=' ') season = media.get('season') - alt_title = join_nonempty(media.get('subtitle'), media.get('toptitle'), delim=' - ') return { 'id': remove_start(media.get('id'), 'ContentItem-') or video_id, 'display_id': video_id, - 'title': title, + 'title': media.get('name'), 'alt_title': strip_or_none(alt_title or None), 'description': media.get('description'), - 'uploader': strip_or_none(media.get('channel') or None), - 'creator': strip_or_none(media.get('editor') or None), + 'uploader': strip_or_none( + traverse_obj(media, ('program_info', 'channel')) + or media.get('channel') or None), + 'creator': strip_or_none( + traverse_obj(media, ('program_info', 'editor')) + or media.get('editor') or None), 'duration': parse_duration(video.get('duration')), 'timestamp': unified_timestamp(date_published), - 'thumbnails': thumbnails, - 'series': program_info.get('name'), + 'thumbnails': self._get_thumbnails_list(media.get('images'), url), + 'series': traverse_obj(media, ('program_info', 'name')), 'season_number': int_or_none(season), 'season': season if (season and not season.isdigit()) else None, 'episode': media.get('episode_title'), 'episode_number': int_or_none(media.get('episode')), - 'subtitles': subtitles, + 'subtitles': self._extract_subtitles(url, video), 'release_year': int_or_none(traverse_obj(media, ('track_info', 'edit_year'))), **relinker_info } @@ -371,38 +362,39 @@ class RaiPlayLiveIE(RaiPlayIE): # XXX: Do not subclass from concrete IE 'live_status': 'is_live', 'upload_date': '20090502', 'timestamp': 1241276220, + 'formats': 'count:3', }, - 'params': { - 'skip_download': True, - }, + 'params': {'skip_download': True}, }] class RaiPlayPlaylistIE(InfoExtractor): _VALID_URL = r'(?Phttps?://(?:www\.)?raiplay\.it/programmi/(?P[^/?#&]+))(?:/(?P[^?#&]+))?' _TESTS = [{ + # entire series episodes + extras... 'url': 'https://www.raiplay.it/programmi/nondirloalmiocapo/', 'info_dict': { 'id': 'nondirloalmiocapo', 'title': 'Non dirlo al mio capo', 'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b', }, - 'playlist_mincount': 12, + 'playlist_mincount': 30, }, { + # single season 'url': 'https://www.raiplay.it/programmi/nondirloalmiocapo/episodi/stagione-2/', 'info_dict': { 'id': 'nondirloalmiocapo', 'title': 'Non dirlo al mio capo - Stagione 2', 'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b', }, - 'playlist_mincount': 12, + 'playlist_count': 12, }] def _real_extract(self, url): base, playlist_id, extra_id = self._match_valid_url(url).groups() program = self._download_json( - base + '.json', playlist_id, 'Downloading program JSON') + f'{base}.json', playlist_id, 'Downloading program JSON') if extra_id: extra_id = extra_id.upper().rstrip('/') @@ -450,7 +442,7 @@ class RaiPlaySoundIE(RaiBaseIE): 'title': 'Il Ruggito del Coniglio del 10/12/2021', 'alt_title': 'md5:0e6476cd57858bb0f3fcc835d305b455', 'description': 'md5:2a17d2107e59a4a8faa0e18334139ee2', - 'thumbnail': r're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.+\.jpg$', 'uploader': 'rai radio 2', 'duration': 5685, 'series': 'Il Ruggito del Coniglio', @@ -459,9 +451,7 @@ class RaiPlaySoundIE(RaiBaseIE): 'timestamp': 1638346620, 'upload_date': '20211201', }, - 'params': { - 'skip_download': True, - }, + 'params': {'skip_download': True}, }] def _real_extract(self, url): @@ -480,9 +470,6 @@ def _real_extract(self, url): lambda x: x['live']['create_date'])) podcast_info = traverse_obj(media, 'podcast_info', ('live', 'cards', 0)) or {} - thumbnails = [{ - 'url': urljoin(url, thumb_url), - } for thumb_url in (podcast_info.get('images') or {}).values() if thumb_url] return { **info, @@ -494,7 +481,7 @@ def _real_extract(self, url): 'uploader': traverse_obj(media, ('track_info', 'channel'), expected_type=strip_or_none), 'creator': traverse_obj(media, ('track_info', 'editor'), expected_type=strip_or_none), 'timestamp': unified_timestamp(date_published), - 'thumbnails': thumbnails, + 'thumbnails': self._get_thumbnails_list(podcast_info.get('images'), url), 'series': podcast_info.get('title'), 'season_number': int_or_none(media.get('season')), 'episode': media.get('episode_title'), @@ -512,30 +499,30 @@ class RaiPlaySoundLiveIE(RaiPlaySoundIE): # XXX: Do not subclass from concrete 'display_id': 'radio2', 'ext': 'mp4', 'title': r're:Rai Radio 2 \d+-\d+-\d+ \d+:\d+', - 'thumbnail': r're:https://www.raiplaysound.it/dl/img/.+?png', + 'thumbnail': r're:^https://www\.raiplaysound\.it/dl/img/.+\.png', 'uploader': 'rai radio 2', 'series': 'Rai Radio 2', 'creator': 'raiplaysound', 'is_live': True, 'live_status': 'is_live', }, - 'params': { - 'skip_download': 'live', - }, + 'params': {'skip_download': True}, }] class RaiPlaySoundPlaylistIE(InfoExtractor): _VALID_URL = r'(?Phttps?://(?:www\.)?raiplaysound\.it/(?:programmi|playlist|audiolibri)/(?P[^/?#&]+))(?:/(?P[^?#&]+))?' _TESTS = [{ + # entire show 'url': 'https://www.raiplaysound.it/programmi/ilruggitodelconiglio', 'info_dict': { 'id': 'ilruggitodelconiglio', 'title': 'Il Ruggito del Coniglio', - 'description': 'md5:1bbaf631245a7ab1ec4d9fbb3c7aa8f3', + 'description': 'md5:48cff6972435964284614d70474132e6', }, 'playlist_mincount': 65, }, { + # single season 'url': 'https://www.raiplaysound.it/programmi/ilruggitodelconiglio/puntate/prima-stagione-1995', 'info_dict': { 'id': 'ilruggitodelconiglio_puntate_prima-stagione-1995', @@ -568,22 +555,19 @@ def _real_extract(self, url): class RaiIE(RaiBaseIE): _VALID_URL = rf'https?://[^/]+\.(?:rai\.(?:it|tv))/.+?-(?P{RaiBaseIE._UUID_RE})(?:-.+?)?\.html' _TESTS = [{ - # var uniquename = "ContentItem-..." - # data-id="ContentItem-..." 'url': 'https://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html', 'info_dict': { 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9', 'ext': 'mp4', 'title': 'TG PRIMO TEMPO', - 'thumbnail': r're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg', 'duration': 1758, 'upload_date': '20140612', }, - 'skip': 'This content is available only in Italy', + 'params': {'skip_download': True}, + 'expected_warnings': ['Video not available. Likely due to geo-restriction.'] }, { - # with ContentItem in og:url 'url': 'https://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html', - 'md5': '06345bd97c932f19ffb129973d07a020', 'info_dict': { 'id': 'efb17665-691c-45d5-a60c-5301333cbb0c', 'ext': 'mp4', @@ -592,123 +576,51 @@ class RaiIE(RaiBaseIE): 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 2214, 'upload_date': '20161103' - } + }, + 'params': {'skip_download': True}, }, { - # Direct MMS URL + # Direct MMS: Media URL no longer works. 'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-b63a4089-ac28-48cf-bca5-9f5b5bc46df5.html', 'only_matching': True, }] - def _extract_from_content_id(self, content_id, url): + def _real_extract(self, url): + content_id = self._match_id(url) media = self._download_json( f'https://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-{content_id}.html?json', - content_id, 'Downloading video JSON') + content_id, 'Downloading video JSON', fatal=False, expected_status=404) - title = media['name'].strip() + if media is None: + return None - media_type = media['type'] - if 'Audio' in media_type: + if 'Audio' in media['type']: relinker_info = { 'formats': [{ - 'format_id': media.get('formatoAudio'), + 'format_id': join_nonempty('https', media.get('formatoAudio'), delim='-'), 'url': media['audioUrl'], 'ext': media.get('formatoAudio'), + 'vcodec': 'none', + 'acodec': media.get('formatoAudio'), }] } - elif 'Video' in media_type: + elif 'Video' in media['type']: relinker_info = self._extract_relinker_info(media['mediaUri'], content_id) else: raise ExtractorError('not a media file') - 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(url, thumbnail_url), - }) - - subtitles = self._extract_subtitles(url, media) + thumbnails = self._get_thumbnails_list( + {image_type: media.get(image_type) for image_type in ( + 'image', 'image_medium', 'image_300')}, url) return { 'id': content_id, - 'title': title, - 'description': strip_or_none(media.get('desc') or None), + 'title': strip_or_none(media.get('name') or media.get('title')), + 'description': strip_or_none(media.get('desc')) or None, 'thumbnails': thumbnails, - 'uploader': strip_or_none(media.get('author') or None), + 'uploader': strip_or_none(media.get('author')) or None, 'upload_date': unified_strdate(media.get('date')), 'duration': parse_duration(media.get('length')), - 'subtitles': subtitles, - **relinker_info - } - - def _real_extract(self, url): - video_id = self._match_id(url) - - webpage = self._download_webpage(url, video_id) - - content_item_id = None - - content_item_url = self._html_search_meta( - ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url', - 'twitter:player', 'jsonlink'), webpage, default=None) - if content_item_url: - content_item_id = self._search_regex( - rf'ContentItem-({self._UUID_RE})', content_item_url, - 'content item id', default=None) - - if not content_item_id: - content_item_id = self._search_regex( - rf'''(?x) - (?: - (?:initEdizione|drawMediaRaiTV)\(| - <(?:[^>]+\bdata-id|var\s+uniquename)=| - ]+\bsrc= - ) - (["\']) - (?:(?!\1).)*\bContentItem-(?P{self._UUID_RE}) - ''', - webpage, 'content item id', default=None, group='id') - - content_item_ids = set() - if content_item_id: - content_item_ids.add(content_item_id) - if video_id not in content_item_ids: - content_item_ids.add(video_id) - - for content_item_id in content_item_ids: - try: - return self._extract_from_content_id(content_item_id, url) - except GeoRestrictedError: - raise - except ExtractorError: - pass - - relinker_url = self._proto_relative_url(self._search_regex( - r'''(?x) - (?: - var\s+videoURL| - mediaInfo\.mediaUri - )\s*=\s* - ([\'"]) - (?P - (?:https?:)? - //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\? - (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1 - ''', - webpage, 'relinker URL', group='url')) - - relinker_info = self._extract_relinker_info( - urljoin(url, relinker_url), video_id) - - 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, + 'subtitles': self._extract_subtitles(url, media), **relinker_info } @@ -726,7 +638,8 @@ class RaiNewsIE(RaiIE): # XXX: Do not subclass from concrete IE 'duration': 1589, 'upload_date': '20220529', 'uploader': 'rainews', - } + }, + 'params': {'skip_download': True}, }, { # old content with fallback method to extract media urls 'url': 'https://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html', @@ -739,12 +652,14 @@ class RaiNewsIE(RaiIE): # XXX: Do not subclass from concrete IE 'duration': 833, 'upload_date': '20161103' }, + 'params': {'skip_download': True}, 'expected_warnings': ['unable to extract player_data'], }, { # iframe + drm 'url': 'https://www.rainews.it/iframe/video/2022/07/euro2022-europei-calcio-femminile-italia-belgio-gol-0-1-video-4de06a69-de75-4e32-a657-02f0885f8118.html', 'only_matching': True, }] + _PLAYER_TAG = 'news' def _real_extract(self, url): video_id = self._match_id(url) @@ -752,8 +667,8 @@ def _real_extract(self, url): webpage = self._download_webpage(url, video_id) player_data = self._search_json( - r'<rainews-player\s*data=\'', webpage, 'player_data', video_id, - transform_source=clean_html, fatal=False) + rf'<rai{self._PLAYER_TAG}-player\s*data=\'', webpage, 'player_data', video_id, + transform_source=clean_html, default={}) track_info = player_data.get('track_info') relinker_url = traverse_obj(player_data, 'mediapolis', 'content_url') @@ -770,16 +685,36 @@ def _real_extract(self, url): return { 'id': video_id, - 'title': track_info.get('title') or self._og_search_title(webpage), + 'title': player_data.get('title') or track_info.get('title') or self._og_search_title(webpage), 'upload_date': unified_strdate(track_info.get('date')), 'uploader': strip_or_none(track_info.get('editor') or None), **relinker_info } -class RaiSudtirolIE(RaiBaseIE): - _VALID_URL = r'https?://raisudtirol\.rai\.it/.+?media=(?P<id>[TP]tv\d+)' +class RaiCulturaIE(RaiNewsIE): # XXX: Do not subclass from concrete IE + _VALID_URL = rf'https?://(www\.)?raicultura\.it/(?!articoli)[^?#]+-(?P<id>{RaiBaseIE._UUID_RE})(?:-[^/?#]+)?\.html' + _EMBED_REGEX = [rf'<iframe[^>]+data-src="(?P<url>/iframe/[^?#]+?{RaiBaseIE._UUID_RE}\.html)'] _TESTS = [{ + 'url': 'https://www.raicultura.it/letteratura/articoli/2018/12/Alberto-Asor-Rosa-Letteratura-e-potere-05ba8775-82b5-45c5-a89d-dd955fbde1fb.html', + 'info_dict': { + 'id': '05ba8775-82b5-45c5-a89d-dd955fbde1fb', + 'ext': 'mp4', + 'title': 'Alberto Asor Rosa: Letteratura e potere', + 'duration': 1756, + 'upload_date': '20181206', + 'uploader': 'raicultura', + 'formats': 'count:2', + }, + 'params': {'skip_download': True}, + }] + _PLAYER_TAG = 'cultura' + + +class RaiSudtirolIE(RaiBaseIE): + _VALID_URL = r'https?://raisudtirol\.rai\.it/.+media=(?P<id>\w+)' + _TESTS = [{ + # mp4 file 'url': 'https://raisudtirol.rai.it/la/index.php?media=Ptv1619729460', 'info_dict': { 'id': 'Ptv1619729460', @@ -787,34 +722,62 @@ class RaiSudtirolIE(RaiBaseIE): 'title': 'Euro: trasmisciun d\'economia - 29-04-2021 20:51', 'series': 'Euro: trasmisciun d\'economia', 'upload_date': '20210429', - 'thumbnail': r're:https://raisudtirol\.rai\.it/img/.+?\.jpg', + 'thumbnail': r're:https://raisudtirol\.rai\.it/img/.+\.jpg', 'uploader': 'raisudtirol', - } + 'formats': 'count:1', + }, + 'params': {'skip_download': True}, + }, { + # m3u manifest + 'url': 'https://raisudtirol.rai.it/it/kidsplayer.php?lang=it&media=GUGGUG_P1.smil', + 'info_dict': { + 'id': 'GUGGUG_P1', + 'ext': 'mp4', + 'title': 'GUGGUG! La Prospettiva - Die Perspektive', + 'uploader': 'raisudtirol', + 'formats': 'count:6', + }, + 'params': {'skip_download': True}, }] def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - video_date = self._html_search_regex(r'<span class="med_data">(.+?)</span>', webpage, 'video_date', fatal=False) - video_title = self._html_search_regex(r'<span class="med_title">(.+?)</span>', webpage, 'video_title', fatal=False) - video_url = self._html_search_regex(r'sources:\s*\[\{file:\s*"(.+?)"\}\]', webpage, 'video_url') - video_thumb = self._html_search_regex(r'image: \'(.+?)\'', webpage, 'video_thumb', fatal=False) + video_date = self._html_search_regex( + r'<span class="med_data">(.+?)</span>', webpage, 'video_date', default=None) + video_title = self._html_search_regex([ + r'<span class="med_title">(.+?)</span>', r'title: \'(.+?)\','], + webpage, 'video_title', default=None) + video_url = self._html_search_regex([ + r'sources:\s*\[\{file:\s*"(.+?)"\}\]', + r'<source\s+src="(.+?)"\s+type="application/x-mpegURL"'], + webpage, 'video_url', default=None) - return { - 'id': video_id, - 'title': join_nonempty(video_title, video_date, delim=' - '), - 'series': video_title, - 'upload_date': unified_strdate(video_date), - 'thumbnail': urljoin('https://raisudtirol.rai.it/', video_thumb), - 'uploader': 'raisudtirol', - 'formats': [{ + ext = determine_ext(video_url) + if ext == 'm3u8': + formats = self._extract_m3u8_formats(video_url, video_id) + elif ext == 'mp4': + formats = [{ 'format_id': 'https-mp4', 'url': self._proto_relative_url(video_url), 'width': 1024, 'height': 576, 'fps': 25, - 'vcodec': 'h264', - 'acodec': 'aac', - }], + 'vcodec': 'avc1', + 'acodec': 'mp4a', + }] + else: + formats = [] + self.raise_no_formats(f'Unrecognized media file: {video_url}') + + return { + 'id': video_id, + 'title': join_nonempty(video_title, video_date, delim=' - '), + 'series': video_title if video_date else None, + 'upload_date': unified_strdate(video_date), + 'thumbnail': urljoin('https://raisudtirol.rai.it/', self._html_search_regex( + r'image: \'(.+?)\'', webpage, 'video_thumb', default=None)), + 'uploader': 'raisudtirol', + 'formats': formats, }