diff options
Diffstat (limited to 'hypervideo_dl/extractor/polskieradio.py')
| -rw-r--r-- | hypervideo_dl/extractor/polskieradio.py | 399 |
1 files changed, 291 insertions, 108 deletions
diff --git a/hypervideo_dl/extractor/polskieradio.py b/hypervideo_dl/extractor/polskieradio.py index 99244f6..5bf92b9 100644 --- a/hypervideo_dl/extractor/polskieradio.py +++ b/hypervideo_dl/extractor/polskieradio.py @@ -2,24 +2,24 @@ import itertools import json import math import re +import urllib.parse from .common import InfoExtractor -from ..compat import ( - compat_str, - compat_urllib_parse_unquote, - compat_urlparse -) +from ..compat import compat_str from ..utils import ( - extract_attributes, ExtractorError, InAdvancePagedList, + determine_ext, + extract_attributes, int_or_none, js_to_json, parse_iso8601, strip_or_none, - unified_timestamp, + traverse_obj, unescapeHTML, + unified_timestamp, url_or_none, + urljoin, ) @@ -42,34 +42,17 @@ class PolskieRadioBaseExtractor(InfoExtractor): 'duration': int_or_none(media.get('length')), 'vcodec': 'none' if media.get('provider') == 'audio' else None, }) - entry_title = compat_urllib_parse_unquote(media['desc']) + entry_title = urllib.parse.unquote(media['desc']) if entry_title: entry['title'] = entry_title yield entry -class PolskieRadioIE(PolskieRadioBaseExtractor): - _VALID_URL = r'https?://(?:www\.)?polskieradio(?:24)?\.pl/\d+/\d+/Artykul/(?P<id>[0-9]+)' - _TESTS = [{ # Old-style single broadcast. - 'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943,Prof-Andrzej-Nowak-o-historii-nie-da-sie-myslec-beznamietnie', - 'info_dict': { - 'id': '1587943', - 'title': 'Prof. Andrzej Nowak: o historii nie da się myśleć beznamiętnie', - 'description': 'md5:12f954edbf3120c5e7075e17bf9fc5c5', - }, - 'playlist': [{ - 'md5': '2984ee6ce9046d91fc233bc1a864a09a', - 'info_dict': { - 'id': '1540576', - 'ext': 'mp3', - 'title': 'md5:d4623290d4ac983bf924061c75c23a0d', - 'timestamp': 1456594200, - 'upload_date': '20160227', - 'duration': 2364, - 'thumbnail': r're:^https?://static\.prsa\.pl/images/.*\.jpg$' - }, - }], - }, { # New-style single broadcast. +class PolskieRadioLegacyIE(PolskieRadioBaseExtractor): + # legacy sites + IE_NAME = 'polskieradio:legacy' + _VALID_URL = r'https?://(?:www\.)?polskieradio(?:24)?\.pl/\d+/\d+/[Aa]rtykul/(?P<id>\d+)' + _TESTS = [{ 'url': 'https://www.polskieradio.pl/8/2382/Artykul/2534482,Zagarysci-Poezja-jak-spoiwo', 'info_dict': { 'id': '2534482', @@ -97,16 +80,6 @@ class PolskieRadioIE(PolskieRadioBaseExtractor): 'title': 'Pogłos 29 października godz. 23:01', }, }, { - 'url': 'http://polskieradio.pl/9/305/Artykul/1632955,Bardzo-popularne-slowo-remis', - 'only_matching': True, - }, { - 'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943', - 'only_matching': True, - }, { - # with mp4 video - 'url': 'http://www.polskieradio.pl/9/299/Artykul/1634903,Brexit-Leszek-Miller-swiat-sie-nie-zawali-Europa-bedzie-trwac-dalej', - 'only_matching': True, - }, { 'url': 'https://polskieradio24.pl/130/4503/Artykul/2621876,Narusza-nasza-suwerennosc-Publicysci-o-uzaleznieniu-funduszy-UE-od-praworzadnosci', 'only_matching': True, }] @@ -114,7 +87,9 @@ class PolskieRadioIE(PolskieRadioBaseExtractor): def _real_extract(self, url): playlist_id = self._match_id(url) - webpage = self._download_webpage(url, playlist_id) + webpage, urlh = self._download_webpage_handle(url, playlist_id) + if PolskieRadioIE.suitable(urlh.url): + return self.url_result(urlh.url, PolskieRadioIE, playlist_id) content = self._search_regex( r'(?s)<div[^>]+class="\s*this-article\s*"[^>]*>(.+?)<div[^>]+class="tags"[^>]*>', @@ -153,26 +128,201 @@ class PolskieRadioIE(PolskieRadioBaseExtractor): return self.playlist_result(entries, playlist_id, title, description) -class PolskieRadioCategoryIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?polskieradio\.pl/\d+(?:,[^/]+)?/(?P<id>\d+)' +class PolskieRadioIE(PolskieRadioBaseExtractor): + # new next.js sites + _VALID_URL = r'https?://(?:[^/]+\.)?(?:polskieradio(?:24)?|radiokierowcow)\.pl/artykul/(?P<id>\d+)' + _TESTS = [{ + # articleData, attachments + 'url': 'https://jedynka.polskieradio.pl/artykul/1587943', + 'info_dict': { + 'id': '1587943', + 'title': 'Prof. Andrzej Nowak: o historii nie da się myśleć beznamiętnie', + 'description': 'md5:12f954edbf3120c5e7075e17bf9fc5c5', + }, + 'playlist': [{ + 'md5': '2984ee6ce9046d91fc233bc1a864a09a', + 'info_dict': { + 'id': '7a85d429-5356-4def-a347-925e4ae7406b', + 'ext': 'mp3', + 'title': 'md5:d4623290d4ac983bf924061c75c23a0d', + }, + }], + }, { + # post, legacy html players + 'url': 'https://trojka.polskieradio.pl/artykul/2589163,Czy-wciaz-otrzymujemy-zdjecia-z-sond-Voyager', + 'info_dict': { + 'id': '2589163', + 'title': 'Czy wciąż otrzymujemy zdjęcia z sond Voyager?', + 'description': 'md5:cf1a7f348d63a2db9c0d7a63d1669473', + }, + 'playlist': [{ + 'info_dict': { + 'id': '2577880', + 'ext': 'mp3', + 'title': 'md5:a57d10a0c02abd34dd675cb33707ad5a', + 'duration': 321, + }, + }], + }, { + # data, legacy + 'url': 'https://radiokierowcow.pl/artykul/2694529', + 'info_dict': { + 'id': '2694529', + 'title': 'Zielona fala reliktem przeszłości?', + 'description': 'md5:f20a9a7ed9cb58916c54add94eae3bc0', + }, + 'playlist_count': 3, + }, { + 'url': 'https://trojka.polskieradio.pl/artykul/1632955', + 'only_matching': True, + }, { + # with mp4 video + 'url': 'https://trojka.polskieradio.pl/artykul/1634903', + 'only_matching': True, + }, { + 'url': 'https://jedynka.polskieradio.pl/artykul/3042436,Polityka-wschodnia-ojca-i-syna-Wladyslawa-Lokietka-i-Kazimierza-Wielkiego', + 'only_matching': True, + }] + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + article_data = traverse_obj( + self._search_nextjs_data(webpage, playlist_id), ( + 'props', 'pageProps', (('data', 'articleData'), 'post', 'data')), get_all=False) + + title = strip_or_none(article_data['title']) + + description = strip_or_none(article_data.get('lead')) + + entries = [{ + 'url': entry['file'], + 'ext': determine_ext(entry.get('fileName')), + 'id': self._search_regex( + r'([a-f\d]{8}-(?:[a-f\d]{4}-){3}[a-f\d]{12})', entry['file'], 'entry id'), + 'title': strip_or_none(entry.get('description')) or title, + } for entry in article_data.get('attachments') or () if entry.get('fileType') in ('Audio', )] + + if not entries: + # some legacy articles have no json attachments, but players in body + entries = self._extract_webpage_player_entries(article_data['content'], playlist_id, { + 'title': title, + }) + + return self.playlist_result(entries, playlist_id, title, description) + + +class PolskieRadioAuditionIE(InfoExtractor): + # new next.js sites + IE_NAME = 'polskieradio:audition' + _VALID_URL = r'https?://(?:[^/]+\.)?polskieradio\.pl/audycj[ae]/(?P<id>\d+)' _TESTS = [{ - 'url': 'http://www.polskieradio.pl/7/5102,HISTORIA-ZYWA', + # articles, PR1 + 'url': 'https://jedynka.polskieradio.pl/audycje/5102', 'info_dict': { 'id': '5102', - 'title': 'HISTORIA ŻYWA', + 'title': 'Historia żywa', + 'thumbnail': r're:https://static\.prsa\.pl/images/.+', }, 'playlist_mincount': 38, }, { - 'url': 'http://www.polskieradio.pl/7/4807', + # episodes, PR1 + 'url': 'https://jedynka.polskieradio.pl/audycje/5769', 'info_dict': { - 'id': '4807', - 'title': 'Vademecum 1050. rocznicy Chrztu Polski' + 'id': '5769', + 'title': 'AgroFakty', + 'thumbnail': r're:https://static\.prsa\.pl/images/.+', }, - 'playlist_mincount': 5 + 'playlist_mincount': 269, }, { - 'url': 'http://www.polskieradio.pl/7/129,Sygnaly-dnia?ref=source', - 'only_matching': True + # both episodes and articles, PR3 + 'url': 'https://trojka.polskieradio.pl/audycja/8906', + 'info_dict': { + 'id': '8906', + 'title': 'Trójka budzi', + 'thumbnail': r're:https://static\.prsa\.pl/images/.+', + }, + 'playlist_mincount': 722, }, { + # some articles were "promoted to main page" and thus link to old frontend + 'url': 'https://trojka.polskieradio.pl/audycja/305', + 'info_dict': { + 'id': '305', + 'title': 'Co w mowie piszczy?', + 'thumbnail': r're:https://static\.prsa\.pl/images/.+', + }, + 'playlist_count': 1523, + }] + + def _call_lp3(self, path, query, video_id, note): + return self._download_json( + f'https://lp3test.polskieradio.pl/{path}', video_id, note, + query=query, headers={'x-api-key': '9bf6c5a2-a7d0-4980-9ed7-a3f7291f2a81'}) + + def _entries(self, playlist_id, has_episodes, has_articles): + for i in itertools.count(1) if has_episodes else []: + page = self._call_lp3( + 'AudioArticle/GetListByCategoryId', { + 'categoryId': playlist_id, + 'PageSize': 10, + 'skip': i, + 'format': 400, + }, playlist_id, f'Downloading episode list page {i}') + if not traverse_obj(page, 'data'): + break + for episode in page['data']: + yield { + 'id': str(episode['id']), + 'url': episode['file'], + 'title': episode.get('title'), + 'duration': int_or_none(episode.get('duration')), + 'timestamp': parse_iso8601(episode.get('datePublic')), + } + + for i in itertools.count(1) if has_articles else []: + page = self._call_lp3( + 'Article/GetListByCategoryId', { + 'categoryId': playlist_id, + 'PageSize': 9, + 'skip': i, + 'format': 400, + }, playlist_id, f'Downloading article list page {i}') + if not traverse_obj(page, 'data'): + break + for article in page['data']: + yield { + '_type': 'url_transparent', + 'id': str(article['id']), + 'url': article['url'], + 'title': article.get('shortTitle'), + 'description': traverse_obj(article, ('description', 'lead')), + 'timestamp': parse_iso8601(article.get('datePublic')), + } + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + page_props = traverse_obj( + self._search_nextjs_data(self._download_webpage(url, playlist_id), playlist_id), + ('props', 'pageProps', ('data', None)), get_all=False) + + has_episodes = bool(traverse_obj(page_props, 'episodes', 'audios')) + has_articles = bool(traverse_obj(page_props, 'articles')) + + return self.playlist_result( + self._entries(playlist_id, has_episodes, has_articles), playlist_id, + title=traverse_obj(page_props, ('details', 'name')), + description=traverse_obj(page_props, ('details', 'description', 'lead')), + thumbnail=traverse_obj(page_props, ('details', 'photo'))) + + +class PolskieRadioCategoryIE(InfoExtractor): + # legacy sites + IE_NAME = 'polskieradio:category' + _VALID_URL = r'https?://(?:www\.)?polskieradio\.pl/(?:\d+(?:,[^/]+)?/|[^/]+/Tag)(?P<id>\d+)' + _TESTS = [{ 'url': 'http://www.polskieradio.pl/37,RedakcjaKatolicka/4143,Kierunek-Krakow', 'info_dict': { 'id': '4143', @@ -187,8 +337,35 @@ class PolskieRadioCategoryIE(InfoExtractor): }, 'playlist_mincount': 61 }, { - 'url': 'http://www.polskieradio.pl/7,Jedynka/5102,HISTORIA-ZYWA', - 'only_matching': True, + # billennium tabs + 'url': 'https://www.polskieradio.pl/8/2385', + 'info_dict': { + 'id': '2385', + 'title': 'Droga przez mąkę', + }, + 'playlist_mincount': 111, + }, { + 'url': 'https://www.polskieradio.pl/10/4930', + 'info_dict': { + 'id': '4930', + 'title': 'Teraz K-pop!', + }, + 'playlist_mincount': 392, + }, { + # post back pages, audio content directly without articles + 'url': 'https://www.polskieradio.pl/8,dwojka/7376,nowa-mowa', + 'info_dict': { + 'id': '7376', + 'title': 'Nowa mowa', + }, + 'playlist_mincount': 244, + }, { + 'url': 'https://www.polskieradio.pl/Krzysztof-Dziuba/Tag175458', + 'info_dict': { + 'id': '175458', + 'title': 'Krzysztof Dziuba', + }, + 'playlist_mincount': 420, }, { 'url': 'http://www.polskieradio.pl/8,Dwojka/196,Publicystyka', 'only_matching': True, @@ -196,35 +373,73 @@ class PolskieRadioCategoryIE(InfoExtractor): @classmethod def suitable(cls, url): - return False if PolskieRadioIE.suitable(url) else super(PolskieRadioCategoryIE, cls).suitable(url) + return False if PolskieRadioLegacyIE.suitable(url) else super().suitable(url) def _entries(self, url, page, category_id): content = page + is_billennium_tabs = 'onclick="TB_LoadTab(' in page + is_post_back = 'onclick="__doPostBack(' in page + pagination = page if is_billennium_tabs else None for page_num in itertools.count(2): for a_entry, entry_id in re.findall( - r'(?s)<article[^>]+>.*?(<a[^>]+href=["\']/\d+/\d+/Artykul/(\d+)[^>]+>).*?</article>', + r'(?s)<article[^>]+>.*?(<a[^>]+href=["\'](?:(?:https?)?://[^/]+)?/\d+/\d+/Artykul/(\d+)[^>]+>).*?</article>', content): entry = extract_attributes(a_entry) - href = entry.get('href') - if not href: - continue - yield self.url_result( - compat_urlparse.urljoin(url, href), PolskieRadioIE.ie_key(), - entry_id, entry.get('title')) - mobj = re.search( - r'<div[^>]+class=["\']next["\'][^>]*>\s*<a[^>]+href=(["\'])(?P<url>(?:(?!\1).)+)\1', - content) - if not mobj: - break - next_url = compat_urlparse.urljoin(url, mobj.group('url')) - content = self._download_webpage( - next_url, category_id, 'Downloading page %s' % page_num) + if entry.get('href'): + yield self.url_result( + urljoin(url, entry['href']), PolskieRadioLegacyIE, entry_id, entry.get('title')) + for a_entry in re.findall(r'<span data-media=({[^ ]+})', content): + yield traverse_obj(self._parse_json(a_entry, category_id), { + 'url': 'file', + 'id': 'uid', + 'duration': 'length', + 'title': ('title', {urllib.parse.unquote}), + 'description': ('desc', {urllib.parse.unquote}), + }) + if is_billennium_tabs: + params = self._search_json( + r'<div[^>]+class=["\']next["\'][^>]*>\s*<a[^>]+onclick=["\']TB_LoadTab\(', + pagination, 'next page params', category_id, default=None, close_objects=1, + contains_pattern='.+', transform_source=lambda x: '[%s' % js_to_json(unescapeHTML(x))) + if not params: + break + tab_content = self._download_json( + 'https://www.polskieradio.pl/CMS/TemplateBoxesManagement/TemplateBoxTabContent.aspx/GetTabContent', + category_id, f'Downloading page {page_num}', headers={'content-type': 'application/json'}, + data=json.dumps(dict(zip(( + 'boxInstanceId', 'tabId', 'categoryType', 'sectionId', 'categoryId', 'pagerMode', + 'subjectIds', 'tagIndexId', 'queryString', 'name', 'openArticlesInParentTemplate', + 'idSectionFromUrl', 'maxDocumentAge', 'showCategoryForArticle', 'pageNumber' + ), params))).encode())['d'] + content, pagination = tab_content['Content'], tab_content.get('PagerContent') + elif is_post_back: + target = self._search_regex( + r'onclick=(?:["\'])__doPostBack\((?P<q1>["\'])(?P<target>[\w$]+)(?P=q1)\s*,\s*(?P<q2>["\'])Next(?P=q2)', + content, 'pagination postback target', group='target', default=None) + if not target: + break + content = self._download_webpage( + url, category_id, f'Downloading page {page_num}', + data=urllib.parse.urlencode({ + **self._hidden_inputs(content), + '__EVENTTARGET': target, + '__EVENTARGUMENT': 'Next', + }).encode()) + else: + next_url = urljoin(url, self._search_regex( + r'<div[^>]+class=["\']next["\'][^>]*>\s*<a[^>]+href=(["\'])(?P<url>(?:(?!\1).)+)\1', + content, 'next page url', group='url', default=None)) + if not next_url: + break + content = self._download_webpage(next_url, category_id, f'Downloading page {page_num}') def _real_extract(self, url): category_id = self._match_id(url) - webpage = self._download_webpage(url, category_id) + webpage, urlh = self._download_webpage_handle(url, category_id) + if PolskieRadioAuditionIE.suitable(urlh.url): + return self.url_result(urlh.url, PolskieRadioAuditionIE, category_id) title = self._html_search_regex( - r'<title>([^<]+) - [^<]+ - [^<]+</title>', + r'<title>([^<]+)(?: - [^<]+ - [^<]+| w [Pp]olskie[Rr]adio\.pl\s*)</title>', webpage, 'title', fatal=False) return self.playlist_result( self._entries(url, webpage, category_id), @@ -358,7 +573,7 @@ class PolskieRadioPodcastListIE(PolskieRadioPodcastBaseExtractor): 'entries': InAdvancePagedList( get_page, math.ceil(data['itemCount'] / self._PAGE_SIZE), self._PAGE_SIZE), 'id': str(data['id']), - 'title': data['title'], + 'title': data.get('title'), 'description': data.get('description'), 'uploader': data.get('announcer'), } @@ -374,6 +589,10 @@ class PolskieRadioPodcastIE(PolskieRadioPodcastBaseExtractor): 'ext': 'mp3', 'title': 'Theresa May rezygnuje. Co dalej z brexitem?', 'description': 'md5:e41c409a29d022b70ef0faa61dbded60', + 'episode': 'Theresa May rezygnuje. Co dalej z brexitem?', + 'duration': 2893, + 'thumbnail': 'https://static.prsa.pl/images/58649376-c8a0-4ba2-a714-78b383285f5f.jpg', + 'series': 'Raport o stanie świata', }, }] @@ -389,39 +608,3 @@ class PolskieRadioPodcastIE(PolskieRadioPodcastBaseExtractor): 'Content-Type': 'application/json', }) return self._parse_episode(data[0]) - - -class PolskieRadioRadioKierowcowIE(PolskieRadioBaseExtractor): - _VALID_URL = r'https?://(?:www\.)?radiokierowcow\.pl/artykul/(?P<id>[0-9]+)' - IE_NAME = 'polskieradio:kierowcow' - - _TESTS = [{ - 'url': 'https://radiokierowcow.pl/artykul/2694529', - 'info_dict': { - 'id': '2694529', - 'title': 'Zielona fala reliktem przeszłości?', - 'description': 'md5:343950a8717c9818fdfd4bd2b8ca9ff2', - }, - 'playlist_count': 3, - }] - - def _real_extract(self, url): - media_id = self._match_id(url) - webpage = self._download_webpage(url, media_id) - nextjs_build = self._search_nextjs_data(webpage, media_id)['buildId'] - article = self._download_json( - f'https://radiokierowcow.pl/_next/data/{nextjs_build}/artykul/{media_id}.json?articleId={media_id}', - media_id) - data = article['pageProps']['data'] - title = data['title'] - entries = self._extract_webpage_player_entries(data['content'], media_id, { - 'title': title, - }) - - return { - '_type': 'playlist', - 'id': media_id, - 'entries': entries, - 'title': title, - 'description': data.get('lead'), - } |