diff options
| -rw-r--r-- | test/helper.py | 31 | ||||
| -rw-r--r-- | test/test_InfoExtractor.py | 3 | ||||
| -rw-r--r-- | test/test_download.py | 22 | ||||
| -rw-r--r-- | test/test_jsinterp.py | 8 | ||||
| -rw-r--r-- | test/test_youtube_signature.py | 4 | ||||
| -rwxr-xr-x | youtube_dl/YoutubeDL.py | 49 | ||||
| -rw-r--r-- | youtube_dl/extractor/common.py | 64 | ||||
| -rw-r--r-- | youtube_dl/extractor/extractors.py | 18 | ||||
| -rw-r--r-- | youtube_dl/extractor/mixcloud.py | 30 | ||||
| -rw-r--r-- | youtube_dl/extractor/orf.py | 1103 | ||||
| -rw-r--r-- | youtube_dl/extractor/palcomp3.py | 9 | ||||
| -rw-r--r-- | youtube_dl/extractor/vidlii.py | 59 | ||||
| -rw-r--r-- | youtube_dl/jsinterp.py | 85 | ||||
| -rw-r--r-- | youtube_dl/utils.py | 2 |
14 files changed, 898 insertions, 589 deletions
diff --git a/test/helper.py b/test/helper.py index 5b7e3dfe2..6f2129eff 100644 --- a/test/helper.py +++ b/test/helper.py @@ -5,9 +5,9 @@ import hashlib import json import os.path import re -import types import ssl import sys +import types import unittest import youtube_dl.extractor @@ -181,18 +181,18 @@ def expect_value(self, got, expected, field): op, _, expected_num = expected.partition(':') expected_num = int(expected_num) if op == 'mincount': - assert_func = assertGreaterEqual + assert_func = self.assertGreaterEqual msg_tmpl = 'Expected %d items in field %s, but only got %d' elif op == 'maxcount': - assert_func = assertLessEqual + assert_func = self.assertLessEqual msg_tmpl = 'Expected maximum %d items in field %s, but got %d' elif op == 'count': - assert_func = assertEqual + assert_func = self.assertEqual msg_tmpl = 'Expected exactly %d items in field %s, but got %d' else: assert False assert_func( - self, len(got), expected_num, + len(got), expected_num, msg_tmpl % (expected_num, field, len(got))) return self.assertEqual( @@ -262,27 +262,6 @@ def assertRegexpMatches(self, text, regexp, msg=None): self.assertTrue(m, msg) -def assertGreaterEqual(self, got, expected, msg=None): - if not (got >= expected): - if msg is None: - msg = '%r not greater than or equal to %r' % (got, expected) - self.assertTrue(got >= expected, msg) - - -def assertLessEqual(self, got, expected, msg=None): - if not (got <= expected): - if msg is None: - msg = '%r not less than or equal to %r' % (got, expected) - self.assertTrue(got <= expected, msg) - - -def assertEqual(self, got, expected, msg=None): - if not (got == expected): - if msg is None: - msg = '%r not equal to %r' % (got, expected) - self.assertTrue(got == expected, msg) - - def expect_warnings(ydl, warnings_re): real_warning = ydl.report_warning diff --git a/test/test_InfoExtractor.py b/test/test_InfoExtractor.py index d55d6ad54..09100a1d6 100644 --- a/test/test_InfoExtractor.py +++ b/test/test_InfoExtractor.py @@ -153,6 +153,9 @@ class TestInfoExtractor(unittest.TestCase): ''' search = self.ie._search_nextjs_data(html, 'testID') self.assertEqual(search['props']['pageProps']['video']['id'], 'testid') + search = self.ie._search_nextjs_data( + 'no next.js data here, move along', 'testID', default={'status': 0}) + self.assertEqual(search['status'], 0) def test_search_nuxt_data(self): html = ''' diff --git a/test/test_download.py b/test/test_download.py index df8b370cf..f7d6a23bc 100644 --- a/test/test_download.py +++ b/test/test_download.py @@ -9,8 +9,6 @@ import unittest sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) from test.helper import ( - assertGreaterEqual, - assertLessEqual, expect_warnings, get_params, gettestcases, @@ -36,12 +34,20 @@ from youtube_dl.utils import ( ExtractorError, error_to_compat_str, format_bytes, + IDENTITY, + preferredencoding, UnavailableVideoError, ) from youtube_dl.extractor import get_info_extractor RETRIES = 3 +# Some unittest APIs require actual str +if not isinstance('TEST', str): + _encode_str = lambda s: s.encode(preferredencoding()) +else: + _encode_str = IDENTITY + class YoutubeDL(youtube_dl.YoutubeDL): def __init__(self, *args, **kwargs): @@ -102,7 +108,7 @@ def generator(test_case, tname): def print_skipping(reason): print('Skipping %s: %s' % (test_case['name'], reason)) - self.skipTest(reason) + self.skipTest(_encode_str(reason)) if not ie.working(): print_skipping('IE marked as not _WORKING') @@ -187,16 +193,14 @@ def generator(test_case, tname): expect_info_dict(self, res_dict, test_case.get('info_dict', {})) if 'playlist_mincount' in test_case: - assertGreaterEqual( - self, + self.assertGreaterEqual( len(res_dict['entries']), test_case['playlist_mincount'], 'Expected at least %d in playlist %s, but got only %d' % ( test_case['playlist_mincount'], test_case['url'], len(res_dict['entries']))) if 'playlist_maxcount' in test_case: - assertLessEqual( - self, + self.assertLessEqual( len(res_dict['entries']), test_case['playlist_maxcount'], 'Expected at most %d in playlist %s, but got %d' % ( @@ -243,8 +247,8 @@ def generator(test_case, tname): if params.get('test'): expected_minsize = max(expected_minsize, 10000) got_fsize = os.path.getsize(tc_filename) - assertGreaterEqual( - self, got_fsize, expected_minsize, + self.assertGreaterEqual( + got_fsize, expected_minsize, 'Expected %s to be at least %s, but it\'s only %s ' % (tc_filename, format_bytes(expected_minsize), format_bytes(got_fsize))) diff --git a/test/test_jsinterp.py b/test/test_jsinterp.py index 91b12f544..da8e98020 100644 --- a/test/test_jsinterp.py +++ b/test/test_jsinterp.py @@ -577,9 +577,11 @@ class TestJSInterpreter(unittest.TestCase): def test_unary_operators(self): jsi = JSInterpreter('function f(){return 2 - - - 2;}') self.assertEqual(jsi.call_function('f'), 0) - # fails - # jsi = JSInterpreter('function f(){return 2 + - + - - 2;}') - # self.assertEqual(jsi.call_function('f'), 0) + jsi = JSInterpreter('function f(){return 2 + - + - - 2;}') + self.assertEqual(jsi.call_function('f'), 0) + # https://github.com/ytdl-org/youtube-dl/issues/32815 + jsi = JSInterpreter('function f(){return 0 - 7 * - 6;}') + self.assertEqual(jsi.call_function('f'), 42) """ # fails so far def test_packed(self): diff --git a/test/test_youtube_signature.py b/test/test_youtube_signature.py index f45dfec7c..cafba7a5c 100644 --- a/test/test_youtube_signature.py +++ b/test/test_youtube_signature.py @@ -158,6 +158,10 @@ _NSIG_TESTS = [ 'https://www.youtube.com/s/player/b7910ca8/player_ias.vflset/en_US/base.js', '_hXMCwMt9qE310D', 'LoZMgkkofRMCZQ', ), + ( + 'https://www.youtube.com/s/player/590f65a6/player_ias.vflset/en_US/base.js', + '1tm7-g_A9zsI8_Lay_', 'xI4Vem4Put_rOg', + ), ] diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 6f2aba5ac..dad44435f 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -1039,8 +1039,8 @@ class YoutubeDL(object): elif result_type in ('playlist', 'multi_video'): # Protect from infinite recursion due to recursively nested playlists # (see https://github.com/ytdl-org/youtube-dl/issues/27833) - webpage_url = ie_result['webpage_url'] - if webpage_url in self._playlist_urls: + webpage_url = ie_result.get('webpage_url') # not all pl/mv have this + if webpage_url and webpage_url in self._playlist_urls: self.to_screen( '[download] Skipping already downloaded playlist: %s' % ie_result.get('title') or ie_result.get('id')) @@ -1048,6 +1048,10 @@ class YoutubeDL(object): self._playlist_level += 1 self._playlist_urls.add(webpage_url) + new_result = dict((k, v) for k, v in extra_info.items() if k not in ie_result) + if new_result: + new_result.update(ie_result) + ie_result = new_result try: return self.__process_playlist(ie_result, download) finally: @@ -1593,6 +1597,28 @@ class YoutubeDL(object): self.cookiejar.add_cookie_header(pr) return pr.get_header('Cookie') + def _fill_common_fields(self, info_dict, final=True): + + for ts_key, date_key in ( + ('timestamp', 'upload_date'), + ('release_timestamp', 'release_date'), + ): + if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None: + # Working around out-of-range timestamp values (e.g. negative ones on Windows, + # see http://bugs.python.org/issue1646728) + try: + upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key]) + info_dict[date_key] = compat_str(upload_date.strftime('%Y%m%d')) + except (ValueError, OverflowError, OSError): + pass + + # Auto generate title fields corresponding to the *_number fields when missing + # in order to always have clean titles. This is very common for TV series. + if final: + for field in ('chapter', 'season', 'episode'): + if info_dict.get('%s_number' % field) is not None and not info_dict.get(field): + info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field]) + def process_video_result(self, info_dict, download=True): assert info_dict.get('_type', 'video') == 'video' @@ -1660,24 +1686,7 @@ class YoutubeDL(object): if 'display_id' not in info_dict and 'id' in info_dict: info_dict['display_id'] = info_dict['id'] - for ts_key, date_key in ( - ('timestamp', 'upload_date'), - ('release_timestamp', 'release_date'), - ): - if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None: - # Working around out-of-range timestamp values (e.g. negative ones on Windows, - # see http://bugs.python.org/issue1646728) - try: - upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key]) - info_dict[date_key] = compat_str(upload_date.strftime('%Y%m%d')) - except (ValueError, OverflowError, OSError): - pass - - # Auto generate title fields corresponding to the *_number fields when missing - # in order to always have clean titles. This is very common for TV series. - for field in ('chapter', 'season', 'episode'): - if info_dict.get('%s_number' % field) is not None and not info_dict.get(field): - info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field]) + self._fill_common_fields(info_dict) for cc_kind in ('subtitles', 'automatic_captions'): cc = info_dict.get(cc_kind) diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 7fae9e57b..9b0016d07 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -1169,10 +1169,10 @@ class InfoExtractor(object): def _get_netrc_login_info(self, netrc_machine=None): username = None password = None - netrc_machine = netrc_machine or self._NETRC_MACHINE if self._downloader.params.get('usenetrc', False): try: + netrc_machine = netrc_machine or self._NETRC_MACHINE info = netrc.netrc().authenticators(netrc_machine) if info is not None: username = info[0] @@ -1180,7 +1180,7 @@ class InfoExtractor(object): else: raise netrc.NetrcParseError( 'No authenticators for %s' % netrc_machine) - except (IOError, netrc.NetrcParseError) as err: + except (AttributeError, IOError, netrc.NetrcParseError) as err: self._downloader.report_warning( 'parsing .netrc: %s' % error_to_compat_str(err)) @@ -1490,14 +1490,18 @@ class InfoExtractor(object): return dict((k, v) for k, v in info.items() if v is not None) def _search_nextjs_data(self, webpage, video_id, **kw): - nkw = dict((k, v) for k, v in kw.items() if k in ('transform_source', 'fatal')) - kw.pop('transform_source', None) - next_data = self._search_regex( - r'''<script[^>]+\bid\s*=\s*('|")__NEXT_DATA__\1[^>]*>(?P<nd>[^<]+)</script>''', - webpage, 'next.js data', group='nd', **kw) - if not next_data: - return {} - return self._parse_json(next_data, video_id, **nkw) + # ..., *, transform_source=None, fatal=True, default=NO_DEFAULT + + # TODO: remove this backward compat + default = kw.get('default', NO_DEFAULT) + if default == '{}': + kw['default'] = {} + kw = compat_kwargs(kw) + + return self._search_json( + r'''<script\s[^>]*?\bid\s*=\s*('|")__NEXT_DATA__\1[^>]*>''', + webpage, 'next.js data', video_id, end_pattern='</script>', + **kw) def _search_nuxt_data(self, webpage, video_id, *args, **kwargs): """Parses Nuxt.js metadata. This works as long as the function __NUXT__ invokes is a pure function""" @@ -3029,7 +3033,6 @@ class InfoExtractor(object): transform_source=transform_source, default=None) def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs): - # allow passing `transform_source` through to _find_jwplayer_data() transform_source = kwargs.pop('transform_source', None) kwfind = compat_kwargs({'transform_source': transform_source}) if transform_source else {} @@ -3296,12 +3299,16 @@ class InfoExtractor(object): return ret @classmethod - def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2): - """ Merge two subtitle dictionaries, language by language. """ - ret = dict(subtitle_dict1) - for lang in subtitle_dict2: - ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang]) - return ret + def _merge_subtitles(cls, subtitle_dict1, *subtitle_dicts, **kwargs): + """ Merge subtitle dictionaries, language by language. """ + + # ..., * , target=None + target = kwargs.get('target') or dict(subtitle_dict1) + + for subtitle_dict in subtitle_dicts: + for lang in subtitle_dict: + target[lang] = cls._merge_subtitle_items(target.get(lang, []), subtitle_dict[lang]) + return target def extract_automatic_captions(self, *args, **kwargs): if (self._downloader.params.get('writeautomaticsub', False) @@ -3334,6 +3341,29 @@ class InfoExtractor(object): def _generic_title(self, url): return compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]) + def _yes_playlist(self, playlist_id, video_id, *args, **kwargs): + # smuggled_data=None, *, playlist_label='playlist', video_label='video' + smuggled_data = args[0] if len(args) == 1 else kwargs.get('smuggled_data') + playlist_label = kwargs.get('playlist_label', 'playlist') + video_label = kwargs.get('video_label', 'video') + + if not playlist_id or not video_id: + return not video_id + + no_playlist = (smuggled_data or {}).get('force_noplaylist') + if no_playlist is not None: + return not no_playlist + + video_id = '' if video_id is True else ' ' + video_id + noplaylist = self.get_param('noplaylist') + self.to_screen( + 'Downloading just the {0}{1} because of --no-playlist'.format(video_label, video_id) + if noplaylist else + 'Downloading {0}{1} - add --no-playlist to download just the {2}{3}'.format( + playlist_label, '' if playlist_id is True else ' ' + playlist_id, + video_label, video_id)) + return not noplaylist + class SearchInfoExtractor(InfoExtractor): """ diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 03d035a27..3da5f8020 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -898,21 +898,13 @@ from .ooyala import ( ) from .ora import OraTVIE from .orf import ( - ORFTVthekIE, - ORFFM4IE, + ORFONIE, + ORFONLiveIE, ORFFM4StoryIE, - ORFOE1IE, - ORFOE3IE, - ORFNOEIE, - ORFWIEIE, - ORFBGLIE, - ORFOOEIE, - ORFSTMIE, - ORFKTNIE, - ORFSBGIE, - ORFTIRIE, - ORFVBGIE, ORFIPTVIE, + ORFPodcastIE, + ORFRadioIE, + ORFRadioCollectionIE, ) from .outsidetv import OutsideTVIE from .packtpub import ( diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 69319857d..2b5e2c15c 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -1,3 +1,4 @@ +# coding: utf-8 from __future__ import unicode_literals import itertools @@ -10,7 +11,7 @@ from ..compat import ( compat_ord, compat_str, compat_urllib_parse_unquote, - compat_zip + compat_zip as zip, ) from ..utils import ( int_or_none, @@ -24,7 +25,7 @@ class MixcloudBaseIE(InfoExtractor): def _call_api(self, object_type, object_fields, display_id, username, slug=None): lookup_key = object_type + 'Lookup' return self._download_json( - 'https://www.mixcloud.com/graphql', display_id, query={ + 'https://app.mixcloud.com/graphql', display_id, query={ 'query': '''{ %s(lookup: {username: "%s"%s}) { %s @@ -44,7 +45,7 @@ class MixcloudIE(MixcloudBaseIE): 'ext': 'm4a', 'title': 'Cryptkeeper', 'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.', - 'uploader': 'Daniel Holbach', + 'uploader': 'dholbach', # was: 'Daniel Holbach', 'uploader_id': 'dholbach', 'thumbnail': r're:https?://.*\.jpg', 'view_count': int, @@ -57,7 +58,7 @@ class MixcloudIE(MixcloudBaseIE): 'id': 'gillespeterson_caribou-7-inch-vinyl-mix-chat', 'ext': 'mp3', 'title': 'Caribou 7 inch Vinyl Mix & Chat', - 'description': 'md5:2b8aec6adce69f9d41724647c65875e8', + 'description': r're:Last week Dan Snaith aka Caribou swung by the Brownswood.{136}', 'uploader': 'Gilles Peterson Worldwide', 'uploader_id': 'gillespeterson', 'thumbnail': 're:https?://.*', @@ -65,6 +66,23 @@ class MixcloudIE(MixcloudBaseIE): 'timestamp': 1422987057, 'upload_date': '20150203', }, + 'params': { + 'skip_download': '404 not found', + }, + }, { + 'url': 'https://www.mixcloud.com/gillespeterson/carnival-m%C3%BAsica-popular-brasileira-mix/', + 'info_dict': { + 'id': 'gillespeterson_carnival-música-popular-brasileira-mix', + 'ext': 'm4a', + 'title': 'Carnival Música Popular Brasileira Mix', + 'description': r're:Gilles was recently in Brazil to play at Boiler Room.{208}', + 'timestamp': 1454347174, + 'upload_date': '20160201', + 'uploader': 'Gilles Peterson Worldwide', + 'uploader_id': 'gillespeterson', + 'thumbnail': 're:https?://.*', + 'view_count': int, + }, }, { 'url': 'https://beta.mixcloud.com/RedLightRadio/nosedrip-15-red-light-radio-01-18-2016/', 'only_matching': True, @@ -76,10 +94,10 @@ class MixcloudIE(MixcloudBaseIE): """Encrypt/Decrypt XOR cipher. Both ways are possible because it's XOR.""" return ''.join([ compat_chr(compat_ord(ch) ^ compat_ord(k)) - for ch, k in compat_zip(ciphertext, itertools.cycle(key))]) + for ch, k in zip(ciphertext, itertools.cycle(key))]) def _real_extract(self, url): - username, slug = re.match(self._VALID_URL, url).groups() + username, slug = self._match_valid_url(url).groups() username, slug = compat_urllib_parse_unquote(username), compat_urllib_parse_unquote(slug) track_id = '%s_%s' % (username, slug) diff --git a/youtube_dl/extractor/orf.py b/youtube_dl/extractor/orf.py index 8d537d7ae..1ee78edbc 100644 --- a/youtube_dl/extractor/orf.py +++ b/youtube_dl/extractor/orf.py @@ -1,407 +1,394 @@ # coding: utf-8 from __future__ import unicode_literals +import base64 +import functools import re from .common import InfoExtractor -from ..compat import compat_str +from .youtube import YoutubeIE from ..utils import ( clean_html, determine_ext, + ExtractorError, float_or_none, - HEADRequest, int_or_none, - orderedSet, - remove_end, - str_or_none, + merge_dicts, + mimetype2ext, + parse_age_limit, + parse_iso8601, strip_jsonp, - unescapeHTML, + txt_or_none, unified_strdate, + update_url_query, url_or_none, ) - - -class ORFTVthekIE(InfoExtractor): - IE_NAME = 'orf:tvthek' - IE_DESC = 'ORF TVthek' - _VALID_URL = r'https?://tvthek\.orf\.at/(?:[^/]+/)+(?P<id>\d+)' +from ..traversal import T, traverse_obj + +k_float_or_none = functools.partial(float_or_none, scale=1000) + + +class ORFRadioBase(InfoExtractor): + STATION_INFO = { + 'fm4': ('fm4', 'fm4', 'orffm4'), + 'noe': ('noe', 'oe2n', 'orfnoe'), + 'wien': ('wie', 'oe2w', 'orfwie'), + 'burgenland': ('bgl', 'oe2b', 'orfbgl'), + 'ooe': ('ooe', 'oe2o', 'orfooe'), + 'steiermark': ('stm', 'oe2st', 'orfstm'), + 'kaernten': ('ktn', 'oe2k', 'orfktn'), + 'salzburg': ('sbg', 'oe2s', 'orfsbg'), + 'tirol': ('tir', 'oe2t', 'orftir'), + 'vorarlberg': ('vbg', 'oe2v', 'orfvbg'), + 'oe3': ('oe3', 'oe3', 'orfoe3'), + 'oe1': ('oe1', 'oe1', 'orfoe1'), + } + _ID_NAMES = ('id', 'guid', 'program') + + @classmethod + def _get_item_id(cls, data): + return traverse_obj(data, *cls._ID_NAMES, expected_type=txt_or_none) + + @classmethod + def _get_api_payload(cls, data, expected_id, in_payload=False): + if expected_id not in traverse_obj(data, ('payload',)[:1 if in_payload else 0] + (cls._ID_NAMES, T(txt_or_none))): + raise ExtractorError('Unexpected API data result', video_id=expected_id) + return data['payload'] + + @staticmethod + def _extract_podcast_upload(data): + return traverse_obj(data, { + 'url': ('enclosures', 0, 'url'), + 'ext': ('enclosures', 0, 'type', T(mimetype2ext)), + 'filesize': ('enclosures', 0, 'length', T(int_or_none)), + 'title': ('title', T(txt_or_none)), + 'description': ('description', T(clean_html)), + 'timestamp': (('published', 'postDate'), T(parse_iso8601)), + 'duration': ('duration', T(k_float_or_none)), + 'series': ('podcast', 'title'), + 'uploader': ((('podcast', 'author'), 'station'), T(txt_or_none)), + 'uploader_id': ('podcast', 'channel', T(txt_or_none)), + }, get_all=False) + + @classmethod + def _entries(cls, data, station, item_type=None): + if item_type in ('upload', 'podcast-episode'): + yield merge_dicts({ + 'id': cls._get_item_id(data), + 'ext': 'mp3', + 'vcodec': 'none', + }, cls._extract_podcast_upload(data), rev=True) + return + + loop_station = cls.STATION_INFO[station][1] + for info in traverse_obj(data, ((('streams', Ellipsis), 'stream'), T(lambda v: v if v['loopStreamId'] else None))): + item_id = info['loopStreamId'] + host = info.get('host') or 'loopstream01.apa.at' + yield merge_dicts({ + 'id': item_id.replace('.mp3', ''), + 'ext': 'mp3', + 'url': update_url_query('https://{0}/'.format(host), { + 'channel': loop_station, + 'id': item_id, + }), + 'vcodec': 'none', + # '_old_archive_ids': [make_archive_id(old_ie, video_id)], + }, traverse_obj(data, { + 'title': ('title', T(txt_or_none)), + 'description': ('subtitle', T(clean_html)), + 'uploader': 'station', + 'series': ('programTitle', T(txt_or_none)), + }), traverse_obj(info, { + 'duration': (('duration', + (None, T(lambda x: x['end'] - x['start']))), + T(k_float_or_none), any), + 'timestamp': (('start', 'startISO'), T(parse_iso8601), any), + })) + + +class ORFRadioIE(ORFRadioBase): + IE_NAME = 'orf:sound' + _STATION_RE = '|'.join(map(re.escape, ORFRadioBase.STATION_INFO.keys())) + + _VALID_URL = ( + r'https?://sound\.orf\.at/radio/(?P<station>{0})/sendung/(?P<id>\d+)(?:/(?P<show>\w+))?'.format(_STATION_RE), + r'https?://(?P<station>{0})\.orf\.at/player/(?P<date>\d{{8}})/(?P<id>\d+)'.format(_STATION_RE), + ) _TESTS = [{ - 'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389', + 'url': 'https://sound.orf.at/radio/ooe/sendung/37802/guten-morgen-oberoesterreich-am-feiertag', + 'info_dict': { + 'id': '37802', + 'title': 'Guten Morgen Oberösterreich am Feiertag', + 'description': 'Oberösterreichs meistgehörte regionale Frühsendung.\nRegionale Nachrichten zu jeder halben Stunde.\nModeration: Wolfgang Lehner\nNachrichten: Stephan Schnabl', + }, 'playlist': [{ - 'md5': '2942210346ed779588f428a92db88712', + 'md5': 'f9ff8517dd681b642a2c900e2c9e6085', 'info_dict': { - 'id': '8896777', - 'ext': 'mp4', - 'title': 'Aufgetischt: Mit der Steirischen Tafelrunde', - 'description': 'md5:c1272f0245537812d4e36419c207b67d', - 'duration': 2668, - 'upload_date': '20141208', - }, + 'id': '2024-05-30_0559_tl_66_7DaysThu1_443862', + 'ext': 'mp3', + 'title': 'Guten Morgen Oberösterreich am Feiertag', + 'description': 'Oberösterreichs meistgehörte regionale Frühsendung.\nRegionale Nachrichten zu jeder halben Stunde.\nModeration: Wolfgang Lehner\nNachrichten: Stephan Schnabl', + 'timestamp': 1717041587, + 'upload_date': '20240530', + 'uploader': 'ooe', + 'duration': 14413.0, + } }], - 'skip': 'Blocked outside of Austria / Germany', + 'skip': 'Shows from ORF Sound are only available for 30 days.' }, { - 'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256', + 'url': 'https://oe1.orf.at/player/20240531/758136', + 'md5': '2397717aaf3ae9c22a4f090ee3b8d374', 'info_dict': { - 'id': '7982259', - 'ext': 'mp4', - 'title': 'Best of Ingrid Thurnher', - 'upload_date': '20140527', - 'description': 'Viele Jahre war Ingrid Thurnher das "Gesicht" der ZIB 2. Vor ihrem Wechsel zur ZIB 2 im Jahr 1995 moderierte sie unter anderem "Land und Leute", "Österreich-Bild" und "Niederösterreich heute".', - }, - 'params': { - 'skip_download': True, # rtsp downloads + 'id': '2024-05-31_1905_tl_51_7DaysFri35_2413387', + 'ext': 'mp3', + 'title': '"Who Cares?"', + 'description': 'Europas größte Netzkonferenz re:publica 2024', + 'timestamp': 1717175100, + 'upload_date': '20240531', + 'uploader': 'oe1', + 'duration': 1500, }, - 'skip': 'Blocked outside of Austria / Germany', - }, { - 'url': 'http://tvthek.orf.at/topic/Fluechtlingskrise/10463081/Heimat-Fremde-Heimat/13879132/Senioren-betreuen-Migrantenkinder/13879141', - 'only_matching': True, - }, { - 'url': 'http://tvthek.orf.at/profile/Universum/35429', - 'only_matching': True, + 'skip': 'Shows from ORF Sound are only available for 30 days.' }] def _real_extract(self, url): - playlist_id = self._match_id(url) - webpage = self._download_webpage(url, playlist_id) + m = self._match_valid_url(url) + station, show_id = m.group('station', 'id') + api_station, _, _ = self.STATION_INFO[station] + if 'date' in m.groupdict(): + data = self._download_json( + 'https://audioapi.orf.at/{0}/json/4.0/broadcast/{1}/{2}?_o={3}.orf.at'.format( + api_station, show_id, m.group('date'), station), show_id) + show_id = data['id'] + else: + data = self._download_json( + 'https://audioapi.orf.at/{0}/api/json/5.0/broadcast/{1}?_o=sound.orf.at'.format( + api_station, show_id), show_id) - data_jsb = self._parse_json( - self._search_regex( - r'<div[^>]+class=(["\']).*?VideoPlaylist.*?\1[^>]+data-jsb=(["\'])(?P<json>.+?)\2', - webpage, 'playlist', group='json'), - playlist_id, transform_source=unescapeHTML)['playlist']['videos'] + data = self._get_api_payload(data, show_id, in_payload=True) - entries = [] - for sd in data_jsb: - video_id, title = sd.get('id'), sd.get('title') - if not video_id or not title: - continue - video_id = compat_str(video_id) - formats = [] - for fd in sd['sources']: - src = url_or_none(fd.get('src')) - if not src: - continue - format_id_list = [] - for key in ('delivery', 'quality', 'quality_string'): - value = fd.get(key) - if value: - format_id_list.append(value) - format_id = '-'.join(format_id_list) - ext = determine_ext(src) - if ext == 'm3u8': - m3u8_formats = self._extract_m3u8_formats( - src, video_id, 'mp4', m3u8_id=format_id, fatal=False) - if any('/geoprotection' in f['url'] for f in m3u8_formats): - self.raise_geo_restricted() - formats.extend(m3u8_formats) - elif ext == 'f4m': - formats.extend(self._extract_f4m_formats( - src, video_id, f4m_id=format_id, fatal=False)) - elif ext == 'mpd': - formats.extend(self._extract_mpd_formats( - src, video_id, mpd_id=format_id, fatal=False)) - else: - formats.append({ - 'format_id': format_id, - 'url': src, - 'protocol': fd.get('protocol'), - }) - - # Check for geoblocking. - # There is a property is_geoprotection, but that's always false - geo_str = sd.get('geoprotection_string') - if geo_str: - try: - http_url = next( - f['url'] - for f in formats - if re.match(r'^https?://.*\.mp4$', f['url'])) - except StopIteration: - pass - else: - req = HEADRequest(http_url) - self._request_webpage( - req, video_id, - note='Testing for geoblocking', - errnote=(( - 'This video seems to be blocked outside of %s. ' - 'You may want to try the streaming-* formats.') - % geo_str), - fatal=False) - - self._check_formats(formats, video_id) - self._sort_formats(formats) + # site sends ISO8601 GMT date-times with separate TZ offset, ignored + # TODO: should `..._date` be calculated relative to TZ? - subtitles = {} - for sub in sd.get('subtitles', []): - sub_src = sub.get('src') - if not sub_src: - continue - subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({ - 'url': sub_src, - }) - - upload_date = unified_strdate(sd.get('created_date')) + return merge_dicts( + {'_type': 'multi_video'}, + self.playlist_result( + self._entries(data, station), show_id, + txt_or_none(data.get('title')), + clean_html(data.get('subtitle')))) - thumbnails = [] - preview = sd.get('preview_image_url') - if preview: - thumbnails.append({ - 'id': 'preview', - 'url': preview, - 'preference': 0, - }) - image = sd.get('image_full_url') - if not image and len(data_jsb) == 1: - image = self._og_search_thumbnail(webpage) - if image: - thumbnails.append({ - 'id': 'full', - 'url': image, - 'preference': 1, - }) - entries.append({ - '_type': 'video', - 'id': video_id, - 'title': title, - 'formats': formats, - 'subtitles': subtitles, - 'description': sd.get('description'), - 'duration': int_or_none(sd.get('duration_in_seconds')), - 'upload_date': upload_date, - 'thumbnails': thumbnails, - }) - - return { - '_type': 'playlist', - 'entries': entries, - 'id': playlist_id, - } - - -class ORFRadioIE(InfoExtractor): - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - show_date = mobj.group('date') - show_id = mobj.group('show') +class ORFRadioCollectionIE(ORFRadioBase): + IE_NAME = 'orf:collection' + _VALID_URL = r'https?://sound\.orf\.at/collection/(?P<coll_id>\d+)(?:/(?P<item_id>\d+))?' - data = self._download_json( - 'http://audioapi.orf.at/%s/api/json/current/broadcast/%s/%s' - % (self._API_STATION, show_id, show_date), show_id) - - entries = [] - for info in data['streams']: - loop_stream_id = str_or_none(info.get('loopStreamId')) - if not loop_stream_id: - continue - title = str_or_none(data.get('title')) - if not title: - continue - start = int_or_none(info.get('start'), scale=1000) - end = int_or_none(info.get('end'), scale=1000) - duration = end - start if end and start else None - entries.append({ - 'id': loop_stream_id.replace('.mp3', ''), - 'url': 'https://loopstream01.apa.at/?channel=%s&id=%s' % (self._LOOP_STATION, loop_stream_id), - 'title': title, - 'description': clean_html(data.get('subtitle')), - 'duration': duration, - 'timestamp': start, + _TESTS = [{ + 'url': 'https://sound.orf.at/collection/4/61908/was-das-uberschreiten-des-15-limits-bedeutet', + 'info_dict': { + 'id': '2577582', + }, + 'playlist': [{ + 'md5': '5789cec7d75575ff58d19c0428c80eb3', + 'info_dict': { + 'id': '2024-06-06_1659_tl_54_7DaysThu6_153926', 'ext': 'mp3', - 'series': data.get('programTitle'), - }) - - return { - '_type': 'playlist', - 'id': show_id, - 'title': data.get('title'), - 'description': clean_html(data.get('subtitle')), - 'entries': entries, - } - - -class ORFFM4IE(ORFRadioIE): - IE_NAME = 'orf:fm4' - IE_DESC = 'radio FM4' - _VALID_URL = r'https?://(?P<station>fm4)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>4\w+)' - _API_STATION = 'fm4' - _LOOP_STATION = 'fm4' - - _TEST = { - 'url': 'http://fm4.orf.at/player/20170107/4CC', - 'md5': '2b0be47375432a7ef104453432a19212', + 'title': 'Klimakrise: Was das Überschreiten des 1,5°-Limits bedeutet', + 'timestamp': 1717686674, + 'upload_date': '20240606', + 'uploader': 'fm4', + }, + }], + 'skip': 'Shows from ORF Sound are only available for 30 days.' + }, { + # persistent playlist (FM4 Highlights) + 'url': 'https://sound.orf.at/collection/4/', 'info_dict': { - 'id': '2017-01-07_2100_tl_54_7DaysSat18_31295', - 'ext': 'mp3', - 'title': 'Solid Steel Radioshow', - 'description': 'Die Mixshow von Coldcut und Ninja Tune.', - 'duration': 3599, - 'timestamp': 1483819257, - 'upload_date': '20170107', + 'id': '4', }, - 'skip': 'Shows from ORF radios are only available for 7 days.', - 'only_matching': True, - } - - -class ORFNOEIE(ORFRadioIE): - IE_NAME = 'orf:noe' - IE_DESC = 'Radio Niederösterreich' - _VALID_URL = r'https?://(?P<station>noe)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'noe' - _LOOP_STATION = 'oe2n' - - _TEST = { - 'url': 'https://noe.orf.at/player/20200423/NGM', - 'only_matching': True, - } - - -class ORFWIEIE(ORFRadioIE): - IE_NAME = 'orf:wien' - IE_DESC = 'Radio Wien' - _VALID_URL = r'https?://(?P<station>wien)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'wie' - _LOOP_STATION = 'oe2w' - - _TEST = { - 'url': 'https://wien.orf.at/player/20200423/WGUM', - 'only_matching': True, - } - - -class ORFBGLIE(ORFRadioIE): - IE_NAME = 'orf:burgenland' - IE_DESC = 'Radio Burgenland' - _VALID_URL = r'https?://(?P<station>burgenland)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'bgl' - _LOOP_STATION = 'oe2b' - - _TEST = { - 'url': 'https://burgenland.orf.at/player/20200423/BGM', - 'only_matching': True, - } - - -class ORFOOEIE(ORFRadioIE): - IE_NAME = 'orf:oberoesterreich' - IE_DESC = 'Radio Oberösterreich' - _VALID_URL = r'https?://(?P<station>ooe)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'ooe' - _LOOP_STATION = 'oe2o' + 'playlist_mincount': 10, + 'playlist_maxcount': 13, + }] - _TEST = { - 'url': 'https://ooe.orf.at/player/20200423/OGMO', - 'only_matching': True, - } + def _real_extract(self, url): + coll_id, item_id = self._match_valid_url(url).group('coll_id', 'item_id') + data = self._download_json( + 'https://collector.orf.at/api/frontend/collections/{0}?_o=sound.orf.at'.format( + coll_id), coll_id) + data = self._get_api_payload(data, coll_id, in_payload=True) + + def yield_items(): + for item in traverse_obj(data, ( + 'content', 'items', lambda _, v: any(k in v['target']['params'] for k in self._ID_NAMES))): + if item_id is None or item_id == txt_or_none(item.get('id')): + target = item['target'] + typed_item_id = self._get_item_id(target['params']) + station = target['params'].get('station') + item_type = target.get('type') + if typed_item_id and (station or item_type): + yield station, typed_item_id, item_type + if item_id is not None: + break + else: + if item_id is not None: + raise ExtractorError('Item not found in collection', + video_id=coll_id, expected=True) + + def item_playlist(station, typed_item_id, item_type): + if item_type == 'upload': + item_data = self._download_json('https://audioapi.orf.at/radiothek/api/2.0/upload/{0}?_o=sound.orf.at'.format( + typed_item_id), typed_item_id) + elif item_type == 'podcast-episode': + item_data = self._download_json('https://audioapi.orf.at/radiothek/api/2.0/episode/{0}?_o=sound.orf.at'.format( + typed_item_id), typed_item_id) + else: + api_station, _, _ = self.STATION_INFO[station] + item_data = self._download_json( + 'https://audioapi.orf.at/{0}/api/json/5.0/{1}/{2}?_o=sound.orf.at'.format( + api_station, item_type or 'broadcastitem', typed_item_id), typed_item_id) + item_data = self._get_api_payload(item_data, typed_item_id, in_payload=True) -class ORFSTMIE(ORFRadioIE): - IE_NAME = 'orf:steiermark' - IE_DESC = 'Radio Steiermark' - _VALID_URL = r'https?://(?P<station>steiermark)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'stm' - _LOOP_STATION = 'oe2st' + return merge_dicts( + {'_type': 'multi_video'}, + self.playlist_result( + self._entries(item_data, station, item_type), typed_item_id, + txt_or_none(data.get('title')), + clean_html(data.get('subtitle')))) - _TEST = { - 'url': 'https://steiermark.orf.at/player/20200423/STGMS', - 'only_matching': True, - } + def yield_item_entries(): + for station, typed_id, item_type in yield_items(): + yield item_playlist(station, typed_id, item_type) + if item_id is not None: + # coll_id = '/'.join((coll_id, item_id)) + return next(yield_item_entries()) -class ORFKTNIE(ORFRadioIE): - IE_NAME = 'orf:kaernten' - IE_DESC = 'Radio Kärnten' - _VALID_URL = r'https?://(?P<station>kaernten)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'ktn' - _LOOP_STATION = 'oe2k' + return self.playlist_result(yield_item_entries(), coll_id, data.get('title')) - _TEST = { - 'url': 'https://kaernten.orf.at/player/20200423/KGUMO', - 'only_matching': True, - } +class ORFPodcastIE(ORFRadioBase): + IE_NAME = 'orf:podcast' + _STATION_RE = '|'.join(map(re.escape, (x[0] for x in ORFRadioBase.STATION_INFO.values()))) + '|tv' + _VALID_URL = r'https?://sound\.orf\.at/podcast/(?P<station>{0})/(?P<show>[\w-]+)/(?P<id>[\w-]+)'.format(_STATION_RE) + _TESTS = [{ + 'url': 'https://sound.orf.at/podcast/stm/der-kraeutertipp-von-christine-lackner/rotklee', + 'md5': '1f2bab2ba90c2ce0c2754196ea78b35f', + 'info_dict': { + 'id': 'der-kraeutertipp-von-christine-lackner/rotklee', + 'ext': 'mp3', + 'title': 'Rotklee', + 'description': 'In der Natur weit verbreitet - in der Medizin längst anerkennt: Rotklee. Dieser Podcast begleitet die Sendung "Radio Steiermark am Vormittag", Radio Steiermark, 28. Mai 2024.', + 'timestamp': 1716891761, + 'upload_date': '20240528', + 'uploader_id': 'stm_kraeutertipp', + 'uploader': 'ORF Radio Steiermark', + 'duration': 101, + 'series': 'Der Kräutertipp von Christine Lackner', + }, + 'skip': 'ORF podcasts are only available for a limited time' + }] -class ORFSBGIE(ORFRadioIE): - IE_NAME = 'orf:salzburg' - IE_DESC = 'Radio Salzburg' - _VALID_URL = r'https?://(?P<station>salzburg)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'sbg' - _LOOP_STATION = 'oe2s' + _ID_NAMES = ('slug', 'guid') - _TEST = { - 'url': 'https://salzburg.orf.at/player/20200423/SGUM', - 'only_matching': True, - } + def _real_extract(self, url): + station, show, show_id = self._match_valid_url(url).group('station', 'show', 'id') + data = self._download_json( + 'https://audioapi.orf.at/radiothek/api/2.0/podcast/{0}/{1}/{2}'.format( + station, show, show_id), show_id) + data = self._get_api_payload(data, show_id, in_payload=True) + return merge_dicts({ + 'id': '/'.join((show, show_id)), + 'ext': 'mp3', + 'vcodec': 'none', + }, self._extract_podcast_upload(data), rev=True) -class ORFTIRIE(ORFRadioIE): - IE_NAME = 'orf:tirol' - IE_DESC = 'Radio Tirol' - _VALID_URL = r'https?://(?P<station>tirol)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'tir' - _LOOP_STATION = 'oe2t' - _TEST = { - 'url': 'https://tirol.orf.at/player/20200423/TGUMO', - 'only_matching': True, - } +class ORFIPTVBase(InfoExtractor): + _TITLE_STRIP_RE = '' + def _extract_video(self, video_id, webpage, fatal=False): -class ORFVBGIE(ORFRadioIE): - IE_NAME = 'orf:vorarlberg' - IE_DESC = 'Radio Vorarlberg' - _VALID_URL = r'https?://(?P<station>vorarlberg)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'vbg' - _LOOP_STATION = 'oe2v' + data = self._download_json( + 'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id, + video_id)[0] - _TEST = { - 'url': 'https://vorarlberg.orf.at/player/20200423/VGUM', - 'only_matching': True, - } + video = traverse_obj(data, ( + 'sources', ('default', 'q8c'), + T(lambda x: x if x['loadBalancerUrl'] else None), + any)) + load_balancer_url = video['loadBalancerUrl'] -class ORFOE3IE(ORFRadioIE): - IE_NAME = 'orf:oe3' - IE_DESC = 'Radio Österreich 3' - _VALID_URL = r'https?://(?P<station>oe3)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'oe3' - _LOOP_STATION = 'oe3' + try: + rendition = self._download_json( + load_balancer_url, video_id, transform_source=strip_jsonp) + except ExtractorError: + rendition = None + + if not rendition: + rendition = { + 'redirect': { + 'smil': re.sub( + r'(/)jsonp(/.+\.)mp4$', r'\1dash\2smil/manifest.mpd', + load_balancer_url), + }, + } - _TEST = { - 'url': 'https://oe3.orf.at/player/20200424/3WEK', - 'only_matching': True, - } + f = traverse_obj(video, { + 'abr': ('audioBitrate', T(int_or_none)), + 'vbr': ('bitrate', T(int_or_none)), + 'fps': ('videoFps', T(int_or_none)), + 'width': ('videoWidth', T(int_or_none)), + 'height': ('videoHeight', T(int_or_none)), + }) + formats = [] + for format_id, format_url in traverse_obj(rendition, ( + 'redirect', T(dict.items), Ellipsis)): + if format_id == 'rtmp': + ff = f.copy() + ff.update({ + 'url': format_url, + 'format_id': format_id, + }) + formats.append(ff) + elif determine_ext(format_url) == 'f4m': + formats.extend(self._extract_f4m_formats( + format_url, video_id, f4m_id=format_id)) + elif determine_ext(format_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', m3u8_id=format_id, + entry_protocol='m3u8_native')) + elif determine_ext(format_url) == 'mpd': + formats.extend(self._extract_mpd_formats( + format_url, video_id, mpd_id=format_id)) -class ORFOE1IE(ORFRadioIE): - IE_NAME = 'orf:oe1' - IE_DESC = 'Radio Österreich 1' - _VALID_URL = r'https?://(?P<station>oe1)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)' - _API_STATION = 'oe1' - _LOOP_STATION = 'oe1' + if formats or fatal: + self._sort_formats(formats) + else: + return - _TEST = { - 'url': 'http://oe1.orf.at/player/20170108/456544', - 'md5': '34d8a6e67ea888293741c86a099b745b', - 'info_dict': { - 'id': '2017-01-08_0759_tl_51_7DaysSun6_256141', - 'ext': 'mp3', - 'title': 'Morgenjournal', - 'duration': 609, - 'timestamp': 1483858796, - 'upload_date': '20170108', - }, - 'skip': 'Shows from ORF radios are only available for 7 days.' - } + return merge_dicts({ + 'id': video_id, + 'title': re.sub(self._TITLE_STRIP_RE, '', self._og_search_title(webpage)), + 'description': self._og_search_description(webpage), + 'upload_date': unified_strdate(self._html_search_meta( + 'dc.date', webpage, 'upload date', fatal=False)), + 'formats': formats, + }, traverse_obj(data, { + 'duration': ('duration', T(k_float_or_none)), + 'thumbnail': ('sources', 'default', 'preview', T(url_or_none)), + }), rev=True) -class ORFIPTVIE(InfoExtractor): +class ORFIPTVIE(ORFIPTVBase): IE_NAME = 'orf:iptv' IE_DESC = 'iptv.ORF.at' + _WORKING = False # URLs redirect to orf.at/ _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)' + _TITLE_STRIP_RE = r'\s+-\s+iptv\.ORF\.at\S*$' _TEST = { 'url': 'http://iptv.orf.at/stories/2275236/', @@ -426,74 +413,32 @@ class ORFIPTVIE(InfoExtractor): video_id = self._search_regex( r'data-video(?:id)?="(\d+)"', webpage, 'video id') - data = self._download_json( - 'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id, - video_id)[0] - - duration = float_or_none(data['duration'], 1000) + return self._extract_video(video_id, webpage) - video = data['sources']['default'] - load_balancer_url = video['loadBalancerUrl'] - abr = int_or_none(video.get('audioBitrate')) - vbr = int_or_none(video.get('bitrate')) - fps = int_or_none(video.get('videoFps')) - width = int_or_none(video.get('videoWidth')) - height = int_or_none(video.get('videoHeight')) - thumbnail = video.get('preview') - - rendition = self._download_json( - load_balancer_url, video_id, transform_source=strip_jsonp) - - f = { - 'abr': abr, - 'vbr': vbr, - 'fps': fps, - 'width': width, - 'height': height, - } - formats = [] - for format_id, format_url in rendition['redirect'].items(): - if format_id == 'rtmp': - ff = f.copy() - ff.update({ - 'url': format_url, - 'format_id': format_id, - }) - formats.append(ff) - elif determine_ext(format_url) == 'f4m': - formats.extend(self._extract_f4m_formats( - format_url, video_id, f4m_id=format_id)) - elif determine_ext(format_url) == 'm3u8': - formats.extend(self._extract_m3u8_formats( - format_url, video_id, 'mp4', m3u8_id=format_id)) - else: - continue - self._sort_formats(formats) - - title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at') - description = self._og_search_description(webpage) - upload_date = unified_strdate(self._html_search_meta( - 'dc.date', webpage, 'upload date')) - - return { - 'id': video_id, - 'title': title, - 'description': description, - 'duration': duration, - 'thumbnail': thumbnail, - 'upload_date': upload_date, - 'formats': formats, - } - - -class ORFFM4StoryIE(InfoExtractor): +class ORFFM4StoryIE(ORFIPTVBase): IE_NAME = 'orf:fm4:story' IE_DESC = 'fm4.orf.at stories' _VALID_URL = r'https?://fm4\.orf\.at/stories/(?P<id>\d+)' + _TITLE_STRIP_RE = r'\s+-\s+fm4\.ORF\.at\s*$' - _TEST = { + _TESTS = [{ + 'url': 'https://fm4.orf.at/stories/3041554/', + 'add_ie': ['Youtube'], + 'info_dict': { + 'id': '3041554', + 'title': 'Is The EU Green Deal In Mortal Danger?', + }, + 'playlist_count': 4, + 'params': { + 'format': 'bestvideo', + }, + }, { 'url': 'http://fm4.orf.at/stories/2865738/', + 'info_dict': { + 'id': '2865738', + 'title': 'Manu Delago und Inner Tongue live', + }, 'playlist': [{ 'md5': 'e1c2c706c45c7b34cf478bbf409907ca', 'info_dict': { @@ -510,83 +455,311 @@ class ORFFM4StoryIE(InfoExtractor): 'info_dict': { 'id': '547798', 'ext': 'flv', - 'title': 'Manu Delago und Inner Tongue live (2)', + 'title': 'Manu Delago und Inner Tongue https://vod-ww.mdn.ors.at/cms-worldwide_episodes_nas/_definst_/nas/cms-worldwide_episodes/online/14228823_0005.smil/chunklist_b992000_vo.m3u8live (2)', 'duration': 1504.08, 'thumbnail': r're:^https?://.*\.jpg$', 'upload_date': '20170913', 'description': 'Manu Delago und Inner Tongue haben bei der FM4 Soundpark Session live alles gegeben. Hier gibt es Fotos und die gesamte Session als Video.', }, }], - } + 'skip': 'Videos gone', + }] def _real_extract(self, url): story_id = self._match_id(url) webpage = self._download_webpage(url, story_id) entries = [] - all_ids = orderedSet(re.findall(r'data-video(?:id)?="(\d+)"', webpage)) - for idx, video_id in enumerate(all_ids): - data = self._download_json( - 'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id, - video_id)[0] + seen_ids = set() + for idx, video_id in enumerate(re.findall(r'data-video(?:id)?="(\d+)"', webpage)): + if video_id in seen_ids: + continue + seen_ids.add(video_id) + entry = self._extract_video(video_id, webpage, fatal=False) + if not entry: + continue + + if idx >= 1: + # Titles are duplicates, make them unique + entry['title'] = '%s (%d)' % (entry['title'], idx) - duration = float_or_none(data['duration'], 1000) + entries.append(entry) - video = data['sources']['q8c'] - load_balancer_url = video['loadBalancerUrl'] - abr = int_or_none(video.get('audioBitrate')) - vbr = int_or_none(video.get('bitrate')) - fps = int_or_none(video.get('videoFps')) - width = int_or_none(video.get('videoWidth')) - height = int_or_none(video.get('videoHeight')) - thumbnail = video.get('preview') + seen_ids = set() + for yt_id in re.findall( + r'data-id\s*=\s*["\']([\w-]+)[^>]+\bclass\s*=\s*["\']youtube\b', + webpage): + if yt_id in seen_ids: + continue + seen_ids.add(yt_id) + if YoutubeIE.suitable(yt_id): + entries.append(self.url_result(yt_id, ie='Youtube', video_id=yt_id)) + + return self.playlist_result( + entries, story_id, + re.sub(self._TITLE_STRIP_RE, '', self._og_search_title(webpage, default='') or None)) + + +class ORFONBase(InfoExtractor): + _ENC_PFX = '3dSlfek03nsLKdj4Jsd' + _API_PATH = 'episode' + + def _call_api(self, video_id, **kwargs): + encrypted_id = base64.b64encode('{0}{1}'.format( + self._ENC_PFX, video_id).encode('utf-8')).decode('ascii') + return self._download_json( + 'https://api-tvthek.orf.at/api/v4.3/public/{0}/encrypted/{1}'.format( + self._API_PATH, encrypted_id), + video_id, **kwargs) + + @classmethod + def _parse_metadata(cls, api_json): + return traverse_obj(api_json, { + 'id': ('id', T(int), T(txt_or_none)), + 'age_limit': ('age_classification', T(parse_age_limit)), + 'duration': ((('exact_duration', T(k_float_or_none)), + ('duration_second', T(float_or_none))),), + 'title': (('title', 'headline'), T(txt_or_none)), + 'description': (('description', 'teaser_text'), T(txt_or_none)), + # 'media_type': ('video_type', T(txt_or_none)), + 'thumbnail': ('_embedded', 'image', 'public_urls', 'highlight_teaser', 'url', T(url_or_none)), + 'timestamp': (('date', 'episode_date'), T(parse_iso8601)), + 'release_timestamp': ('release_date', T(parse_iso8601)), + # 'modified_timestamp': ('updated_at', T(parse_iso8601)), + }, get_all=False) + + def _extract_video(self, video_id, segment_id): + # Not a segmented episode: return single video + # Segmented episode without valid segment id: return entire playlist + # Segmented episode with valid segment id and yes-playlist: return entire playlist + # Segmented episode with valid segment id and no-playlist: return single video corresponding to segment id + # If a multi_video playlist would be returned, but an unsegmented source exists, that source is chosen instead. + + api_json = self._call_api(video_id) + + if traverse_obj(api_json, 'is_drm_protected'): + self.report_drm(video_id) + + # updates formats, subtitles + def extract_sources(src_json, video_id): + for manifest_type in traverse_obj(src_json, ('sources', T(dict.keys), Ellipsis)): + for manifest_url in traverse_obj(src_json, ('sources', manifest_type, Ellipsis, 'src', T(url_or_none))): + if manifest_type == 'hls': + fmts, subs = self._extract_m3u8_formats( + manifest_url, video_id, fatal=False, m3u8_id='hls', + ext='mp4', entry_protocol='m3u8_native'), {} + for f in fmts: + if '_vo.' in f['url']: + f['acodec'] = 'none' + elif manifest_type == 'dash': + fmts, subs = self._extract_mpd_formats_and_subtitles( + manifest_url, video_id, fatal=False, mpd_id='dash') + else: + continue + formats.extend(fmts) + self._merge_subtitles(subs, target=subtitles) + + formats, subtitles = [], {} + if segment_id is None: + extract_sources(api_json, video_id) + if not formats: + segments = traverse_obj(api_json, ( + '_embedded', 'segments', lambda _, v: v['id'])) + if len(segments) > 1 and segment_id is not None: + if not self._yes_playlist(video_id, segment_id, playlist_label='collection', video_label='segment'): + segments = [next(s for s in segments if txt_or_none(s['id']) == segment_id)] + + entries = [] + for seg in segments: + formats, subtitles = [], {} + extract_sources(seg, segment_id) + self._sort_formats(formats) + entries.append(merge_dicts({ + 'formats': formats, + 'subtitles': subtitles, + }, self._parse_metadata(seg), rev=True)) + result = merge_dicts( + {'_type': 'multi_video' if len(entries) > 1 else 'playlist'}, + self._parse_metadata(api_json), + self.playlist_result(entries, video_id)) + # not yet processed in core for playlist/multi + self._downloader._fill_common_fields(result) + return result + else: + self._sort_formats(formats) - rendition = self._download_json( - load_balancer_url, video_id, transform_source=strip_jsonp) + for sub_url in traverse_obj(api_json, ( + '_embedded', 'subtitle', + ('xml_url', 'sami_url', 'stl_url', 'ttml_url', 'srt_url', 'vtt_url'), + T(url_or_none))): + self._merge_subtitles({'de': [{'url': sub_url}]}, target=subtitles) - f = { - 'abr': abr, - 'vbr': vbr, - 'fps': fps, - 'width': width, - 'height': height, - } + return merge_dicts({ + 'id': video_id, + 'formats': formats, + 'subtitles': subtitles, + # '_old_archive_ids': [self._downloader._make_archive_id({'ie_key': 'ORFTVthek', 'id': video_id})], + }, self._parse_metadata(api_json), rev=True) - formats = [] - for format_id, format_url in rendition['redirect'].items(): - if format_id == 'rtmp': - ff = f.copy() - ff.update({ - 'url': format_url, - 'format_id': format_id, - }) - formats.append(ff) - elif determine_ext(format_url) == 'f4m': - formats.extend(self._extract_f4m_formats( - format_url, video_id, f4m_id=format_id)) - elif determine_ext(format_url) == 'm3u8': - formats.extend(self._extract_m3u8_formats( - format_url, video_id, 'mp4', m3u8_id=format_id)) - else: - continue - self._sort_formats(formats) + def _real_extract(self, url): + video_id, segment_id = self._match_valid_url(url).group('id', 'segment') + webpage = self._download_webpage(url, video_id) - title = remove_end(self._og_search_title(webpage), ' - fm4.ORF.at') - if idx >= 1: - # Titles are duplicates, make them unique - title += ' (' + str(idx + 1) + ')' - description = self._og_search_description(webpage) - upload_date = unified_strdate(self._html_search_meta( - 'dc.date', webpage, 'upload date')) - - entries.append({ - 'id': video_id, - 'title': title, - 'description': description, - 'duration': duration, - 'thumbnail': thumbnail, - 'upload_date': upload_date, - 'formats': formats, - }) - - return self.playlist_result(entries) + # ORF doesn't like 410 or 404 + if self._search_regex(r'<div\b[^>]*>\s*(Nicht mehr verfügbar)\s*</div>', webpage, 'Availability', default=False): + raise ExtractorError('Content is no longer available', expected=True, video_id=video_id) + + return merge_dicts({ + 'id': video_id, + 'title': self._html_search_meta(['og:title', 'twitter:title'], webpage, default=None), + 'description': self._html_search_meta( + ['description', 'og:description', 'twitter:description'], webpage, default=None), + }, self._search_json_ld(webpage, video_id, default={}), + self._extract_video(video_id, segment_id), + rev=True) + + +class ORFONIE(ORFONBase): + IE_NAME = 'orf:on' + _VALID_URL = r'https?://on\.orf\.at/video/(?P<id>\d+)(?:/(?P<segment>\d+))?' + _TESTS = [{ + 'url': 'https://on.orf.at/video/14210000/school-of-champions-48', + 'info_dict': { + 'id': '14210000', + 'ext': 'mp4', + 'duration': 2651.08, + 'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0167/98/thumb_16697671_segments_highlight_teaser.jpeg', + 'title': 'School of Champions (4/8)', + 'description': r're:(?s)Luca hat sein ganzes Leben in den Bergen Südtirols verbracht und ist bei seiner Mutter aufgewachsen, .{1029} Leo$', + # 'media_type': 'episode', + 'timestamp': 1706558922, + 'upload_date': '20240129', + 'release_timestamp': 1706472362, + 'release_date': '20240128', + # 'modified_timestamp': 1712756663, + # 'modified_date': '20240410', + # '_old_archive_ids': ['orftvthek 14210000'], + }, + 'params': { + 'format': 'bestvideo', + }, + 'skip': 'Available until 2024-08-12', + }, { + 'url': 'https://on.orf.at/video/3220355', + 'md5': '925a93b2b9a37da5c9b979d7cf71aa2e', + 'info_dict': { + 'id': '3220355', + 'ext': 'mp4', + 'duration': 445.04, + 'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0002/60/thumb_159573_segments_highlight_teaser.png', + 'title': '50 Jahre Burgenland: Der Festumzug', + 'description': r're:(?s)Aus allen Landesteilen zogen festlich geschmückte Wagen und Musikkapellen .{270} Jenakowitsch$', + # 'media_type': 'episode', + 'timestamp': 52916400, + 'upload_date': '19710905', + 'release_timestamp': 52916400, + 'release_date': '19710905', + # 'modified_timestamp': 1498536049, + # 'modified_date': '20170627', + # '_old_archive_ids': ['orftvthek 3220355'], + }, + }, { + # Video with multiple segments selecting the second segment + 'url': 'https://on.orf.at/video/14226549/15639808/jugendbande-einbrueche-aus-langeweile', + 'md5': 'fc151bba8c05ea77ab5693617e4a33d3', + 'info_dict': { + 'id': '15639808', + 'ext': 'mp4', + 'duration': 97.707, + 'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0175/43/thumb_17442704_segments_highlight_teaser.jpg', + 'title': 'Jugendbande: Einbrüche aus Langeweile', + 'description': r're:Jugendbande: Einbrüche aus Langeweile \| Neuer Kinder- und .{259} Wanda$', + # 'media_type': 'segment', + 'timestamp': 1715792400, + 'upload_date': '20240515', + # 'modified_timestamp': 1715794394, + # 'modified_date': '20240515', + # '_old_archive_ids': ['orftvthek 15639808'], + }, + 'params': { + 'noplaylist': True, + 'format': 'bestvideo', + }, + 'skip': 'Available until 2024-06-14', + }, { + # Video with multiple segments and no combined version + 'url': 'https://on.orf.at/video/14227864/formel-1-grosser-preis-von-monaco-2024', + 'info_dict': { + '_type': 'multi_video', + 'id': '14227864', + 'duration': 18410.52, + 'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0176/04/thumb_17503881_segments_highlight_teaser.jpg', + 'title': 'Formel 1: Großer Preis von Monaco 2024', + 'description': 'md5:aeeb010710ccf70ce28ccb4482243d4f', + # 'media_type': 'episode', + 'timestamp': 1716721200, + 'upload_date': '20240526', + 'release_timestamp': 1716721802, + 'release_date': '20240526', + # 'modified_timestamp': 1716884702, + # 'modified_date': '20240528', + }, + 'playlist_count': 42, + 'skip': 'Gone: Nicht mehr verfügbar', + }, { + # Video with multiple segments, but with combined version + 'url': 'https://on.orf.at/video/14228172', + 'info_dict': { + 'id': '14228172', + 'ext': 'mp4', + 'duration': 3294.878, + 'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0176/29/thumb_17528242_segments_highlight_teaser.jpg', + 'title': 'Willkommen Österreich mit Stermann & Grissemann', + 'description': r're:Zum Saisonfinale freuen sich die urlaubsreifen Gastgeber Stermann und .{1863} Geschichten\.$', + # 'media_type': 'episode', + 'timestamp': 1716926584, + 'upload_date': '20240528', + 'release_timestamp': 1716919202, + 'release_date': '20240528', + # 'modified_timestamp': 1716968045, + # 'modified_date': '20240529', + # '_old_archive_ids': ['orftvthek 14228172'], + }, + 'params': { + 'format': 'bestvideo', + }, + 'skip': 'Gone: Nicht mehr verfügbar', + }] + + +class ORFONLiveIE(ORFONBase): + _ENC_PFX = '8876324jshjd7293ktd' + _API_PATH = 'livestream' + _VALID_URL = r'https?://on\.orf\.at/livestream/(?P<id>\d+)(?:/(?P<segment>\d+))?' + _TESTS = [{ + 'url': 'https://on.orf.at/livestream/14320204/pressekonferenz-neos-zu-aktuellen-entwicklungen', + 'info_dict': { + 'id': '14320204', + 'ext': 'mp4', + 'title': 'Pressekonferenz: Neos zu aktuellen Entwicklungen', + 'description': r're:(?s)Neos-Chefin Beate Meinl-Reisinger informi.{598}ng\."', + 'timestamp': 1716886335, + 'upload_date': '20240528', + # 'modified_timestamp': 1712756663, + # 'modified_date': '20240410', + # '_old_archive_ids': ['orftvthek 14210000'], + }, + 'params': { + 'format': 'bestvideo', + }, + }] + + @classmethod + def _parse_metadata(cls, api_json): + return merge_dicts( + super(ORFONLiveIE, cls)._parse_metadata(api_json), + traverse_obj(api_json, { + 'timestamp': ('updated_at', T(parse_iso8601)), + 'release_timestamp': ('start', T(parse_iso8601)), + 'is_live': True, + })) diff --git a/youtube_dl/extractor/palcomp3.py b/youtube_dl/extractor/palcomp3.py index fb29d83f9..60f7a4d48 100644 --- a/youtube_dl/extractor/palcomp3.py +++ b/youtube_dl/extractor/palcomp3.py @@ -8,7 +8,7 @@ from ..compat import compat_str from ..utils import ( int_or_none, str_or_none, - try_get, + traverse_obj, ) @@ -109,7 +109,7 @@ class PalcoMP3ArtistIE(PalcoMP3BaseIE): } name''' - @ classmethod + @classmethod def suitable(cls, url): return False if re.match(PalcoMP3IE._VALID_URL, url) else super(PalcoMP3ArtistIE, cls).suitable(url) @@ -118,7 +118,8 @@ class PalcoMP3ArtistIE(PalcoMP3BaseIE): artist = self._call_api(artist_slug, self._ARTIST_FIELDS_TMPL)['artist'] def entries(): - for music in (try_get(artist, lambda x: x['musics']['nodes'], list) or []): + for music in traverse_obj(artist, ( + 'musics', 'nodes', lambda _, m: m['musicID'])): yield self._parse_music(music) return self.playlist_result( @@ -137,7 +138,7 @@ class PalcoMP3VideoIE(PalcoMP3BaseIE): 'title': 'Maiara e Maraisa - Você Faz Falta Aqui - DVD Ao Vivo Em Campo Grande', 'description': 'md5:7043342c09a224598e93546e98e49282', 'upload_date': '20161107', - 'uploader_id': 'maiaramaraisaoficial', + 'uploader_id': '@maiaramaraisaoficial', 'uploader': 'Maiara e Maraisa', } }] diff --git a/youtube_dl/extractor/vidlii.py b/youtube_dl/extractor/vidlii.py index f4774256b..47f328e87 100644 --- a/youtube_dl/extractor/vidlii.py +++ b/youtube_dl/extractor/vidlii.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor + from ..utils import ( float_or_none, get_element_by_id, @@ -11,6 +12,7 @@ from ..utils import ( strip_or_none, unified_strdate, urljoin, + str_to_int, ) @@ -36,6 +38,26 @@ class VidLiiIE(InfoExtractor): 'tags': ['Vidlii', 'Jan', 'Videogames'], } }, { + # HD + 'url': 'https://www.vidlii.com/watch?v=2Ng8Abj2Fkl', + 'md5': '450e7da379c884788c3a4fa02a3ce1a4', + 'info_dict': { + 'id': '2Ng8Abj2Fkl', + 'ext': 'mp4', + 'title': 'test', + 'description': 'md5:cc55a86032a7b6b3cbfd0f6b155b52e9', + 'thumbnail': 'https://www.vidlii.com/usfi/thmp/2Ng8Abj2Fkl.jpg', + 'uploader': 'VidLii', + 'uploader_url': 'https://www.vidlii.com/user/VidLii', + 'upload_date': '20200927', + 'duration': 5, + 'view_count': int, + 'comment_count': int, + 'average_rating': float, + 'categories': ['Film & Animation'], + 'tags': list, + }, + }, { 'url': 'https://www.vidlii.com/embed?v=tJluaH4BJ3v&a=0', 'only_matching': True, }] @@ -46,11 +68,32 @@ class VidLiiIE(InfoExtractor): webpage = self._download_webpage( 'https://www.vidlii.com/watch?v=%s' % video_id, video_id) - video_url = self._search_regex( - r'src\s*:\s*(["\'])(?P<url>(?:https?://)?(?:(?!\1).)+)\1', webpage, - 'video url', group='url') + formats = [] + + def add_format(format_url, height=None): + height = int(self._search_regex(r'(\d+)\.mp4', + format_url, 'height', default=360)) + + formats.append({ + 'url': format_url, + 'format_id': '%dp' % height if height else None, + 'height': height, + }) + + sources = re.findall( + r'src\s*:\s*(["\'])(?P<url>(?:https?://)?(?:(?!\1).)+)\1', + webpage) + + formats = [] + if len(sources) > 1: + add_format(sources[1][1]) + self._check_formats(formats, video_id) + if len(sources) > 0: + add_format(sources[0][1]) + + self._sort_formats(formats) - title = self._search_regex( + title = self._html_search_regex( (r'<h1>([^<]+)</h1>', r'<title>([^<]+) - VidLii<'), webpage, 'title') @@ -82,9 +125,9 @@ class VidLiiIE(InfoExtractor): default=None) or self._search_regex( r'duration\s*:\s*(\d+)', webpage, 'duration', fatal=False)) - view_count = int_or_none(self._search_regex( - (r'<strong>(\d+)</strong> views', - r'Views\s*:\s*<strong>(\d+)</strong>'), + view_count = str_to_int(self._html_search_regex( + (r'<strong>([\d,.]+)</strong> views', + r'Views\s*:\s*<strong>([\d,.]+)</strong>'), webpage, 'view count', fatal=False)) comment_count = int_or_none(self._search_regex( @@ -109,7 +152,7 @@ class VidLiiIE(InfoExtractor): return { 'id': video_id, - 'url': video_url, + 'formats': formats, 'title': title, 'description': description, 'thumbnail': thumbnail, diff --git a/youtube_dl/jsinterp.py b/youtube_dl/jsinterp.py index 86d902248..02adf6678 100644 --- a/youtube_dl/jsinterp.py +++ b/youtube_dl/jsinterp.py @@ -14,6 +14,7 @@ from .utils import ( remove_quotes, unified_timestamp, variadic, + write_string, ) from .compat import ( compat_basestring, @@ -53,15 +54,16 @@ def wraps_op(op): # NB In principle NaN cannot be checked by membership. # Here all NaN values are actually this one, so _NaN is _NaN, -# although _NaN != _NaN. +# although _NaN != _NaN. Ditto Infinity. _NaN = float('nan') +_Infinity = float('inf') def _js_bit_op(op): def zeroise(x): - return 0 if x in (None, JS_Undefined, _NaN) else x + return 0 if x in (None, JS_Undefined, _NaN, _Infinity) else x @wraps_op(op) def wrapped(a, b): @@ -84,7 +86,7 @@ def _js_arith_op(op): def _js_div(a, b): if JS_Undefined in (a, b) or not (a or b): return _NaN - return operator.truediv(a or 0, b) if b else float('inf') + return operator.truediv(a or 0, b) if b else _Infinity def _js_mod(a, b): @@ -220,6 +222,42 @@ class LocalNameSpace(ChainMap): return 'LocalNameSpace%s' % (self.maps, ) +class Debugger(object): + ENABLED = False + + @staticmethod + def write(*args, **kwargs): + level = kwargs.get('level', 100) + + def truncate_string(s, left, right=0): + if s is None or len(s) <= left + right: + return s + return '...'.join((s[:left - 3], s[-right:] if right else '')) + + write_string('[debug] JS: {0}{1}\n'.format( + ' ' * (100 - level), + ' '.join(truncate_string(compat_str(x), 50, 50) for x in args))) + + @classmethod + def wrap_interpreter(cls, f): + def interpret_statement(self, stmt, local_vars, allow_recursion, *args, **kwargs): + if cls.ENABLED and stmt.strip(): + cls.write(stmt, level=allow_recursion) + try: + ret, should_ret = f(self, stmt, local_vars, allow_recursion, *args, **kwargs) + except Exception as e: + if cls.ENABLED: + if isinstance(e, ExtractorError): + e = e.orig_msg + cls.write('=> Raises:', e, '<-|', stmt, level=allow_recursion) + raise + if cls.ENABLED and stmt.strip(): + if should_ret or not repr(ret) == stmt: + cls.write(['->', '=>'][should_ret], repr(ret), '<-|', stmt, level=allow_recursion) + return ret, should_ret + return interpret_statement + + class JSInterpreter(object): __named_object_counter = 0 @@ -307,8 +345,7 @@ class JSInterpreter(object): def __op_chars(cls): op_chars = set(';,[') for op in cls._all_operators(): - for c in op[0]: - op_chars.add(c) + op_chars.update(op[0]) return op_chars def _named_object(self, namespace, obj): @@ -326,9 +363,8 @@ class JSInterpreter(object): # collections.Counter() is ~10% slower in both 2.7 and 3.9 counters = dict((k, 0) for k in _MATCHING_PARENS.values()) start, splits, pos, delim_len = 0, 0, 0, len(delim) - 1 - in_quote, escaping, skipping = None, False, 0 - after_op, in_regex_char_group = True, False - + in_quote, escaping, after_op, in_regex_char_group = None, False, True, False + skipping = 0 for idx, char in enumerate(expr): paren_delta = 0 if not in_quote: @@ -382,10 +418,12 @@ class JSInterpreter(object): return separated[0][1:].strip(), separated[1].strip() @staticmethod - def _all_operators(): - return itertools.chain( - # Ref: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Operator_Precedence - _SC_OPERATORS, _LOG_OPERATORS, _COMP_OPERATORS, _OPERATORS) + def _all_operators(_cached=[]): + if not _cached: + _cached.extend(itertools.chain( + # Ref: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Operator_Precedence + _SC_OPERATORS, _LOG_OPERATORS, _COMP_OPERATORS, _OPERATORS)) + return _cached def _operator(self, op, left_val, right_expr, expr, local_vars, allow_recursion): if op in ('||', '&&'): @@ -416,7 +454,7 @@ class JSInterpreter(object): except Exception as e: if allow_undefined: return JS_Undefined - raise self.Exception('Cannot get index {idx:.100}'.format(**locals()), expr=repr(obj), cause=e) + raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e) def _dump(self, obj, namespace): try: @@ -438,6 +476,7 @@ class JSInterpreter(object): _FINALLY_RE = re.compile(r'finally\s*\{') _SWITCH_RE = re.compile(r'switch\s*\(') + @Debugger.wrap_interpreter def interpret_statement(self, stmt, local_vars, allow_recursion=100): if allow_recursion < 0: raise self.Exception('Recursion limit reached') @@ -511,7 +550,6 @@ class JSInterpreter(object): expr = self._dump(inner, local_vars) + outer if expr.startswith('('): - m = re.match(r'\((?P<d>[a-z])%(?P<e>[a-z])\.length\+(?P=e)\.length\)%(?P=e)\.length', expr) if m: # short-cut eval of frequently used `(d%e.length+e.length)%e.length`, worth ~6% on `pytest -k test_nsig` @@ -693,7 +731,7 @@ class JSInterpreter(object): (?P<op>{_OPERATOR_RE})? =(?!=)(?P<expr>.*)$ )|(?P<return> - (?!if|return|true|false|null|undefined)(?P<name>{_NAME_RE})$ + (?!if|return|true|false|null|undefined|NaN|Infinity)(?P<name>{_NAME_RE})$ )|(?P<indexing> (?P<in>{_NAME_RE})\[(?P<idx>.+)\]$ )|(?P<attribute> @@ -727,11 +765,12 @@ class JSInterpreter(object): raise JS_Break() elif expr == 'continue': raise JS_Continue() - elif expr == 'undefined': return JS_Undefined, should_return elif expr == 'NaN': return _NaN, should_return + elif expr == 'Infinity': + return _Infinity, should_return elif md.get('return'): return local_vars[m.group('name')], should_return @@ -760,18 +799,28 @@ class JSInterpreter(object): right_expr = separated.pop() # handle operators that are both unary and binary, minimal BODMAS if op in ('+', '-'): + # simplify/adjust consecutive instances of these operators undone = 0 while len(separated) > 1 and not separated[-1].strip(): undone += 1 separated.pop() if op == '-' and undone % 2 != 0: right_expr = op + right_expr + elif op == '+': + while len(separated) > 1 and separated[-1].strip() in self.OP_CHARS: + right_expr = separated.pop() + right_expr + # hanging op at end of left => unary + (strip) or - (push right) left_val = separated[-1] for dm_op in ('*', '%', '/', '**'): bodmas = tuple(self._separate(left_val, dm_op, skip_delims=skip_delim)) if len(bodmas) > 1 and not bodmas[-1].strip(): expr = op.join(separated) + op + right_expr - right_expr = None + if len(separated) > 1: + separated.pop() + right_expr = op.join((left_val, right_expr)) + else: + separated = [op.join((left_val, right_expr))] + right_expr = None break if right_expr is None: continue @@ -797,6 +846,8 @@ class JSInterpreter(object): def eval_method(): if (variable, member) == ('console', 'debug'): + if Debugger.ENABLED: + Debugger.write(self.interpret_expression('[{}]'.format(arg_str), local_vars, allow_recursion)) return types = { 'String': compat_str, diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index cd4303566..113c913df 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -2406,7 +2406,7 @@ class ExtractorError(YoutubeDLError): """ tb, if given, is the original traceback (so that it can be printed out). If expected is set, this is a normal error message and most likely not a bug in youtube-dl. """ - + self.orig_msg = msg if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError): expected = True if video_id is not None: |