X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/e76f531201cd41dfc0ce00be28bcc5c575c7acc5..3f49721bd802c357ee9e5c1b6f07e0b68ac47fc2:/youtube_dl/extractor/soundcloud.py?ds=inline diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index 2b60d35..5a201ea 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -1,14 +1,17 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals import re import itertools -from .common import InfoExtractor +from .common import ( + InfoExtractor, + SearchInfoExtractor +) from ..compat import ( compat_str, compat_urlparse, - compat_urllib_parse, + compat_urllib_parse_urlencode, ) from ..utils import ( ExtractorError, @@ -29,7 +32,7 @@ class SoundcloudIE(InfoExtractor): _VALID_URL = r'''(?x)^(?:https?://)? (?:(?:(?:www\.|m\.)?soundcloud\.com/ (?P[\w\d-]+)/ - (?!(?:tracks|sets(?:/[^/?#]+)?|reposts|likes|spotlight)/?(?:$|[?#])) + (?!(?:tracks|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#])) (?P[\w\d-]+)/? (?P<token>[^?]+?)?(?:[?].*)?$) |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+) @@ -50,6 +53,7 @@ class SoundcloudIE(InfoExtractor): 'uploader': 'E.T. ExTerrestrial Music', 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1', 'duration': 143, + 'license': 'all-rights-reserved', } }, # not streamable song @@ -63,6 +67,7 @@ class SoundcloudIE(InfoExtractor): 'uploader': 'The Royal Concept', 'upload_date': '20120521', 'duration': 227, + 'license': 'all-rights-reserved', }, 'params': { # rtmp @@ -81,6 +86,7 @@ class SoundcloudIE(InfoExtractor): 'description': 'test chars: \"\'/\\ä↭', 'upload_date': '20131209', 'duration': 9, + 'license': 'all-rights-reserved', }, }, # private link (alt format) @@ -95,6 +101,7 @@ class SoundcloudIE(InfoExtractor): 'description': 'test chars: \"\'/\\ä↭', 'upload_date': '20131209', 'duration': 9, + 'license': 'all-rights-reserved', }, }, # downloadable song @@ -109,13 +116,20 @@ class SoundcloudIE(InfoExtractor): 'uploader': 'oddsamples', 'upload_date': '20140109', 'duration': 17, + 'license': 'cc-by-sa', }, }, ] - _CLIENT_ID = '02gUJC0hH2ct1EGOcYXQIzRFU91c72Ea' + _CLIENT_ID = 'fDoItMDbsbZz8dY16ZzARCZmzgHBPotA' _IPHONE_CLIENT_ID = '376f225bf427445fc4bfb6b99b72e0bf' + @staticmethod + def _extract_urls(webpage): + return [m.group('url') for m in re.finditer( + r'<iframe[^>]+src=(["\'])(?P<url>(?:https?://)?(?:w\.)?soundcloud\.com/player.+?)\1', + webpage)] + def report_resolve(self, video_id): """Report information extraction.""" self.to_screen('%s: Resolving id' % video_id) @@ -129,20 +143,20 @@ class SoundcloudIE(InfoExtractor): name = full_title or track_id if quiet: self.report_extraction(name) - - thumbnail = info['artwork_url'] - if thumbnail is not None: + thumbnail = info.get('artwork_url') + if isinstance(thumbnail, compat_str): thumbnail = thumbnail.replace('-large', '-t500x500') ext = 'mp3' result = { 'id': track_id, - 'uploader': info['user']['username'], - 'upload_date': unified_strdate(info['created_at']), + 'uploader': info.get('user', {}).get('username'), + 'upload_date': unified_strdate(info.get('created_at')), 'title': info['title'], - 'description': info['description'], + 'description': info.get('description'), 'thumbnail': thumbnail, 'duration': int_or_none(info.get('duration'), 1000), 'webpage_url': info.get('permalink_url'), + 'license': info.get('license'), } formats = [] if info.get('downloadable', False): @@ -212,13 +226,13 @@ class SoundcloudIE(InfoExtractor): raise ExtractorError('Invalid URL: %s' % url) track_id = mobj.group('track_id') - token = None + if track_id is not None: info_json_url = 'http://api.soundcloud.com/tracks/' + track_id + '.json?client_id=' + self._CLIENT_ID full_title = track_id token = mobj.group('secret_token') if token: - info_json_url += "&secret_token=" + token + info_json_url += '&secret_token=' + token elif mobj.group('player'): query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) real_url = query['url'][0] @@ -246,7 +260,20 @@ class SoundcloudIE(InfoExtractor): return self._extract_info_dict(info, full_title, secret_token=token) -class SoundcloudSetIE(SoundcloudIE): +class SoundcloudPlaylistBaseIE(SoundcloudIE): + @staticmethod + def _extract_id(e): + return compat_str(e['id']) if e.get('id') else None + + def _extract_track_entries(self, tracks): + return [ + self.url_result( + track['permalink_url'], SoundcloudIE.ie_key(), + video_id=self._extract_id(track)) + for track in tracks if track.get('permalink_url')] + + +class SoundcloudSetIE(SoundcloudPlaylistBaseIE): _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[\w\d-]+)(?:/(?P<token>[^?/]+))?' IE_NAME = 'soundcloud:set' _TESTS = [{ @@ -256,6 +283,9 @@ class SoundcloudSetIE(SoundcloudIE): 'title': 'The Royal Concept EP', }, 'playlist_mincount': 6, + }, { + 'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token', + 'only_matching': True, }] def _real_extract(self, url): @@ -282,7 +312,7 @@ class SoundcloudSetIE(SoundcloudIE): msgs = (compat_str(err['error_message']) for err in info['errors']) raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs)) - entries = [self.url_result(track['permalink_url'], 'Soundcloud') for track in info['tracks']] + entries = self._extract_track_entries(info['tracks']) return { '_type': 'playlist', @@ -292,7 +322,7 @@ class SoundcloudSetIE(SoundcloudIE): } -class SoundcloudUserIE(SoundcloudIE): +class SoundcloudUserIE(SoundcloudPlaylistBaseIE): _VALID_URL = r'''(?x) https?:// (?:(?:www|m)\.)?soundcloud\.com/ @@ -309,21 +339,21 @@ class SoundcloudUserIE(SoundcloudIE): 'id': '114582580', 'title': 'The Akashic Chronicler (All)', }, - 'playlist_mincount': 111, + 'playlist_mincount': 74, }, { 'url': 'https://soundcloud.com/the-akashic-chronicler/tracks', 'info_dict': { 'id': '114582580', 'title': 'The Akashic Chronicler (Tracks)', }, - 'playlist_mincount': 50, + 'playlist_mincount': 37, }, { 'url': 'https://soundcloud.com/the-akashic-chronicler/sets', 'info_dict': { 'id': '114582580', 'title': 'The Akashic Chronicler (Playlists)', }, - 'playlist_mincount': 3, + 'playlist_mincount': 2, }, { 'url': 'https://soundcloud.com/the-akashic-chronicler/reposts', 'info_dict': { @@ -342,7 +372,7 @@ class SoundcloudUserIE(SoundcloudIE): 'url': 'https://soundcloud.com/grynpyret/spotlight', 'info_dict': { 'id': '7098329', - 'title': 'Grynpyret (Spotlight)', + 'title': 'GRYNPYRET (Spotlight)', }, 'playlist_mincount': 1, }] @@ -380,47 +410,48 @@ class SoundcloudUserIE(SoundcloudIE): resource = mobj.group('rsrc') or 'all' base_url = self._BASE_URL_MAP[resource] % user['id'] - next_href = None + COMMON_QUERY = { + 'limit': 50, + 'client_id': self._CLIENT_ID, + 'linked_partitioning': '1', + } + + query = COMMON_QUERY.copy() + query['offset'] = 0 + + next_href = base_url + '?' + compat_urllib_parse_urlencode(query) entries = [] for i in itertools.count(): - if not next_href: - data = compat_urllib_parse.urlencode({ - 'offset': i * 50, - 'limit': 50, - 'client_id': self._CLIENT_ID, - 'linked_partitioning': '1', - 'representation': 'speedy', - }) - next_href = base_url + '?' + data - response = self._download_json( next_href, uploader, 'Downloading track page %s' % (i + 1)) collection = response['collection'] - if not collection: - self.to_screen('%s: End page received' % uploader) break def resolve_permalink_url(candidates): for cand in candidates: if isinstance(cand, dict): permalink_url = cand.get('permalink_url') + entry_id = self._extract_id(cand) if permalink_url and permalink_url.startswith('http'): - return permalink_url + return permalink_url, entry_id for e in collection: - permalink_url = resolve_permalink_url((e, e.get('track'), e.get('playlist'))) + permalink_url, entry_id = resolve_permalink_url((e, e.get('track'), e.get('playlist'))) if permalink_url: - entries.append(self.url_result(permalink_url)) + entries.append(self.url_result(permalink_url, video_id=entry_id)) - if 'next_href' in response: - next_href = response['next_href'] - if not next_href: - break - else: - next_href = None + next_href = response.get('next_href') + if not next_href: + break + + parsed_next_href = compat_urlparse.urlparse(response['next_href']) + qs = compat_urlparse.parse_qs(parsed_next_href.query) + qs.update(COMMON_QUERY) + next_href = compat_urlparse.urlunparse( + parsed_next_href._replace(query=compat_urllib_parse_urlencode(qs, True))) return { '_type': 'playlist', @@ -430,7 +461,7 @@ class SoundcloudUserIE(SoundcloudIE): } -class SoundcloudPlaylistIE(SoundcloudIE): +class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE): _VALID_URL = r'https?://api\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$' IE_NAME = 'soundcloud:playlist' _TESTS = [{ @@ -456,11 +487,11 @@ class SoundcloudPlaylistIE(SoundcloudIE): if token: data_dict['secret_token'] = token - data = compat_urllib_parse.urlencode(data_dict) + data = compat_urllib_parse_urlencode(data_dict) data = self._download_json( base_url + data, playlist_id, 'Downloading playlist') - entries = [self.url_result(track['permalink_url'], 'Soundcloud') for track in data['tracks']] + entries = self._extract_track_entries(data['tracks']) return { '_type': 'playlist', @@ -469,3 +500,60 @@ class SoundcloudPlaylistIE(SoundcloudIE): 'description': data.get('description'), 'entries': entries, } + + +class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE): + IE_NAME = 'soundcloud:search' + IE_DESC = 'Soundcloud search' + _MAX_RESULTS = float('inf') + _TESTS = [{ + 'url': 'scsearch15:post-avant jazzcore', + 'info_dict': { + 'title': 'post-avant jazzcore', + }, + 'playlist_count': 15, + }] + + _SEARCH_KEY = 'scsearch' + _MAX_RESULTS_PER_PAGE = 200 + _DEFAULT_RESULTS_PER_PAGE = 50 + _API_V2_BASE = 'https://api-v2.soundcloud.com' + + def _get_collection(self, endpoint, collection_id, **query): + limit = min( + query.get('limit', self._DEFAULT_RESULTS_PER_PAGE), + self._MAX_RESULTS_PER_PAGE) + query['limit'] = limit + query['client_id'] = self._CLIENT_ID + query['linked_partitioning'] = '1' + query['offset'] = 0 + data = compat_urllib_parse_urlencode(query) + next_url = '{0}{1}?{2}'.format(self._API_V2_BASE, endpoint, data) + + collected_results = 0 + + for i in itertools.count(1): + response = self._download_json( + next_url, collection_id, 'Downloading page {0}'.format(i), + 'Unable to download API page') + + collection = response.get('collection', []) + if not collection: + break + + collection = list(filter(bool, collection)) + collected_results += len(collection) + + for item in collection: + yield self.url_result(item['uri'], SoundcloudIE.ie_key()) + + if not collection or collected_results >= limit: + break + + next_url = response.get('next_href') + if not next_url: + break + + def _get_n_results(self, query, n): + tracks = self._get_collection('/search/tracks', query, limit=n, q=query) + return self.playlist_result(tracks, playlist_title=query)