From c96eca426b6b4c963fafd3f3268ea1f39b3e2857 Mon Sep 17 00:00:00 2001 From: Philip Huppert Date: Mon, 5 Oct 2015 00:41:20 +0200 Subject: [PATCH 1/5] [mixcloud] Added support for user uploads, playlists, favorites and listens. Fixes #3750 and #5272 --- youtube_dl/extractor/extractors.py | 6 +- youtube_dl/extractor/mixcloud.py | 200 ++++++++++++++++++++++++++++- 2 files changed, 203 insertions(+), 3 deletions(-) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 04c6508f19..b06b717f57 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -411,7 +411,11 @@ from .miomio import MioMioIE from .mit import TechTVMITIE, MITIE, OCWMITIE from .mitele import MiTeleIE -from .mixcloud import MixcloudIE +from .mixcloud import ( + MixcloudIE, + MixcloudUserIE, + MixcloudPlaylistIE +) from .mlb import MLBIE from .mnet import MnetIE from .mpora import MporaIE diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 9638cc9e6d..20e64bab5e 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -3,18 +3,22 @@ import re from .common import InfoExtractor -from ..compat import compat_urllib_parse_unquote +from ..compat import ( + compat_urllib_parse_unquote, + compat_urllib_request +) from ..utils import ( ExtractorError, HEADRequest, NO_DEFAULT, parse_count, str_to_int, + clean_html ) class MixcloudIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/([^/]+)' + _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)' IE_NAME = 'mixcloud' _TESTS = [{ @@ -115,3 +119,195 @@ def _real_extract(self, url): 'view_count': view_count, 'like_count': like_count, } + + +class MixcloudUserIE(InfoExtractor): + """ + Information extractor for Mixcloud users. + It can retrieve a list of a user's uploads, favorites or listens. + """ + + _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P[^/]+)/(?Puploads|favorites|listens)?/?$' + IE_NAME = 'mixcloud:user' + + _TESTS = [{ + 'url': 'http://www.mixcloud.com/dholbach/', + 'info_dict': { + 'id': 'dholbach/uploads', + 'title': 'Daniel Holbach (uploads)', + 'description': 'md5:327af72d1efeb404a8216c27240d1370', + }, + 'playlist_mincount': 11 + }, { + 'url': 'http://www.mixcloud.com/dholbach/uploads/', + 'info_dict': { + 'id': 'dholbach/uploads', + 'title': 'Daniel Holbach (uploads)', + 'description': 'md5:327af72d1efeb404a8216c27240d1370', + }, + 'playlist_mincount': 11 + }, { + 'url': 'http://www.mixcloud.com/dholbach/favorites/', + 'info_dict': { + 'id': 'dholbach/favorites', + 'title': 'Daniel Holbach (favorites)', + 'description': 'md5:327af72d1efeb404a8216c27240d1370', + }, + 'playlist_mincount': 244 + }, { + 'url': 'http://www.mixcloud.com/dholbach/listens/', + 'info_dict': { + 'id': 'dholbach/listens', + 'title': 'Daniel Holbach (listens)', + 'description': 'md5:327af72d1efeb404a8216c27240d1370', + }, + 'playlist_mincount': 846 + }] + + def _fetch_tracks(self, base_url, video_id, dl_note=None, dl_errnote=None): + # retrieve all fragments of a list of tracks with fake AJAX calls + track_urls = [] + current_page = 1 + while True: + # fake a AJAX request to retrieve a list fragment + page_url = base_url + "?page=%d&list=main&_ajax=1" % current_page + req = compat_urllib_request.Request(page_url, headers={"X-Requested-With": "XMLHttpRequest"}, method="GET") + resp = self._download_webpage(req, video_id, note=dl_note + " (page %d)" % current_page, errnote=dl_errnote) + + # extract all track URLs from fragment + urls = re.findall(r'm-play-button m-url="(?P[^"]+)"', resp) + # clean up URLs + urls = map(clean_html, urls) + # create absolute URLs + urls = map(lambda u: "https://www.mixcloud.com" + u, urls) + track_urls.extend(urls) + + # advance to next fragment, if any + if " m-next-page-url=" in resp: + current_page += 1 + else: + break + + return track_urls + + def _handle_track_urls(self, urls): + return map(lambda u: self.url_result(u, "Mixcloud"), urls) + + def _get_user_description(self, page_content): + return self._html_search_regex( + r'
.*?

(?P.*?)

', + page_content, + "user description", + group="description", + fatal=False, + default="") + + def _get_username(self, page_content): + return self._og_search_title(page_content) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user_id = mobj.group("user") + list_type = mobj.group("type") + + # if only a profile URL was supplied, default to download all uploads + if list_type is None: + list_type = "uploads" + + video_id = "%s/%s" % (user_id, list_type) + + # download the user's profile to retrieve some metadata + profile = self._download_webpage("https://www.mixcloud.com/%s/" % user_id, + video_id, + note="Downloading user profile", + errnote="Unable to download user profile") + + username = self._get_username(profile) + description = self._get_user_description(profile) + + # retrieve all page fragments of uploads, favorites or listens + track_urls = self._fetch_tracks( + "https://www.mixcloud.com/%s/%s/" % (user_id, list_type), + video_id, + dl_note="Downloading list of %s" % list_type, + dl_errnote="Unable to download list of %s" % list_type) + + # let MixcloudIE handle each track URL + entries = self._handle_track_urls(track_urls) + + return { + '_type': 'playlist', + 'entries': entries, + 'title': "%s (%s)" % (username, list_type), + 'id': video_id, + "description": description + } + + +class MixcloudPlaylistIE(MixcloudUserIE): + """ + Information extractor for Mixcloud playlists. + """ + + _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P[^/]+)/playlists/(?P[^/]+)/?$' + IE_NAME = 'mixcloud:playlist' + + _TESTS = [{ + 'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/', + 'info_dict': { + 'id': 'RedBullThre3style/playlists/tokyo-finalists-2015', + 'title': 'National Champions 2015', + 'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3', + }, + 'playlist_mincount': 16 + }, { + 'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/', + 'info_dict': { + 'id': 'maxvibes/playlists/jazzcat-on-ness-radio', + 'title': 'Jazzcat on Ness Radio', + 'description': 'md5:c2c51a1f1b8bb5442f2ca67c3dc4af27', + }, + 'playlist_mincount': 23 + }] + + def _get_playlist_title(self, page_content): + return self._html_search_regex( + r'(?P.*?)</span>', + page_content, + "playlist title", + group="title", + fatal=True + ) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user_id = mobj.group("user") + playlist_id = mobj.group("playlist") + video_id = "%s/playlists/%s" % (user_id, playlist_id) + + # download the playlist page to retrieve some metadata + profile = self._download_webpage(url, + user_id, + note="Downloading playlist page", + errnote="Unable to download playlist page") + + description = self._get_user_description(profile) + playlist_title = self._get_playlist_title(profile) + + # retrieve all page fragments of playlist + track_urls = self._fetch_tracks( + "https://www.mixcloud.com/%s/playlists/%s/" % (user_id, playlist_id), + video_id, + dl_note="Downloading tracklist of %s" % playlist_title, + dl_errnote="Unable to tracklist of %s" % playlist_title) + + # let MixcloudIE handle each track + entries = self._handle_track_urls(track_urls) + + return { + '_type': 'playlist', + 'entries': entries, + 'title': playlist_title, + 'id': video_id, + "description": description + } From f896e1ccefc5d946c83f91bf517801882f4184dc Mon Sep 17 00:00:00 2001 From: Philip Huppert <philip@zeilen-sprung.de> Date: Sat, 9 Apr 2016 14:30:01 +0200 Subject: [PATCH 2/5] [mixcloud] fixed some tests --- youtube_dl/extractor/mixcloud.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 20e64bab5e..dcc4ddf25f 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -25,7 +25,7 @@ class MixcloudIE(InfoExtractor): 'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/', 'info_dict': { 'id': 'dholbach-cryptkeeper', - 'ext': 'mp3', + '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', @@ -265,7 +265,7 @@ class MixcloudPlaylistIE(MixcloudUserIE): 'info_dict': { 'id': 'maxvibes/playlists/jazzcat-on-ness-radio', 'title': 'Jazzcat on Ness Radio', - 'description': 'md5:c2c51a1f1b8bb5442f2ca67c3dc4af27', + 'description': 'md5:7bbbf0d6359a0b8cda85224be0f8f263', }, 'playlist_mincount': 23 }] From dcaf00fb3eb716ea146a8e8870c7753b9acf67b8 Mon Sep 17 00:00:00 2001 From: Philip Huppert <philip@zeilen-sprung.de> Date: Sat, 9 Apr 2016 22:20:16 +0200 Subject: [PATCH 3/5] [mixcloud] support older urllib versions --- youtube_dl/extractor/mixcloud.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index dcc4ddf25f..769b68dc36 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -171,7 +171,7 @@ def _fetch_tracks(self, base_url, video_id, dl_note=None, dl_errnote=None): while True: # fake a AJAX request to retrieve a list fragment page_url = base_url + "?page=%d&list=main&_ajax=1" % current_page - req = compat_urllib_request.Request(page_url, headers={"X-Requested-With": "XMLHttpRequest"}, method="GET") + req = compat_urllib_request.Request(page_url, headers={"X-Requested-With": "XMLHttpRequest"}) resp = self._download_webpage(req, video_id, note=dl_note + " (page %d)" % current_page, errnote=dl_errnote) # extract all track URLs from fragment From 6d671695092236af75c49e8a74d19b771e1b4d79 Mon Sep 17 00:00:00 2001 From: Philip Huppert <philip@zeilen-sprung.de> Date: Sun, 10 Apr 2016 15:53:17 +0200 Subject: [PATCH 4/5] [mixcloud] improved extraction of user description --- youtube_dl/extractor/mixcloud.py | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 769b68dc36..30b33e7e98 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -195,12 +195,10 @@ def _handle_track_urls(self, urls): def _get_user_description(self, page_content): return self._html_search_regex( - r'<div class="description-text">.*?<p>(?P<description>.*?)</p></div></div></div>', + r'<div class="description-text">.*?<p>(.*?)</p></div></div></div>', page_content, "user description", - group="description", - fatal=False, - default="") + fatal=False) def _get_username(self, page_content): return self._og_search_title(page_content) From 9c250931f5e1e68a835065c0cc5fa58e3f1e4734 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan <yan12125@gmail.com> Date: Thu, 14 Apr 2016 22:19:52 +0800 Subject: [PATCH 5/5] [mixcloud] Improve and simplify mixcloud:user and mixcloud:playlist --- youtube_dl/extractor/mixcloud.py | 198 ++++++++++++------------------- 1 file changed, 76 insertions(+), 122 deletions(-) diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 30b33e7e98..5cf42198d5 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -1,19 +1,21 @@ from __future__ import unicode_literals +import functools import re from .common import InfoExtractor from ..compat import ( compat_urllib_parse_unquote, - compat_urllib_request + compat_urlparse, ) from ..utils import ( + clean_html, ExtractorError, HEADRequest, + OnDemandPagedList, NO_DEFAULT, parse_count, str_to_int, - clean_html ) @@ -121,191 +123,143 @@ def _real_extract(self, url): } -class MixcloudUserIE(InfoExtractor): - """ - Information extractor for Mixcloud users. - It can retrieve a list of a user's uploads, favorites or listens. - """ +class MixcloudPlaylistBaseIE(InfoExtractor): + _PAGE_SIZE = 24 + def _fetch_tracks_page(self, path, video_id, page_name, current_page): + resp = self._download_webpage( + 'https://www.mixcloud.com/%s/' % path, video_id, + note='Download %s (page %d)' % (page_name, current_page + 1), + errnote='Unable to download %s' % page_name, + query={'page': (current_page + 1), 'list': 'main', '_ajax': '1'}, + headers={'X-Requested-With': 'XMLHttpRequest'}) + + for url in re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', resp): + yield self.url_result( + compat_urlparse.urljoin('https://www.mixcloud.com', clean_html(url)), + MixcloudIE.ie_key()) + + def _get_user_description(self, page_content): + return self._html_search_regex( + r'<div[^>]+class="description-text"[^>]*>(.+?)</div>', + page_content, 'user description', fatal=False) + + +class MixcloudUserIE(MixcloudPlaylistBaseIE): _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$' IE_NAME = 'mixcloud:user' _TESTS = [{ 'url': 'http://www.mixcloud.com/dholbach/', 'info_dict': { - 'id': 'dholbach/uploads', + 'id': 'dholbach_uploads', 'title': 'Daniel Holbach (uploads)', 'description': 'md5:327af72d1efeb404a8216c27240d1370', }, - 'playlist_mincount': 11 + 'playlist_mincount': 11, }, { 'url': 'http://www.mixcloud.com/dholbach/uploads/', 'info_dict': { - 'id': 'dholbach/uploads', + 'id': 'dholbach_uploads', 'title': 'Daniel Holbach (uploads)', 'description': 'md5:327af72d1efeb404a8216c27240d1370', }, - 'playlist_mincount': 11 + 'playlist_mincount': 11, }, { 'url': 'http://www.mixcloud.com/dholbach/favorites/', 'info_dict': { - 'id': 'dholbach/favorites', + 'id': 'dholbach_favorites', 'title': 'Daniel Holbach (favorites)', 'description': 'md5:327af72d1efeb404a8216c27240d1370', }, - 'playlist_mincount': 244 + 'params': { + 'playlist_items': '1-100', + }, + 'playlist_mincount': 100, }, { 'url': 'http://www.mixcloud.com/dholbach/listens/', 'info_dict': { - 'id': 'dholbach/listens', + 'id': 'dholbach_listens', 'title': 'Daniel Holbach (listens)', 'description': 'md5:327af72d1efeb404a8216c27240d1370', }, - 'playlist_mincount': 846 + 'params': { + 'playlist_items': '1-100', + }, + 'playlist_mincount': 100, }] - def _fetch_tracks(self, base_url, video_id, dl_note=None, dl_errnote=None): - # retrieve all fragments of a list of tracks with fake AJAX calls - track_urls = [] - current_page = 1 - while True: - # fake a AJAX request to retrieve a list fragment - page_url = base_url + "?page=%d&list=main&_ajax=1" % current_page - req = compat_urllib_request.Request(page_url, headers={"X-Requested-With": "XMLHttpRequest"}) - resp = self._download_webpage(req, video_id, note=dl_note + " (page %d)" % current_page, errnote=dl_errnote) - - # extract all track URLs from fragment - urls = re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', resp) - # clean up URLs - urls = map(clean_html, urls) - # create absolute URLs - urls = map(lambda u: "https://www.mixcloud.com" + u, urls) - track_urls.extend(urls) - - # advance to next fragment, if any - if " m-next-page-url=" in resp: - current_page += 1 - else: - break - - return track_urls - - def _handle_track_urls(self, urls): - return map(lambda u: self.url_result(u, "Mixcloud"), urls) - - def _get_user_description(self, page_content): - return self._html_search_regex( - r'<div class="description-text">.*?<p>(.*?)</p></div></div></div>', - page_content, - "user description", - fatal=False) - - def _get_username(self, page_content): - return self._og_search_title(page_content) - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - user_id = mobj.group("user") - list_type = mobj.group("type") + user_id = mobj.group('user') + list_type = mobj.group('type') # if only a profile URL was supplied, default to download all uploads if list_type is None: - list_type = "uploads" + list_type = 'uploads' - video_id = "%s/%s" % (user_id, list_type) + video_id = '%s_%s' % (user_id, list_type) - # download the user's profile to retrieve some metadata - profile = self._download_webpage("https://www.mixcloud.com/%s/" % user_id, - video_id, - note="Downloading user profile", - errnote="Unable to download user profile") + profile = self._download_webpage( + 'https://www.mixcloud.com/%s/' % user_id, video_id, + note='Downloading user profile', + errnote='Unable to download user profile') - username = self._get_username(profile) + username = self._og_search_title(profile) description = self._get_user_description(profile) - # retrieve all page fragments of uploads, favorites or listens - track_urls = self._fetch_tracks( - "https://www.mixcloud.com/%s/%s/" % (user_id, list_type), - video_id, - dl_note="Downloading list of %s" % list_type, - dl_errnote="Unable to download list of %s" % list_type) + entries = OnDemandPagedList( + functools.partial( + self._fetch_tracks_page, + '%s/%s' % (user_id, list_type), video_id, 'list of %s' % list_type), + self._PAGE_SIZE, use_cache=True) - # let MixcloudIE handle each track URL - entries = self._handle_track_urls(track_urls) - - return { - '_type': 'playlist', - 'entries': entries, - 'title': "%s (%s)" % (username, list_type), - 'id': video_id, - "description": description - } + return self.playlist_result( + entries, video_id, '%s (%s)' % (username, list_type), description) -class MixcloudPlaylistIE(MixcloudUserIE): - """ - Information extractor for Mixcloud playlists. - """ - +class MixcloudPlaylistIE(MixcloudPlaylistBaseIE): _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$' IE_NAME = 'mixcloud:playlist' _TESTS = [{ 'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/', 'info_dict': { - 'id': 'RedBullThre3style/playlists/tokyo-finalists-2015', + 'id': 'RedBullThre3style_tokyo-finalists-2015', 'title': 'National Champions 2015', 'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3', }, - 'playlist_mincount': 16 + 'playlist_mincount': 16, }, { 'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/', 'info_dict': { - 'id': 'maxvibes/playlists/jazzcat-on-ness-radio', + 'id': 'maxvibes_jazzcat-on-ness-radio', 'title': 'Jazzcat on Ness Radio', 'description': 'md5:7bbbf0d6359a0b8cda85224be0f8f263', }, 'playlist_mincount': 23 }] - def _get_playlist_title(self, page_content): - return self._html_search_regex( - r'<span class="main-list-title list-playlist-title ">(?P<title>.*?)</span>', - page_content, - "playlist title", - group="title", - fatal=True - ) - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - user_id = mobj.group("user") - playlist_id = mobj.group("playlist") - video_id = "%s/playlists/%s" % (user_id, playlist_id) + user_id = mobj.group('user') + playlist_id = mobj.group('playlist') + video_id = '%s_%s' % (user_id, playlist_id) - # download the playlist page to retrieve some metadata - profile = self._download_webpage(url, - user_id, - note="Downloading playlist page", - errnote="Unable to download playlist page") + profile = self._download_webpage( + url, user_id, + note='Downloading playlist page', + errnote='Unable to download playlist page') description = self._get_user_description(profile) - playlist_title = self._get_playlist_title(profile) + playlist_title = self._html_search_regex( + r'<span[^>]+class="[^"]*list-playlist-title[^"]*"[^>]*>(.*?)</span>', + profile, 'playlist title') - # retrieve all page fragments of playlist - track_urls = self._fetch_tracks( - "https://www.mixcloud.com/%s/playlists/%s/" % (user_id, playlist_id), - video_id, - dl_note="Downloading tracklist of %s" % playlist_title, - dl_errnote="Unable to tracklist of %s" % playlist_title) + entries = OnDemandPagedList( + functools.partial( + self._fetch_tracks_page, + '%s/playlists/%s' % (user_id, playlist_id), video_id, 'tracklist'), + self._PAGE_SIZE) - # let MixcloudIE handle each track - entries = self._handle_track_urls(track_urls) - - return { - '_type': 'playlist', - 'entries': entries, - 'title': playlist_title, - 'id': video_id, - "description": description - } + return self.playlist_result(entries, video_id, playlist_title, description)