X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/b238854ce845f3796daac74edab2e8a373e8ba1a..4d287e7783d80aaa7ad181bff64c48b595d7ec05:/youtube_dl/extractor/bandcamp.py diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index 3a32c14..8692949 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -1,129 +1,181 @@ +from __future__ import unicode_literals + import json import re from .common import InfoExtractor -from ..utils import ( +from ..compat import ( compat_str, compat_urlparse, +) +from ..utils import ( ExtractorError, ) class BandcampIE(InfoExtractor): - IE_NAME = u'Bandcamp' - _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P.*)' + _VALID_URL = r'https?://.*?\.bandcamp\.com/track/(?P<title>.*)' _TESTS = [{ - u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song', - u'file': u'1812978515.mp3', - u'md5': u'cdeb30cdae1921719a3cbcab696ef53c', - u'info_dict': { - u"title": u"youtube-dl test song \"'/\\\u00e4\u21ad" + 'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song', + 'md5': 'c557841d5e50261777a6585648adf439', + 'info_dict': { + 'id': '1812978515', + 'ext': 'mp3', + 'title': "youtube-dl \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad", + 'duration': 9.8485, + }, + '_skip': 'There is a limit of 200 free downloads / month for the test song' + }, { + 'url': 'http://benprunty.bandcamp.com/track/lanius-battle', + 'md5': '2b68e5851514c20efdff2afc5603b8b4', + 'info_dict': { + 'id': '2650410135', + 'ext': 'mp3', + 'title': 'Lanius (Battle)', + 'uploader': 'Ben Prunty Music', }, - u'skip': u'There is a limit of 200 free downloads / month for the test song' }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) title = mobj.group('title') webpage = self._download_webpage(url, title) - # We get the link to the free download page m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage) - if m_download is None: + if not m_download: m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage) - if m_trackinfo: - json_code = m_trackinfo.group(1) - data = json.loads(json_code) + if m_trackinfo: + json_code = m_trackinfo.group(1) + data = json.loads(json_code)[0] + + formats = [] + for format_id, format_url in data['file'].items(): + ext, abr_str = format_id.split('-', 1) + formats.append({ + 'format_id': format_id, + 'url': format_url, + 'ext': ext, + 'vcodec': 'none', + 'acodec': ext, + 'abr': int(abr_str), + }) + + self._sort_formats(formats) - for d in data: - formats = [{ - 'format_id': 'format_id', - 'url': format_url, - 'ext': format_id.partition('-')[0] - } for format_id, format_url in sorted(d['file'].items())] return { - 'id': compat_str(d['id']), - 'title': d['title'], + 'id': compat_str(data['id']), + 'title': data['title'], 'formats': formats, + 'duration': float(data['duration']), } - else: - raise ExtractorError(u'No free songs found') + else: + raise ExtractorError('No free songs found') download_link = m_download.group(1) - id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$', - webpage, re.MULTILINE|re.DOTALL).group('id') + video_id = self._search_regex( + r'(?ms)var TralbumData = {.*?id: (?P<id>\d+),?$', + webpage, 'video id') - download_webpage = self._download_webpage(download_link, id, - 'Downloading free downloads page') - # We get the dictionary of the track from some javascrip code - info = re.search(r'items: (.*?),$', - download_webpage, re.MULTILINE).group(1) - info = json.loads(info)[0] + download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page') + # We get the dictionary of the track from some javascript code + all_info = self._parse_json(self._search_regex( + r'(?sm)items: (.*?),$', download_webpage, 'items'), video_id) + info = all_info[0] # We pick mp3-320 for now, until format selection can be easily implemented. - mp3_info = info[u'downloads'][u'mp3-320'] + mp3_info = info['downloads']['mp3-320'] # If we try to use this url it says the link has expired - initial_url = mp3_info[u'url'] - re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$' - m_url = re.match(re_url, initial_url) - #We build the url we will use to get the final track url + initial_url = mp3_info['url'] + m_url = re.match( + r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$', + initial_url) + # We build the url we will use to get the final track url # This url is build in Bandcamp in the script download_bunde_*.js - request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), id, m_url.group('ts')) - final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url') + request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts')) + final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url') # If we could correctly generate the .rand field the url would be - #in the "download_url" key - final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1) + # in the "download_url" key + final_url = self._search_regex( + r'"retry_url":"(.*?)"', final_url_webpage, 'final video URL') - track_info = {'id':id, - 'title' : info[u'title'], - 'ext' : 'mp3', - 'url' : final_url, - 'thumbnail' : info[u'thumb_url'], - 'uploader' : info[u'artist'] - } - - return [track_info] + return { + 'id': video_id, + 'title': info['title'], + 'ext': 'mp3', + 'vcodec': 'none', + 'url': final_url, + 'thumbnail': info.get('thumb_url'), + 'uploader': info.get('artist'), + } class BandcampAlbumIE(InfoExtractor): - IE_NAME = u'Bandcamp:album' - _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)' + IE_NAME = 'Bandcamp:album' + _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^?#]+)|/?(?:$|[?#]))' - _TEST = { - u'url': u'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1', - u'playlist': [ + _TESTS = [{ + 'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1', + 'playlist': [ { - u'file': u'1353101989.mp3', - u'md5': u'39bc1eded3476e927c724321ddf116cf', - u'info_dict': { - u'title': u'Intro', + 'md5': '39bc1eded3476e927c724321ddf116cf', + 'info_dict': { + 'id': '1353101989', + 'ext': 'mp3', + 'title': 'Intro', } }, { - u'file': u'38097443.mp3', - u'md5': u'1a2c32e2691474643e912cc6cd4bffaa', - u'info_dict': { - u'title': u'Kero One - Keep It Alive (Blazo remix)', + 'md5': '1a2c32e2691474643e912cc6cd4bffaa', + 'info_dict': { + 'id': '38097443', + 'ext': 'mp3', + 'title': 'Kero One - Keep It Alive (Blazo remix)', } }, ], - u'params': { - u'playlistend': 2 + 'info_dict': { + 'title': 'Jazz Format Mixtape vol.1', + 'id': 'jazz-format-mixtape-vol-1', + 'uploader_id': 'blazo', + }, + 'params': { + 'playlistend': 2 + }, + 'skip': 'Bandcamp imposes download limits.' + }, { + 'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave', + 'info_dict': { + 'title': 'Hierophany of the Open Grave', + 'uploader_id': 'nightbringer', + 'id': 'hierophany-of-the-open-grave', }, - u'skip': u'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test' - } + 'playlist_mincount': 9, + }, { + 'url': 'http://dotscale.bandcamp.com', + 'info_dict': { + 'title': 'Loom', + 'id': 'dotscale', + 'uploader_id': 'dotscale', + }, + 'playlist_mincount': 7, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - title = mobj.group('title') - webpage = self._download_webpage(url, title) + uploader_id = mobj.group('subdomain') + album_id = mobj.group('album_id') + playlist_id = album_id or uploader_id + webpage = self._download_webpage(url, playlist_id) tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage) if not tracks_paths: - raise ExtractorError(u'The page doesn\'t contain any track') + raise ExtractorError('The page doesn\'t contain any tracks') entries = [ self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key()) for t_path in tracks_paths] - title = self._search_regex(r'album_title : "(.*?)"', webpage, u'title') + title = self._search_regex( + r'album_title\s*:\s*"(.*?)"', webpage, 'title', fatal=False) return { '_type': 'playlist', + 'uploader_id': uploader_id, + 'id': playlist_id, 'title': title, 'entries': entries, }