1 from __future__ import unicode_literals
6 from .common import InfoExtractor
14 class BandcampIE(InfoExtractor):
15 _VALID_URL = r'https?://.*?\.bandcamp\.com/track/(?P<title>.*)'
17 'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
18 'md5': 'c557841d5e50261777a6585648adf439',
22 'title': "youtube-dl \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
25 '_skip': 'There is a limit of 200 free downloads / month for the test song'
27 'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
28 'md5': '2b68e5851514c20efdff2afc5603b8b4',
32 'title': 'Lanius (Battle)',
33 'uploader': 'Ben Prunty Music',
37 def _real_extract(self, url):
38 mobj = re.match(self._VALID_URL, url)
39 title = mobj.group('title')
40 webpage = self._download_webpage(url, title)
41 m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
43 m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
45 json_code = m_trackinfo.group(1)
46 data = json.loads(json_code)[0]
49 for format_id, format_url in data['file'].items():
50 ext, abr_str = format_id.split('-', 1)
52 'format_id': format_id,
60 self._sort_formats(formats)
63 'id': compat_str(data['id']),
64 'title': data['title'],
66 'duration': float(data['duration']),
69 raise ExtractorError('No free songs found')
71 download_link = m_download.group(1)
72 video_id = self._search_regex(
73 r'var TralbumData = {.*?id: (?P<id>\d+),?$',
74 webpage, 'video id', flags=re.MULTILINE | re.DOTALL)
76 download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page')
77 # We get the dictionary of the track from some javascript code
78 info = re.search(r'items: (.*?),$', download_webpage, re.MULTILINE).group(1)
79 info = json.loads(info)[0]
80 # We pick mp3-320 for now, until format selection can be easily implemented.
81 mp3_info = info['downloads']['mp3-320']
82 # If we try to use this url it says the link has expired
83 initial_url = mp3_info['url']
84 re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
85 m_url = re.match(re_url, initial_url)
86 # We build the url we will use to get the final track url
87 # This url is build in Bandcamp in the script download_bunde_*.js
88 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'))
89 final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
90 # If we could correctly generate the .rand field the url would be
91 # in the "download_url" key
92 final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
96 'title': info['title'],
100 'thumbnail': info.get('thumb_url'),
101 'uploader': info.get('artist'),
105 class BandcampAlbumIE(InfoExtractor):
106 IE_NAME = 'Bandcamp:album'
107 _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<title>[^?#]+))'
110 'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
113 'md5': '39bc1eded3476e927c724321ddf116cf',
121 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
125 'title': 'Kero One - Keep It Alive (Blazo remix)',
130 'title': 'Jazz Format Mixtape vol.1',
135 'skip': 'Bandcamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
137 'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
139 'title': 'Hierophany of the Open Grave',
141 'playlist_mincount': 9,
144 def _real_extract(self, url):
145 mobj = re.match(self._VALID_URL, url)
146 playlist_id = mobj.group('subdomain')
147 title = mobj.group('title')
148 display_id = title or playlist_id
149 webpage = self._download_webpage(url, display_id)
150 tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
152 raise ExtractorError('The page doesn\'t contain any tracks')
154 self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
155 for t_path in tracks_paths]
156 title = self._search_regex(r'album_title : "(.*?)"', webpage, 'title')
160 'display_id': display_id,