4 from .common import InfoExtractor
12 class BandcampIE(InfoExtractor):
13 _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)'
15 u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
16 u'file': u'1812978515.mp3',
17 u'md5': u'c557841d5e50261777a6585648adf439',
19 u"title": u"youtube-dl \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
22 u'skip': u'There is a limit of 200 free downloads / month for the test song'
25 def _real_extract(self, url):
26 mobj = re.match(self._VALID_URL, url)
27 title = mobj.group('title')
28 webpage = self._download_webpage(url, title)
29 # We get the link to the free download page
30 m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
31 if m_download is None:
32 m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
34 json_code = m_trackinfo.group(1)
35 data = json.loads(json_code)
38 duration = int(round(d['duration']))
40 for format_id, format_url in d['file'].items():
41 ext, _, abr_str = format_id.partition('-')
44 'format_id': format_id,
46 'ext': format_id.partition('-')[0],
48 'acodec': format_id.partition('-')[0],
49 'abr': int(format_id.partition('-')[2]),
52 self._sort_formats(formats)
55 'id': compat_str(d['id']),
61 raise ExtractorError(u'No free songs found')
63 download_link = m_download.group(1)
65 r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
66 webpage, re.MULTILINE | re.DOTALL).group('id')
68 download_webpage = self._download_webpage(download_link, video_id,
69 'Downloading free downloads page')
70 # We get the dictionary of the track from some javascrip code
71 info = re.search(r'items: (.*?),$',
72 download_webpage, re.MULTILINE).group(1)
73 info = json.loads(info)[0]
74 # We pick mp3-320 for now, until format selection can be easily implemented.
75 mp3_info = info[u'downloads'][u'mp3-320']
76 # If we try to use this url it says the link has expired
77 initial_url = mp3_info[u'url']
78 re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
79 m_url = re.match(re_url, initial_url)
80 #We build the url we will use to get the final track url
81 # This url is build in Bandcamp in the script download_bunde_*.js
82 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'))
83 final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
84 # If we could correctly generate the .rand field the url would be
85 #in the "download_url" key
86 final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
90 'title': info[u'title'],
94 'thumbnail': info[u'thumb_url'],
95 'uploader': info[u'artist'],
99 class BandcampAlbumIE(InfoExtractor):
100 IE_NAME = u'Bandcamp:album'
101 _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)'
104 u'url': u'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
107 u'file': u'1353101989.mp3',
108 u'md5': u'39bc1eded3476e927c724321ddf116cf',
114 u'file': u'38097443.mp3',
115 u'md5': u'1a2c32e2691474643e912cc6cd4bffaa',
117 u'title': u'Kero One - Keep It Alive (Blazo remix)',
124 u'skip': u'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
127 def _real_extract(self, url):
128 mobj = re.match(self._VALID_URL, url)
129 title = mobj.group('title')
130 webpage = self._download_webpage(url, title)
131 tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
133 raise ExtractorError(u'The page doesn\'t contain any tracks')
135 self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
136 for t_path in tracks_paths]
137 title = self._search_regex(r'album_title : "(.*?)"', webpage, u'title')