[bandcamp] Add support for multiple formats

This commit is contained in:
Philipp Hagemeister 2013-12-26 14:08:57 +01:00
parent 608d11f515
commit 5ecd3c6a09

View file

@ -10,14 +10,14 @@
class BandcampIE(InfoExtractor): class BandcampIE(InfoExtractor):
IE_NAME = u'Bandcamp'
_VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)' _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)'
_TESTS = [{ _TESTS = [{
u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song', u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
u'file': u'1812978515.mp3', u'file': u'1812978515.mp3',
u'md5': u'cdeb30cdae1921719a3cbcab696ef53c', u'md5': u'c557841d5e50261777a6585648adf439',
u'info_dict': { u'info_dict': {
u"title": u"youtube-dl test song \"'/\\\u00e4\u21ad" u"title": u"youtube-dl \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
u"duration": 10,
}, },
u'skip': u'There is a limit of 200 free downloads / month for the test song' u'skip': u'There is a limit of 200 free downloads / month for the test song'
}] }]
@ -33,26 +33,39 @@ def _real_extract(self, url):
if m_trackinfo: if m_trackinfo:
json_code = m_trackinfo.group(1) json_code = m_trackinfo.group(1)
data = json.loads(json_code) data = json.loads(json_code)
d = data[0]
for d in data: duration = int(round(d['duration']))
formats = [{ formats = []
'format_id': 'format_id', for format_id, format_url in d['file'].items():
ext, _, abr_str = format_id.partition('-')
formats.append({
'format_id': format_id,
'url': format_url, 'url': format_url,
'ext': format_id.partition('-')[0] 'ext': format_id.partition('-')[0],
} for format_id, format_url in sorted(d['file'].items())] 'vcodec': 'none',
'acodec': format_id.partition('-')[0],
'abr': int(format_id.partition('-')[2]),
})
self._sort_formats(formats)
return { return {
'id': compat_str(d['id']), 'id': compat_str(d['id']),
'title': d['title'], 'title': d['title'],
'formats': formats, 'formats': formats,
'duration': duration,
} }
else: else:
raise ExtractorError(u'No free songs found') raise ExtractorError(u'No free songs found')
download_link = m_download.group(1) download_link = m_download.group(1)
id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$', video_id = re.search(
webpage, re.MULTILINE|re.DOTALL).group('id') r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
webpage, re.MULTILINE | re.DOTALL).group('id')
download_webpage = self._download_webpage(download_link, id, download_webpage = self._download_webpage(download_link, video_id,
'Downloading free downloads page') 'Downloading free downloads page')
# We get the dictionary of the track from some javascrip code # We get the dictionary of the track from some javascrip code
info = re.search(r'items: (.*?),$', info = re.search(r'items: (.*?),$',
@ -66,22 +79,22 @@ def _real_extract(self, url):
m_url = re.match(re_url, initial_url) m_url = re.match(re_url, initial_url)
#We build the url we will use to get the final track 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 # 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')) 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, id, 'Requesting download url') final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
# If we could correctly generate the .rand field the url would be # If we could correctly generate the .rand field the url would be
#in the "download_url" key #in the "download_url" key
final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1) final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
track_info = {'id':id, return {
'title' : info[u'title'], 'id': video_id,
'ext' : 'mp3', 'title': info[u'title'],
'url' : final_url, 'ext': 'mp3',
'thumbnail' : info[u'thumb_url'], 'vcodec': 'none',
'uploader' : info[u'artist'] 'url': final_url,
'thumbnail': info[u'thumb_url'],
'uploader': info[u'artist'],
} }
return [track_info]
class BandcampAlbumIE(InfoExtractor): class BandcampAlbumIE(InfoExtractor):
IE_NAME = u'Bandcamp:album' IE_NAME = u'Bandcamp:album'
@ -117,7 +130,7 @@ def _real_extract(self, url):
webpage = self._download_webpage(url, title) webpage = self._download_webpage(url, title)
tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage) tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
if not tracks_paths: if not tracks_paths:
raise ExtractorError(u'The page doesn\'t contain any track') raise ExtractorError(u'The page doesn\'t contain any tracks')
entries = [ entries = [
self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key()) self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
for t_path in tracks_paths] for t_path in tracks_paths]