- id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
- webpage, re.MULTILINE|re.DOTALL).group('id')
-
- download_webpage = self._download_webpage(download_link, id,
- 'Downloading free downloads page')
- # We get the dictionary of the track from some javascrip code
- info = re.search(r'items: (.*?),$',
- download_webpage, re.MULTILINE).group(1)
- info = json.loads(info)[0]
- # We pick mp3-320 for now, until format selection can be easily implemented.
- mp3_info = info[u'downloads'][u'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>.*)$'
- m_url = re.match(re_url, initial_url)
- #We build the url we will use to get the final track url
- # This url is build in Bandcamp in the script download_bunde_*.js
- request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), id, m_url.group('ts'))
- final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
- # If we could correctly generate the .rand field the url would be
- #in the "download_url" key
- final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
-
- track_info = {'id':id,
- 'title' : info[u'title'],
- 'ext' : 'mp3',
- 'url' : final_url,
- 'thumbnail' : info[u'thumb_url'],
- 'uploader' : info[u'artist']
- }
-
- return [track_info]
+ video_id = self._search_regex(
+ r'(?ms)var TralbumData = .*?[{,]\s*id: (?P<id>\d+),?$',
+ webpage, 'video id')
+
+ download_webpage = self._download_webpage(
+ download_link, video_id, 'Downloading free downloads page')
+
+ blob = self._parse_json(
+ self._search_regex(
+ r'data-blob=(["\'])(?P<blob>{.+?})\1', download_webpage,
+ 'blob', group='blob'),
+ video_id, transform_source=unescapeHTML)
+
+ info = blob['digital_items'][0]
+
+ downloads = info['downloads']
+ track = info['title']
+
+ artist = info.get('artist')
+ title = '%s - %s' % (artist, track) if artist else track
+
+ download_formats = {}
+ for f in blob['download_formats']:
+ name, ext = f.get('name'), f.get('file_extension')
+ if all(isinstance(x, compat_str) for x in (name, ext)):
+ download_formats[name] = ext.strip('.')
+
+ formats = []
+ for format_id, f in downloads.items():
+ format_url = f.get('url')
+ if not format_url:
+ continue
+ # Stat URL generation algorithm is reverse engineered from
+ # download_*_bundle_*.js
+ stat_url = update_url_query(
+ format_url.replace('/download/', '/statdownload/'), {
+ '.rand': int(time.time() * 1000 * random.random()),
+ })
+ format_id = f.get('encoding_name') or format_id
+ stat = self._download_json(
+ stat_url, video_id, 'Downloading %s JSON' % format_id,
+ transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
+ fatal=False)
+ if not stat:
+ continue
+ retry_url = stat.get('retry_url')
+ if not isinstance(retry_url, compat_str):
+ continue
+ formats.append({
+ 'url': self._proto_relative_url(retry_url, 'http:'),
+ 'ext': download_formats.get(format_id),
+ 'format_id': format_id,
+ 'format_note': f.get('description'),
+ 'filesize': parse_filesize(f.get('size_mb')),
+ 'vcodec': 'none',
+ })
+ self._sort_formats(formats)
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'thumbnail': info.get('thumb_url'),
+ 'uploader': info.get('artist'),
+ 'artist': artist,
+ 'track': track,
+ 'formats': formats,
+ }
+
+
+class BandcampAlbumIE(InfoExtractor):
+ IE_NAME = 'Bandcamp:album'
+ _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^?#]+)|/?(?:$|[?#]))'
+
+ _TESTS = [{
+ 'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
+ 'playlist': [
+ {
+ 'md5': '39bc1eded3476e927c724321ddf116cf',
+ 'info_dict': {
+ 'id': '1353101989',
+ 'ext': 'mp3',
+ 'title': 'Intro',
+ }
+ },
+ {
+ 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
+ 'info_dict': {
+ 'id': '38097443',
+ 'ext': 'mp3',
+ 'title': 'Kero One - Keep It Alive (Blazo remix)',
+ }
+ },
+ ],
+ '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',
+ },
+ 'playlist_mincount': 9,
+ }, {
+ 'url': 'http://dotscale.bandcamp.com',
+ 'info_dict': {
+ 'title': 'Loom',
+ 'id': 'dotscale',
+ 'uploader_id': 'dotscale',
+ },
+ 'playlist_mincount': 7,
+ }, {
+ # with escaped quote in title
+ 'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
+ 'info_dict': {
+ 'title': '"Entropy" EP',
+ 'uploader_id': 'jstrecords',
+ 'id': 'entropy-ep',
+ },
+ 'playlist_mincount': 3,
+ }]
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ 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('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._html_search_regex(
+ r'album_title\s*:\s*"((?:\\.|[^"\\])+?)"',
+ webpage, 'title', fatal=False)
+ if title:
+ title = title.replace(r'\"', '"')
+ return {
+ '_type': 'playlist',
+ 'uploader_id': uploader_id,
+ 'id': playlist_id,
+ 'title': title,
+ 'entries': entries,
+ }
Rapsys Git / youtubedl / blobdiff