ytdl/youtube_dl/extractor/bandcamp.py

182 lines
6.6 KiB
Python
Raw Normal View History

2014-01-07 10:04:48 +01:00
from __future__ import unicode_literals
2013-06-23 22:24:58 +02:00
import json
import re
from .common import InfoExtractor
from ..compat import (
compat_str,
compat_urlparse,
)
from ..utils import (
2013-06-23 22:24:58 +02:00
ExtractorError,
)
class BandcampIE(InfoExtractor):
_VALID_URL = r'https?://.*?\.bandcamp\.com/track/(?P<title>.*)'
_TESTS = [{
2014-01-07 10:04:48 +01:00
'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,
2013-06-27 20:46:46 +02:00
},
2014-01-07 10:04:48 +01:00
'_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',
},
}]
2013-06-23 22:24:58 +02:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
title = mobj.group('title')
webpage = self._download_webpage(url, title)
m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
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)[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)
return {
'id': compat_str(data['id']),
'title': data['title'],
'formats': formats,
'duration': float(data['duration']),
}
else:
2014-01-07 10:04:48 +01:00
raise ExtractorError('No free songs found')
2013-06-23 22:24:58 +02:00
download_link = m_download.group(1)
video_id = self._search_regex(
r'(?ms)var TralbumData = .*?[{,]\s*id: (?P<id>\d+),?$',
webpage, 'video id')
2013-06-23 22:24:58 +02:00
download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page')
# We get the dictionary of the track from some javascript code
all_info = self._parse_json(self._search_regex(
r'(?sm)items: (.*?),$', download_webpage, 'items'), video_id)
2015-02-10 01:37:14 +01:00
info = all_info[0]
2013-06-23 22:24:58 +02:00
# We pick mp3-320 for now, until format selection can be easily implemented.
2014-01-07 10:04:48 +01:00
mp3_info = info['downloads']['mp3-320']
2013-06-23 22:24:58 +02:00
# If we try to use this url it says the link has expired
2014-01-07 10:04:48 +01:00
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>.*)$',
initial_url)
2014-11-23 20:41:03 +01:00
# We build the url we will use to get the final track url
2013-06-23 22:24:58 +02:00
# This url is build in Bandcamp in the script download_bunde_*.js
request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts'))
2014-01-09 20:23:28 +01:00
final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
2013-06-23 22:24:58 +02:00
# If we could correctly generate the .rand field the url would be
2014-11-23 20:41:03 +01:00
# in the "download_url" key
final_url = self._search_regex(
r'"retry_url":"(.*?)"', final_url_webpage, 'final video URL')
2013-06-23 22:24:58 +02:00
return {
'id': video_id,
2014-01-07 10:04:48 +01:00
'title': info['title'],
'ext': 'mp3',
'vcodec': 'none',
'url': final_url,
'thumbnail': info.get('thumb_url'),
'uploader': info.get('artist'),
}
class BandcampAlbumIE(InfoExtractor):
2014-01-07 10:04:48 +01:00
IE_NAME = 'Bandcamp:album'
_VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^?#]+)|/?(?:$|[?#]))'
_TESTS = [{
2014-01-07 10:04:48 +01:00
'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
'playlist': [
{
2014-01-07 10:04:48 +01:00
'md5': '39bc1eded3476e927c724321ddf116cf',
'info_dict': {
2014-11-12 15:00:54 +01:00
'id': '1353101989',
'ext': 'mp3',
2014-01-07 10:04:48 +01:00
'title': 'Intro',
}
},
{
2014-01-07 10:04:48 +01:00
'md5': '1a2c32e2691474643e912cc6cd4bffaa',
'info_dict': {
2014-11-12 15:00:54 +01:00
'id': '38097443',
'ext': 'mp3',
2014-01-07 10:04:48 +01:00
'title': 'Kero One - Keep It Alive (Blazo remix)',
}
},
],
2014-11-12 15:00:54 +01:00
'info_dict': {
'title': 'Jazz Format Mixtape vol.1',
'id': 'jazz-format-mixtape-vol-1',
'uploader_id': 'blazo',
2014-11-12 15:00:54 +01:00
},
2014-01-07 10:04:48 +01:00
'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,
}]
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:
2014-01-07 10:04:48 +01:00
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\s*:\s*"(.*?)"', webpage, 'title', fatal=False)
return {
'_type': 'playlist',
'uploader_id': uploader_id,
'id': playlist_id,
'title': title,
'entries': entries,
}