Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bandcamp.py

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_str,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13 )
  14
  15
  16 class BandcampIE(InfoExtractor):
  17     _VALID_URL = r'https?://.*?\.bandcamp\.com/track/(?P<title>.*)'
  18     _TESTS = [{
  19         'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  20         'md5': 'c557841d5e50261777a6585648adf439',
  21         'info_dict': {
  22             'id': '1812978515',
  23             'ext': 'mp3',
  24             'title': "youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
  25             'duration': 9.8485,
  26         },
  27         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  28     }, {
  29         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  30         'md5': '2b68e5851514c20efdff2afc5603b8b4',
  31         'info_dict': {
  32             'id': '2650410135',
  33             'ext': 'mp3',
  34             'title': 'Lanius (Battle)',
  35             'uploader': 'Ben Prunty Music',
  36         },
  37     }]
  38
  39     def _real_extract(self, url):
  40         mobj = re.match(self._VALID_URL, url)
  41         title = mobj.group('title')
  42         webpage = self._download_webpage(url, title)
  43         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  44         if not m_download:
  45             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
  46             if m_trackinfo:
  47                 json_code = m_trackinfo.group(1)
  48                 data = json.loads(json_code)[0]
  49
  50                 formats = []
  51                 for format_id, format_url in data['file'].items():
  52                     ext, abr_str = format_id.split('-', 1)
  53                     formats.append({
  54                         'format_id': format_id,
  55                         'url': format_url,
  56                         'ext': ext,
  57                         'vcodec': 'none',
  58                         'acodec': ext,
  59                         'abr': int(abr_str),
  60                     })
  61
  62                 self._sort_formats(formats)
  63
  64                 return {
  65                     'id': compat_str(data['id']),
  66                     'title': data['title'],
  67                     'formats': formats,
  68                     'duration': float(data['duration']),
  69                 }
  70             else:
  71                 raise ExtractorError('No free songs found')
  72
  73         download_link = m_download.group(1)
  74         video_id = self._search_regex(
  75             r'var TralbumData = {.*?id: (?P<id>\d+),?$',
  76             webpage, 'video id', flags=re.MULTILINE | re.DOTALL)
  77
  78         download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page')
  79         # We get the dictionary of the track from some javascript code
  80         info = re.search(r'items: (.*?),$', download_webpage, re.MULTILINE).group(1)
  81         info = json.loads(info)[0]
  82         # We pick mp3-320 for now, until format selection can be easily implemented.
  83         mp3_info = info['downloads']['mp3-320']
  84         # If we try to use this url it says the link has expired
  85         initial_url = mp3_info['url']
  86         re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
  87         m_url = re.match(re_url, initial_url)
  88         # We build the url we will use to get the final track url
  89         # This url is build in Bandcamp in the script download_bunde_*.js
  90         request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts'))
  91         final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
  92         # If we could correctly generate the .rand field the url would be
  93         # in the "download_url" key
  94         final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
  95
  96         return {
  97             'id': video_id,
  98             'title': info['title'],
  99             'ext': 'mp3',
 100             'vcodec': 'none',
 101             'url': final_url,
 102             'thumbnail': info.get('thumb_url'),
 103             'uploader': info.get('artist'),
 104         }
 105
 106
 107 class BandcampAlbumIE(InfoExtractor):
 108     IE_NAME = 'Bandcamp:album'
 109     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<title>[^?#]+)|/?(?:$|[?#]))'
 110
 111     _TESTS = [{
 112         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 113         'playlist': [
 114             {
 115                 'md5': '39bc1eded3476e927c724321ddf116cf',
 116                 'info_dict': {
 117                     'id': '1353101989',
 118                     'ext': 'mp3',
 119                     'title': 'Intro',
 120                 }
 121             },
 122             {
 123                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 124                 'info_dict': {
 125                     'id': '38097443',
 126                     'ext': 'mp3',
 127                     'title': 'Kero One - Keep It Alive (Blazo remix)',
 128                 }
 129             },
 130         ],
 131         'info_dict': {
 132             'title': 'Jazz Format Mixtape vol.1',
 133         },
 134         'params': {
 135             'playlistend': 2
 136         },
 137         'skip': 'Bandcamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
 138     }, {
 139         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 140         'info_dict': {
 141             'title': 'Hierophany of the Open Grave',
 142         },
 143         'playlist_mincount': 9,
 144     }, {
 145         'url': 'http://dotscale.bandcamp.com',
 146         'info_dict': {
 147             'title': 'Loom',
 148         },
 149         'playlist_mincount': 7,
 150     }]
 151
 152     def _real_extract(self, url):
 153         mobj = re.match(self._VALID_URL, url)
 154         playlist_id = mobj.group('subdomain')
 155         title = mobj.group('title')
 156         display_id = title or playlist_id
 157         webpage = self._download_webpage(url, display_id)
 158         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
 159         if not tracks_paths:
 160             raise ExtractorError('The page doesn\'t contain any tracks')
 161         entries = [
 162             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
 163             for t_path in tracks_paths]
 164         title = self._search_regex(
 165             r'album_title\s*:\s*"(.*?)"', webpage, 'title', fatal=False)
 166         return {
 167             '_type': 'playlist',
 168             'id': playlist_id,
 169             'display_id': display_id,
 170             'title': title,
 171             'entries': entries,
 172         }