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

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import random
   5 import re
   6 import time
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_str,
  11     compat_urlparse,
  12 )
  13 from ..utils import (
  14     ExtractorError,
  15     float_or_none,
  16     int_or_none,
  17     parse_filesize,
  18     unescapeHTML,
  19     update_url_query,
  20 )
  21
  22
  23 class BandcampIE(InfoExtractor):
  24     _VALID_URL = r'https?://.*?\.bandcamp\.com/track/(?P<title>.*)'
  25     _TESTS = [{
  26         'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  27         'md5': 'c557841d5e50261777a6585648adf439',
  28         'info_dict': {
  29             'id': '1812978515',
  30             'ext': 'mp3',
  31             'title': "youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
  32             'duration': 9.8485,
  33         },
  34         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  35     }, {
  36         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  37         'md5': '73d0b3171568232574e45652f8720b5c',
  38         'info_dict': {
  39             'id': '2650410135',
  40             'ext': 'mp3',
  41             'title': 'Lanius (Battle)',
  42             'uploader': 'Ben Prunty Music',
  43         },
  44     }]
  45
  46     def _real_extract(self, url):
  47         mobj = re.match(self._VALID_URL, url)
  48         title = mobj.group('title')
  49         webpage = self._download_webpage(url, title)
  50         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  51         if not m_download:
  52             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
  53             if m_trackinfo:
  54                 json_code = m_trackinfo.group(1)
  55                 data = json.loads(json_code)[0]
  56                 track_id = compat_str(data['id'])
  57
  58                 if not data.get('file'):
  59                     raise ExtractorError('Not streamable', video_id=track_id, expected=True)
  60
  61                 formats = []
  62                 for format_id, format_url in data['file'].items():
  63                     ext, abr_str = format_id.split('-', 1)
  64                     formats.append({
  65                         'format_id': format_id,
  66                         'url': self._proto_relative_url(format_url, 'http:'),
  67                         'ext': ext,
  68                         'vcodec': 'none',
  69                         'acodec': ext,
  70                         'abr': int_or_none(abr_str),
  71                     })
  72
  73                 self._sort_formats(formats)
  74
  75                 return {
  76                     'id': track_id,
  77                     'title': data['title'],
  78                     'formats': formats,
  79                     'duration': float_or_none(data.get('duration')),
  80                 }
  81             else:
  82                 raise ExtractorError('No free songs found')
  83
  84         download_link = m_download.group(1)
  85         video_id = self._search_regex(
  86             r'(?ms)var TralbumData = .*?[{,]\s*id: (?P<id>\d+),?$',
  87             webpage, 'video id')
  88
  89         download_webpage = self._download_webpage(
  90             download_link, video_id, 'Downloading free downloads page')
  91
  92         blob = self._parse_json(
  93             self._search_regex(
  94                 r'data-blob=(["\'])(?P<blob>{.+?})\1', download_webpage,
  95                 'blob', group='blob'),
  96             video_id, transform_source=unescapeHTML)
  97
  98         info = blob['digital_items'][0]
  99
 100         downloads = info['downloads']
 101         track = info['title']
 102
 103         artist = info.get('artist')
 104         title = '%s - %s' % (artist, track) if artist else track
 105
 106         download_formats = {}
 107         for f in blob['download_formats']:
 108             name, ext = f.get('name'), f.get('file_extension')
 109             if all(isinstance(x, compat_str) for x in (name, ext)):
 110                 download_formats[name] = ext.strip('.')
 111
 112         formats = []
 113         for format_id, f in downloads.items():
 114             format_url = f.get('url')
 115             if not format_url:
 116                 continue
 117             # Stat URL generation algorithm is reverse engineered from
 118             # download_*_bundle_*.js
 119             stat_url = update_url_query(
 120                 format_url.replace('/download/', '/statdownload/'), {
 121                     '.rand': int(time.time() * 1000 * random.random()),
 122                 })
 123             format_id = f.get('encoding_name') or format_id
 124             stat = self._download_json(
 125                 stat_url, video_id, 'Downloading %s JSON' % format_id,
 126                 transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
 127                 fatal=False)
 128             if not stat:
 129                 continue
 130             retry_url = stat.get('retry_url')
 131             if not isinstance(retry_url, compat_str):
 132                 continue
 133             formats.append({
 134                 'url': self._proto_relative_url(retry_url, 'http:'),
 135                 'ext': download_formats.get(format_id),
 136                 'format_id': format_id,
 137                 'format_note': f.get('description'),
 138                 'filesize': parse_filesize(f.get('size_mb')),
 139                 'vcodec': 'none',
 140             })
 141         self._sort_formats(formats)
 142
 143         return {
 144             'id': video_id,
 145             'title': title,
 146             'thumbnail': info.get('thumb_url'),
 147             'uploader': info.get('artist'),
 148             'artist': artist,
 149             'track': track,
 150             'formats': formats,
 151         }
 152
 153
 154 class BandcampAlbumIE(InfoExtractor):
 155     IE_NAME = 'Bandcamp:album'
 156     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^?#]+)|/?(?:$|[?#]))'
 157
 158     _TESTS = [{
 159         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 160         'playlist': [
 161             {
 162                 'md5': '39bc1eded3476e927c724321ddf116cf',
 163                 'info_dict': {
 164                     'id': '1353101989',
 165                     'ext': 'mp3',
 166                     'title': 'Intro',
 167                 }
 168             },
 169             {
 170                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 171                 'info_dict': {
 172                     'id': '38097443',
 173                     'ext': 'mp3',
 174                     'title': 'Kero One - Keep It Alive (Blazo remix)',
 175                 }
 176             },
 177         ],
 178         'info_dict': {
 179             'title': 'Jazz Format Mixtape vol.1',
 180             'id': 'jazz-format-mixtape-vol-1',
 181             'uploader_id': 'blazo',
 182         },
 183         'params': {
 184             'playlistend': 2
 185         },
 186         'skip': 'Bandcamp imposes download limits.'
 187     }, {
 188         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 189         'info_dict': {
 190             'title': 'Hierophany of the Open Grave',
 191             'uploader_id': 'nightbringer',
 192             'id': 'hierophany-of-the-open-grave',
 193         },
 194         'playlist_mincount': 9,
 195     }, {
 196         'url': 'http://dotscale.bandcamp.com',
 197         'info_dict': {
 198             'title': 'Loom',
 199             'id': 'dotscale',
 200             'uploader_id': 'dotscale',
 201         },
 202         'playlist_mincount': 7,
 203     }, {
 204         # with escaped quote in title
 205         'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
 206         'info_dict': {
 207             'title': '"Entropy" EP',
 208             'uploader_id': 'jstrecords',
 209             'id': 'entropy-ep',
 210         },
 211         'playlist_mincount': 3,
 212     }, {
 213         # not all tracks have songs
 214         'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
 215         'info_dict': {
 216             'id': 'we-are-the-plague',
 217             'title': 'WE ARE THE PLAGUE',
 218             'uploader_id': 'insulters',
 219         },
 220         'playlist_count': 2,
 221     }]
 222
 223     def _real_extract(self, url):
 224         mobj = re.match(self._VALID_URL, url)
 225         uploader_id = mobj.group('subdomain')
 226         album_id = mobj.group('album_id')
 227         playlist_id = album_id or uploader_id
 228         webpage = self._download_webpage(url, playlist_id)
 229         track_elements = re.findall(
 230             r'(?s)<div[^>]*>(.*?<a[^>]+href="([^"]+?)"[^>]+itemprop="url"[^>]*>.*?)</div>', webpage)
 231         if not track_elements:
 232             raise ExtractorError('The page doesn\'t contain any tracks')
 233         # Only tracks with duration info have songs
 234         entries = [
 235             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
 236             for elem_content, t_path in track_elements
 237             if self._html_search_meta('duration', elem_content, default=None)]
 238
 239         title = self._html_search_regex(
 240             r'album_title\s*:\s*"((?:\\.|[^"\\])+?)"',
 241             webpage, 'title', fatal=False)
 242         if title:
 243             title = title.replace(r'\"', '"')
 244         return {
 245             '_type': 'playlist',
 246             'uploader_id': uploader_id,
 247             'id': playlist_id,
 248             'title': title,
 249             'entries': entries,
 250         }