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': '0369ace6b939f0927e62c67a1a8d9fa7',
  38         'info_dict': {
  39             'id': '2650410135',
  40             'ext': 'aiff',
  41             'title': 'Ben Prunty - Lanius (Battle)',
  42             'uploader': 'Ben Prunty',
  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         thumbnail = self._html_search_meta('og:image', webpage, default=None)
  51         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  52         if not m_download:
  53             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
  54             if m_trackinfo:
  55                 json_code = m_trackinfo.group(1)
  56                 data = json.loads(json_code)[0]
  57                 track_id = compat_str(data['id'])
  58
  59                 if not data.get('file'):
  60                     raise ExtractorError('Not streamable', video_id=track_id, expected=True)
  61
  62                 formats = []
  63                 for format_id, format_url in data['file'].items():
  64                     ext, abr_str = format_id.split('-', 1)
  65                     formats.append({
  66                         'format_id': format_id,
  67                         'url': self._proto_relative_url(format_url, 'http:'),
  68                         'ext': ext,
  69                         'vcodec': 'none',
  70                         'acodec': ext,
  71                         'abr': int_or_none(abr_str),
  72                     })
  73
  74                 self._sort_formats(formats)
  75
  76                 return {
  77                     'id': track_id,
  78                     'title': data['title'],
  79                     'thumbnail': thumbnail,
  80                     'formats': formats,
  81                     'duration': float_or_none(data.get('duration')),
  82                 }
  83             else:
  84                 raise ExtractorError('No free songs found')
  85
  86         download_link = m_download.group(1)
  87         video_id = self._search_regex(
  88             r'(?ms)var TralbumData = .*?[{,]\s*id: (?P<id>\d+),?$',
  89             webpage, 'video id')
  90
  91         download_webpage = self._download_webpage(
  92             download_link, video_id, 'Downloading free downloads page')
  93
  94         blob = self._parse_json(
  95             self._search_regex(
  96                 r'data-blob=(["\'])(?P<blob>{.+?})\1', download_webpage,
  97                 'blob', group='blob'),
  98             video_id, transform_source=unescapeHTML)
  99
 100         info = blob['digital_items'][0]
 101
 102         downloads = info['downloads']
 103         track = info['title']
 104
 105         artist = info.get('artist')
 106         title = '%s - %s' % (artist, track) if artist else track
 107
 108         download_formats = {}
 109         for f in blob['download_formats']:
 110             name, ext = f.get('name'), f.get('file_extension')
 111             if all(isinstance(x, compat_str) for x in (name, ext)):
 112                 download_formats[name] = ext.strip('.')
 113
 114         formats = []
 115         for format_id, f in downloads.items():
 116             format_url = f.get('url')
 117             if not format_url:
 118                 continue
 119             # Stat URL generation algorithm is reverse engineered from
 120             # download_*_bundle_*.js
 121             stat_url = update_url_query(
 122                 format_url.replace('/download/', '/statdownload/'), {
 123                     '.rand': int(time.time() * 1000 * random.random()),
 124                 })
 125             format_id = f.get('encoding_name') or format_id
 126             stat = self._download_json(
 127                 stat_url, video_id, 'Downloading %s JSON' % format_id,
 128                 transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
 129                 fatal=False)
 130             if not stat:
 131                 continue
 132             retry_url = stat.get('retry_url')
 133             if not isinstance(retry_url, compat_str):
 134                 continue
 135             formats.append({
 136                 'url': self._proto_relative_url(retry_url, 'http:'),
 137                 'ext': download_formats.get(format_id),
 138                 'format_id': format_id,
 139                 'format_note': f.get('description'),
 140                 'filesize': parse_filesize(f.get('size_mb')),
 141                 'vcodec': 'none',
 142             })
 143         self._sort_formats(formats)
 144
 145         return {
 146             'id': video_id,
 147             'title': title,
 148             'thumbnail': info.get('thumb_url') or thumbnail,
 149             'uploader': info.get('artist'),
 150             'artist': artist,
 151             'track': track,
 152             'formats': formats,
 153         }
 154
 155
 156 class BandcampAlbumIE(InfoExtractor):
 157     IE_NAME = 'Bandcamp:album'
 158     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^?#]+)|/?(?:$|[?#]))'
 159
 160     _TESTS = [{
 161         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 162         'playlist': [
 163             {
 164                 'md5': '39bc1eded3476e927c724321ddf116cf',
 165                 'info_dict': {
 166                     'id': '1353101989',
 167                     'ext': 'mp3',
 168                     'title': 'Intro',
 169                 }
 170             },
 171             {
 172                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 173                 'info_dict': {
 174                     'id': '38097443',
 175                     'ext': 'mp3',
 176                     'title': 'Kero One - Keep It Alive (Blazo remix)',
 177                 }
 178             },
 179         ],
 180         'info_dict': {
 181             'title': 'Jazz Format Mixtape vol.1',
 182             'id': 'jazz-format-mixtape-vol-1',
 183             'uploader_id': 'blazo',
 184         },
 185         'params': {
 186             'playlistend': 2
 187         },
 188         'skip': 'Bandcamp imposes download limits.'
 189     }, {
 190         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 191         'info_dict': {
 192             'title': 'Hierophany of the Open Grave',
 193             'uploader_id': 'nightbringer',
 194             'id': 'hierophany-of-the-open-grave',
 195         },
 196         'playlist_mincount': 9,
 197     }, {
 198         'url': 'http://dotscale.bandcamp.com',
 199         'info_dict': {
 200             'title': 'Loom',
 201             'id': 'dotscale',
 202             'uploader_id': 'dotscale',
 203         },
 204         'playlist_mincount': 7,
 205     }, {
 206         # with escaped quote in title
 207         'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
 208         'info_dict': {
 209             'title': '"Entropy" EP',
 210             'uploader_id': 'jstrecords',
 211             'id': 'entropy-ep',
 212         },
 213         'playlist_mincount': 3,
 214     }, {
 215         # not all tracks have songs
 216         'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
 217         'info_dict': {
 218             'id': 'we-are-the-plague',
 219             'title': 'WE ARE THE PLAGUE',
 220             'uploader_id': 'insulters',
 221         },
 222         'playlist_count': 2,
 223     }]
 224
 225     def _real_extract(self, url):
 226         mobj = re.match(self._VALID_URL, url)
 227         uploader_id = mobj.group('subdomain')
 228         album_id = mobj.group('album_id')
 229         playlist_id = album_id or uploader_id
 230         webpage = self._download_webpage(url, playlist_id)
 231         track_elements = re.findall(
 232             r'(?s)<div[^>]*>(.*?<a[^>]+href="([^"]+?)"[^>]+itemprop="url"[^>]*>.*?)</div>', webpage)
 233         if not track_elements:
 234             raise ExtractorError('The page doesn\'t contain any tracks')
 235         # Only tracks with duration info have songs
 236         entries = [
 237             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
 238             for elem_content, t_path in track_elements
 239             if self._html_search_meta('duration', elem_content, default=None)]
 240
 241         title = self._html_search_regex(
 242             r'album_title\s*:\s*"((?:\\.|[^"\\])+?)"',
 243             webpage, 'title', fatal=False)
 244         if title:
 245             title = title.replace(r'\"', '"')
 246         return {
 247             '_type': 'playlist',
 248             'uploader_id': uploader_id,
 249             'id': playlist_id,
 250             'title': title,
 251             'entries': entries,
 252         }