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