1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-12-05 08:27:43 +00:00
youtube-dl/youtube_dl/extractor/bandcamp.py

392 lines
14 KiB
Python
Raw Normal View History

2020-11-16 16:26:09 +00:00
# coding: utf-8
2014-01-07 09:04:48 +00:00
from __future__ import unicode_literals
import random
2013-06-23 20:24:58 +00:00
import re
import time
2013-06-23 20:24:58 +00:00
from .common import InfoExtractor
2020-11-16 16:26:09 +00:00
from ..compat import compat_str
from ..utils import (
2013-06-23 20:24:58 +00:00
ExtractorError,
2015-10-16 18:51:35 +00:00
float_or_none,
int_or_none,
KNOWN_EXTENSIONS,
parse_filesize,
str_or_none,
try_get,
update_url_query,
2017-06-04 13:47:05 +00:00
unified_strdate,
unified_timestamp,
2018-07-21 12:08:28 +00:00
url_or_none,
2020-11-16 16:26:09 +00:00
urljoin,
2013-06-23 20:24:58 +00:00
)
class BandcampIE(InfoExtractor):
2020-11-16 16:26:09 +00:00
_VALID_URL = r'https?://[^/]+\.bandcamp\.com/track/(?P<id>[^/?#&]+)'
_TESTS = [{
2014-01-07 09:04:48 +00:00
'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
'md5': 'c557841d5e50261777a6585648adf439',
'info_dict': {
'id': '1812978515',
'ext': 'mp3',
2020-11-16 16:26:09 +00:00
'title': "youtube-dl \"'/\\ä↭ - youtube-dl \"'/\\ä↭ - youtube-dl test song \"'/\\ä↭",
'duration': 9.8485,
2020-11-16 16:26:09 +00:00
'uploader': 'youtube-dl "\'/\\ä↭',
'upload_date': '20121129',
'timestamp': 1354224127,
2013-06-27 18:46:46 +00:00
},
2014-01-07 09:04:48 +00:00
'_skip': 'There is a limit of 200 free downloads / month for the test song'
}, {
# free download
'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
'info_dict': {
'id': '2650410135',
2017-04-27 20:13:12 +00:00
'ext': 'aiff',
'title': 'Ben Prunty - Lanius (Battle)',
'thumbnail': r're:^https?://.*\.jpg$',
2017-04-27 20:13:12 +00:00
'uploader': 'Ben Prunty',
'timestamp': 1396508491,
'upload_date': '20140403',
2021-03-09 20:41:21 +00:00
'release_timestamp': 1396483200,
'release_date': '20140403',
'duration': 260.877,
'track': 'Lanius (Battle)',
'track_number': 1,
'track_id': '2650410135',
'artist': 'Ben Prunty',
'album': 'FTL: Advanced Edition Soundtrack',
},
}, {
# no free download, mp3 128
'url': 'https://relapsealumni.bandcamp.com/track/hail-to-fire',
'md5': 'fec12ff55e804bb7f7ebeb77a800c8b7',
'info_dict': {
'id': '2584466013',
'ext': 'mp3',
'title': 'Mastodon - Hail to Fire',
'thumbnail': r're:^https?://.*\.jpg$',
'uploader': 'Mastodon',
'timestamp': 1322005399,
'upload_date': '20111122',
2021-03-09 20:41:21 +00:00
'release_timestamp': 1076112000,
'release_date': '20040207',
'duration': 120.79,
'track': 'Hail to Fire',
'track_number': 5,
'track_id': '2584466013',
'artist': 'Mastodon',
'album': 'Call of the Mastodon',
},
}]
2013-06-23 20:24:58 +00:00
2020-11-16 16:26:09 +00:00
def _extract_data_attr(self, webpage, video_id, attr='tralbum', fatal=True):
return self._parse_json(self._html_search_regex(
r'data-%s=(["\'])({.+?})\1' % attr, webpage,
attr + ' data', group=2), video_id, fatal=fatal)
2013-06-23 20:24:58 +00:00
def _real_extract(self, url):
2020-11-16 16:26:09 +00:00
title = self._match_id(url)
2013-06-23 20:24:58 +00:00
webpage = self._download_webpage(url, title)
2020-11-16 16:26:09 +00:00
tralbum = self._extract_data_attr(webpage, title)
thumbnail = self._og_search_thumbnail(webpage)
track_id = None
track = None
track_number = None
duration = None
formats = []
2020-11-16 16:26:09 +00:00
track_info = try_get(tralbum, lambda x: x['trackinfo'][0], dict)
if track_info:
file_ = track_info.get('file')
if isinstance(file_, dict):
for format_id, format_url in file_.items():
if not url_or_none(format_url):
continue
ext, abr_str = format_id.split('-', 1)
formats.append({
'format_id': format_id,
'url': self._proto_relative_url(format_url, 'http:'),
'ext': ext,
'vcodec': 'none',
'acodec': ext,
2015-10-16 18:51:35 +00:00
'abr': int_or_none(abr_str),
})
track = track_info.get('title')
2020-11-16 16:26:09 +00:00
track_id = str_or_none(
track_info.get('track_id') or track_info.get('id'))
track_number = int_or_none(track_info.get('track_num'))
duration = float_or_none(track_info.get('duration'))
2020-11-16 16:26:09 +00:00
embed = self._extract_data_attr(webpage, title, 'embed', False)
current = tralbum.get('current') or {}
artist = embed.get('artist') or current.get('artist') or tralbum.get('artist')
timestamp = unified_timestamp(
2020-11-16 16:26:09 +00:00
current.get('publish_date') or tralbum.get('album_publish_date'))
2020-11-16 16:26:09 +00:00
download_link = tralbum.get('freeDownloadPage')
if download_link:
2020-11-16 16:26:09 +00:00
track_id = compat_str(tralbum['id'])
download_webpage = self._download_webpage(
download_link, track_id, 'Downloading free downloads page')
2020-11-16 16:26:09 +00:00
blob = self._extract_data_attr(download_webpage, track_id, 'blob')
info = try_get(
blob, (lambda x: x['digital_items'][0],
lambda x: x['download_items'][0]), dict)
if info:
downloads = info.get('downloads')
if isinstance(downloads, dict):
if not track:
track = info.get('title')
if not artist:
artist = info.get('artist')
if not thumbnail:
thumbnail = info.get('thumb_url')
download_formats = {}
download_formats_list = blob.get('download_formats')
if isinstance(download_formats_list, list):
for f in blob['download_formats']:
name, ext = f.get('name'), f.get('file_extension')
if all(isinstance(x, compat_str) for x in (name, ext)):
download_formats[name] = ext.strip('.')
for format_id, f in downloads.items():
format_url = f.get('url')
if not format_url:
continue
# Stat URL generation algorithm is reverse engineered from
# download_*_bundle_*.js
stat_url = update_url_query(
format_url.replace('/download/', '/statdownload/'), {
'.rand': int(time.time() * 1000 * random.random()),
})
format_id = f.get('encoding_name') or format_id
stat = self._download_json(
stat_url, track_id, 'Downloading %s JSON' % format_id,
transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
fatal=False)
if not stat:
continue
retry_url = url_or_none(stat.get('retry_url'))
if not retry_url:
continue
formats.append({
'url': self._proto_relative_url(retry_url, 'http:'),
'ext': download_formats.get(format_id),
'format_id': format_id,
'format_note': f.get('description'),
'filesize': parse_filesize(f.get('size_mb')),
'vcodec': 'none',
})
self._sort_formats(formats)
title = '%s - %s' % (artist, track) if artist else track
if not duration:
duration = float_or_none(self._html_search_meta(
'duration', webpage, default=None))
2013-06-23 20:24:58 +00:00
return {
'id': track_id,
'title': title,
'thumbnail': thumbnail,
'uploader': artist,
'timestamp': timestamp,
2021-03-09 20:41:21 +00:00
'release_timestamp': unified_timestamp(tralbum.get('album_release_date')),
'duration': duration,
'track': track,
'track_number': track_number,
'track_id': track_id,
'artist': artist,
2020-11-16 16:26:09 +00:00
'album': embed.get('album_title'),
'formats': formats,
}
2020-11-16 16:26:09 +00:00
class BandcampAlbumIE(BandcampIE):
2014-01-07 09:04:48 +00:00
IE_NAME = 'Bandcamp:album'
2020-11-16 16:26:09 +00:00
_VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<id>[^/?#&]+))?'
_TESTS = [{
2014-01-07 09:04:48 +00:00
'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
'playlist': [
{
2014-01-07 09:04:48 +00:00
'md5': '39bc1eded3476e927c724321ddf116cf',
'info_dict': {
2014-11-12 14:00:54 +00:00
'id': '1353101989',
'ext': 'mp3',
2020-11-16 16:26:09 +00:00
'title': 'Blazo - Intro',
'timestamp': 1311756226,
'upload_date': '20110727',
'uploader': 'Blazo',
}
},
{
2014-01-07 09:04:48 +00:00
'md5': '1a2c32e2691474643e912cc6cd4bffaa',
'info_dict': {
2014-11-12 14:00:54 +00:00
'id': '38097443',
'ext': 'mp3',
2020-11-16 16:26:09 +00:00
'title': 'Blazo - Kero One - Keep It Alive (Blazo remix)',
'timestamp': 1311757238,
'upload_date': '20110727',
'uploader': 'Blazo',
}
},
],
2014-11-12 14:00:54 +00:00
'info_dict': {
'title': 'Jazz Format Mixtape vol.1',
'id': 'jazz-format-mixtape-vol-1',
'uploader_id': 'blazo',
2014-11-12 14:00:54 +00:00
},
2014-01-07 09:04:48 +00:00
'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,
}, {
# with escaped quote in title
'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
'info_dict': {
'title': '"Entropy" EP',
'uploader_id': 'jstrecords',
'id': 'entropy-ep',
'description': 'md5:0ff22959c943622972596062f2f366a5',
},
'playlist_mincount': 3,
}, {
# not all tracks have songs
'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
'info_dict': {
'id': 'we-are-the-plague',
'title': 'WE ARE THE PLAGUE',
'uploader_id': 'insulters',
'description': 'md5:b3cf845ee41b2b1141dc7bde9237255f',
},
'playlist_count': 2,
}]
2017-06-04 13:47:05 +00:00
@classmethod
def suitable(cls, url):
return (False
if BandcampWeeklyIE.suitable(url) or BandcampIE.suitable(url)
else super(BandcampAlbumIE, cls).suitable(url))
2017-06-04 13:47:05 +00:00
def _real_extract(self, url):
2020-11-16 16:26:09 +00:00
uploader_id, album_id = re.match(self._VALID_URL, url).groups()
playlist_id = album_id or uploader_id
webpage = self._download_webpage(url, playlist_id)
2020-11-16 16:26:09 +00:00
tralbum = self._extract_data_attr(webpage, playlist_id)
track_info = tralbum.get('trackinfo')
if not track_info:
2014-01-07 09:04:48 +00:00
raise ExtractorError('The page doesn\'t contain any tracks')
# Only tracks with duration info have songs
entries = [
2017-08-20 16:32:33 +00:00
self.url_result(
2020-11-16 16:26:09 +00:00
urljoin(url, t['title_link']), BandcampIE.ie_key(),
str_or_none(t.get('track_id') or t.get('id')), t.get('title'))
for t in track_info
if t.get('duration')]
current = tralbum.get('current') or {}
return {
'_type': 'playlist',
'uploader_id': uploader_id,
'id': playlist_id,
'title': current.get('title'),
'description': current.get('about'),
'entries': entries,
}
2017-06-04 13:47:05 +00:00
2020-11-16 16:26:09 +00:00
class BandcampWeeklyIE(BandcampIE):
IE_NAME = 'Bandcamp:weekly'
_VALID_URL = r'https?://(?:www\.)?bandcamp\.com/?\?(?:.*?&)?show=(?P<id>\d+)'
2017-06-04 13:47:05 +00:00
_TESTS = [{
'url': 'https://bandcamp.com/?show=224',
'md5': 'b00df799c733cf7e0c567ed187dea0fd',
'info_dict': {
'id': '224',
'ext': 'opus',
'title': 'BC Weekly April 4th 2017 - Magic Moments',
'description': 'md5:5d48150916e8e02d030623a48512c874',
'duration': 5829.77,
'release_date': '20170404',
'series': 'Bandcamp Weekly',
'episode': 'Magic Moments',
'episode_id': '224',
2020-11-16 16:26:09 +00:00
},
'params': {
'format': 'opus-lo',
},
2017-06-04 13:47:05 +00:00
}, {
'url': 'https://bandcamp.com/?blah/blah@&show=228',
'only_matching': True
}]
def _real_extract(self, url):
2020-11-16 16:26:09 +00:00
show_id = self._match_id(url)
webpage = self._download_webpage(url, show_id)
2017-06-04 13:47:05 +00:00
2020-11-16 16:26:09 +00:00
blob = self._extract_data_attr(webpage, show_id, 'blob')
2017-06-04 13:47:05 +00:00
2020-11-16 16:26:09 +00:00
show = blob['bcw_data'][show_id]
2017-06-04 13:47:05 +00:00
formats = []
for format_id, format_url in show['audio_stream'].items():
2018-07-21 12:08:28 +00:00
if not url_or_none(format_url):
continue
for known_ext in KNOWN_EXTENSIONS:
if known_ext in format_id:
ext = known_ext
break
else:
ext = None
formats.append({
'format_id': format_id,
'url': format_url,
'ext': ext,
'vcodec': 'none',
})
self._sort_formats(formats)
2017-06-04 13:47:05 +00:00
title = show.get('audio_title') or 'Bandcamp Weekly'
subtitle = show.get('subtitle')
if subtitle:
title += ' - %s' % subtitle
2017-06-04 13:47:05 +00:00
return {
2020-11-16 16:26:09 +00:00
'id': show_id,
'title': title,
'description': show.get('desc') or show.get('short_desc'),
2017-06-04 13:47:05 +00:00
'duration': float_or_none(show.get('audio_duration')),
'is_live': False,
'release_date': unified_strdate(show.get('published_date')),
'series': 'Bandcamp Weekly',
'episode': show.get('subtitle'),
2020-11-16 16:26:09 +00:00
'episode_id': show_id,
2017-06-04 13:47:05 +00:00
'formats': formats
}