0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-15 01:15:06 +00:00
yt-dlp/yt_dlp/extractor/audiomack.py

147 lines
6 KiB
Python
Raw Normal View History

import itertools
import time
2014-10-23 21:55:39 +00:00
from .common import InfoExtractor
2014-10-24 04:54:59 +00:00
from .soundcloud import SoundcloudIE
from ..utils import (
ExtractorError,
url_basename,
)
2014-10-23 21:55:39 +00:00
class AudiomackIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?audiomack\.com/(?:song/|(?=.+/song/))(?P<id>[\w/-]+)'
2014-10-24 04:54:59 +00:00
IE_NAME = 'audiomack'
_TESTS = [
2015-01-09 00:35:33 +00:00
# hosted on audiomack
2014-10-24 04:54:59 +00:00
{
'url': 'http://www.audiomack.com/song/roosh-williams/extraordinary',
'info_dict':
{
'id': '310086',
2015-01-09 00:35:33 +00:00
'ext': 'mp3',
'uploader': 'Roosh Williams',
'title': 'Extraordinary',
},
2014-10-24 04:54:59 +00:00
},
2015-01-09 00:35:33 +00:00
# audiomack wrapper around soundcloud song
# Needs new test URL.
2014-10-24 04:54:59 +00:00
{
2014-12-03 13:53:12 +00:00
'add_ie': ['Soundcloud'],
'url': 'http://www.audiomack.com/song/hip-hop-daily/black-mamba-freestyle',
'info_dict': {
'id': '258901379',
'ext': 'mp3',
'description': 'mamba day freestyle for the legend Kobe Bryant ',
'title': 'Black Mamba Freestyle [Prod. By Danny Wolf]',
'uploader': 'ILOVEMAKONNEN',
'upload_date': '20160414',
},
'skip': 'Song has been removed from the site',
2014-12-03 13:53:12 +00:00
},
2014-10-24 04:54:59 +00:00
]
2014-10-23 21:55:39 +00:00
def _real_extract(self, url):
# URLs end with [uploader name]/song/[uploader title]
# this title is whatever the user types in, and is rarely
# the proper song title. Real metadata is in the api response
album_url_tag = self._match_id(url).replace('/song/', '/')
2014-10-23 21:55:39 +00:00
# Request the extended version of the api for extra fields like artist and title
2014-10-25 06:58:03 +00:00
api_response = self._download_json(
2015-01-09 00:35:33 +00:00
'http://www.audiomack.com/api/music/url/song/%s?extended=1&_=%d' % (
album_url_tag, time.time()),
album_url_tag)
2014-10-25 02:07:01 +00:00
# API is inconsistent with errors
2015-01-09 00:35:33 +00:00
if 'url' not in api_response or not api_response['url'] or 'error' in api_response:
raise ExtractorError(f'Invalid url {url}')
2014-10-24 04:54:59 +00:00
2014-11-23 19:41:03 +00:00
# Audiomack wraps a lot of soundcloud tracks in their branded wrapper
# if so, pass the work off to the soundcloud extractor
2015-01-09 00:35:33 +00:00
if SoundcloudIE.suitable(api_response['url']):
2019-01-01 16:56:05 +00:00
return self.url_result(api_response['url'], SoundcloudIE.ie_key())
2014-10-25 06:58:03 +00:00
return {
'id': str(api_response.get('id', album_url_tag)),
'uploader': api_response.get('artist'),
'title': api_response.get('title'),
'url': api_response['url'],
}
2014-10-25 06:58:03 +00:00
class AudiomackAlbumIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?audiomack\.com/(?:album/|(?=.+/album/))(?P<id>[\w/-]+)'
IE_NAME = 'audiomack:album'
_TESTS = [
# Standard album playlist
{
'url': 'http://www.audiomack.com/album/flytunezcom/tha-tour-part-2-mixtape',
'playlist_count': 11,
'info_dict':
{
2015-01-09 00:35:33 +00:00
'id': '812251',
'title': 'Tha Tour: Part 2 (Official Mixtape)',
},
},
# Album playlist ripped from fakeshoredrive with no metadata
{
2015-01-09 00:35:33 +00:00
'url': 'http://www.audiomack.com/album/fakeshoredrive/ppp-pistol-p-project',
2015-01-25 04:15:47 +00:00
'info_dict': {
'title': 'PPP (Pistol P Project)',
'id': '837572',
},
'playlist': [{
'info_dict': {
'title': 'PPP (Pistol P Project) - 8. Real (prod by SYK SENSE )',
'id': '837576',
'ext': 'mp3',
'uploader': 'Lil Herb a.k.a. G Herbo',
},
}, {
'info_dict': {
'title': 'PPP (Pistol P Project) - 10. 4 Minutes Of Hell Part 4 (prod by DY OF 808 MAFIA)',
'id': '837580',
'ext': 'mp3',
2015-01-25 04:15:47 +00:00
'uploader': 'Lil Herb a.k.a. G Herbo',
},
}],
},
]
def _real_extract(self, url):
# URLs end with [uploader name]/album/[uploader title]
# this title is whatever the user types in, and is rarely
# the proper song title. Real metadata is in the api response
album_url_tag = self._match_id(url).replace('/album/', '/')
2015-01-09 00:35:33 +00:00
result = {'_type': 'playlist', 'entries': []}
# There is no one endpoint for album metadata - instead it is included/repeated in each song's metadata
# Therefore we don't know how many songs the album has and must infi-loop until failure
for track_no in itertools.count():
# Get song's metadata
api_response = self._download_json(
'http://www.audiomack.com/api/music/url/album/%s/%d?extended=1&_=%d'
% (album_url_tag, track_no, time.time()), album_url_tag,
note=f'Querying song information ({track_no + 1})')
# Total failure, only occurs when url is totally wrong
# Won't happen in middle of valid playlist (next case)
2015-01-09 00:35:33 +00:00
if 'url' not in api_response or 'error' in api_response:
raise ExtractorError(f'Invalid url for track {track_no} of album url {url}')
# URL is good but song id doesn't exist - usually means end of playlist
2015-01-09 00:35:33 +00:00
elif not api_response['url']:
break
else:
# Pull out the album metadata and add to result (if it exists)
2015-01-09 00:35:33 +00:00
for resultkey, apikey in [('id', 'album_id'), ('title', 'album_title')]:
if apikey in api_response and resultkey not in result:
result[resultkey] = str(api_response[apikey])
song_id = url_basename(api_response['url']).rpartition('.')[0]
result['entries'].append({
'id': str(api_response.get('id', song_id)),
'uploader': api_response.get('artist'),
'title': api_response.get('title', song_id),
'url': api_response['url'],
})
return result