paradizelost/youtube-dl
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[bandcamp] fix extraction

Browse Source
This commit is contained in:
Remita Amine 2020-11-16 17:26:09 +01:00
parent fe13087cd1
commit 3f1748b944
1 changed files with 58 additions and 91 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

149
youtube_dl/extractor/bandcamp.py
View File

@ -1,3 +1,4 @@
# coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import random import random
@ -5,10 +6,7 @@ import re
import time import time
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import compat_str
compat_str,
compat_urlparse,
)
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
float_or_none, float_or_none,
@ -17,30 +15,32 @@ from ..utils import (
parse_filesize, parse_filesize,
str_or_none, str_or_none,
try_get, try_get,
unescapeHTML,
update_url_query, update_url_query,
unified_strdate, unified_strdate,
unified_timestamp, unified_timestamp,
url_or_none, url_or_none,
urljoin,
) )
class BandcampIE(InfoExtractor): class BandcampIE(InfoExtractor):
_VALID_URL = r'https?://[^/]+\.bandcamp\.com/track/(?P<title>[^/?#&]+)' _VALID_URL = r'https?://[^/]+\.bandcamp\.com/track/(?P<id>[^/?#&]+)'
_TESTS = [{ _TESTS = [{
'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song', 'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
'md5': 'c557841d5e50261777a6585648adf439', 'md5': 'c557841d5e50261777a6585648adf439',
'info_dict': { 'info_dict': {
'id': '1812978515', 'id': '1812978515',
'ext': 'mp3', 'ext': 'mp3',
'title': "youtube-dl \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad", 'title': "youtube-dl \"'/\\ä↭ - youtube-dl \"'/\\ä↭ - youtube-dl test song \"'/\\ä↭",
'duration': 9.8485, 'duration': 9.8485,
'uploader': 'youtube-dl "\'/\\ä↭',
'upload_date': '20121129',
'timestamp': 1354224127,
}, },
'_skip': 'There is a limit of 200 free downloads / month for the test song' '_skip': 'There is a limit of 200 free downloads / month for the test song'
}, { }, {
# free download # free download
'url': 'http://benprunty.bandcamp.com/track/lanius-battle', 'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
'md5': '853e35bf34aa1d6fe2615ae612564b36',
'info_dict': { 'info_dict': {
'id': '2650410135', 'id': '2650410135',
'ext': 'aiff', 'ext': 'aiff',
@ -79,11 +79,16 @@ class BandcampIE(InfoExtractor):
}, },
}] }]
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)
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) title = self._match_id(url)
title = mobj.group('title')
webpage = self._download_webpage(url, title) webpage = self._download_webpage(url, title)
thumbnail = self._html_search_meta('og:image', webpage, default=None) tralbum = self._extract_data_attr(webpage, title)
thumbnail = self._og_search_thumbnail(webpage)
track_id = None track_id = None
track = None track = None
@ -91,10 +96,7 @@ class BandcampIE(InfoExtractor):
duration = None duration = None
formats = [] formats = []
track_info = self._parse_json( track_info = try_get(tralbum, lambda x: x['trackinfo'][0], dict)
self._search_regex(
r'trackinfo\s*:\s*\[\s*({.+?})\s*\]\s*,\s*?\n',
webpage, 'track info', default='{}'), title)
if track_info: if track_info:
file_ = track_info.get('file') file_ = track_info.get('file')
if isinstance(file_, dict): if isinstance(file_, dict):
@ -111,37 +113,25 @@ class BandcampIE(InfoExtractor):
'abr': int_or_none(abr_str), 'abr': int_or_none(abr_str),
}) })
track = track_info.get('title') track = track_info.get('title')
track_id = str_or_none(track_info.get('track_id') or track_info.get('id')) 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')) track_number = int_or_none(track_info.get('track_num'))
duration = float_or_none(track_info.get('duration')) duration = float_or_none(track_info.get('duration'))
def extract(key): embed = self._extract_data_attr(webpage, title, 'embed', False)
return self._search_regex( current = tralbum.get('current') or {}
r'\b%s\s*["\']?\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1' % key, artist = embed.get('artist') or current.get('artist') or tralbum.get('artist')
webpage, key, default=None, group='value')
artist = extract('artist')
album = extract('album_title')
timestamp = unified_timestamp( timestamp = unified_timestamp(
extract('publish_date') or extract('album_publish_date')) current.get('publish_date') or tralbum.get('album_publish_date'))
release_date = unified_strdate(extract('album_release_date'))
download_link = self._search_regex( download_link = tralbum.get('freeDownloadPage')
r'freeDownloadPage\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
'download link', default=None, group='url')
if download_link: if download_link:
track_id = self._search_regex( track_id = compat_str(tralbum['id'])
r'(?ms)var TralbumData = .*?[{,]\s*id: (?P<id>\d+),?$',
webpage, 'track id')
download_webpage = self._download_webpage( download_webpage = self._download_webpage(
download_link, track_id, 'Downloading free downloads page') download_link, track_id, 'Downloading free downloads page')
blob = self._parse_json( blob = self._extract_data_attr(download_webpage, track_id, 'blob')
self._search_regex(
r'data-blob=(["\'])(?P<blob>{.+?})\1', download_webpage,
'blob', group='blob'),
track_id, transform_source=unescapeHTML)
info = try_get( info = try_get(
blob, (lambda x: x['digital_items'][0], blob, (lambda x: x['digital_items'][0],
@ -207,20 +197,20 @@ class BandcampIE(InfoExtractor):
'thumbnail': thumbnail, 'thumbnail': thumbnail,
'uploader': artist, 'uploader': artist,
'timestamp': timestamp, 'timestamp': timestamp,
'release_date': release_date, 'release_date': unified_strdate(tralbum.get('album_release_date')),
'duration': duration, 'duration': duration,
'track': track, 'track': track,
'track_number': track_number, 'track_number': track_number,
'track_id': track_id, 'track_id': track_id,
'artist': artist, 'artist': artist,
'album': album, 'album': embed.get('album_title'),
'formats': formats, 'formats': formats,
} }
class BandcampAlbumIE(InfoExtractor): class BandcampAlbumIE(BandcampIE):
IE_NAME = 'Bandcamp:album' IE_NAME = 'Bandcamp:album'
_VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^/?#&]+))?' _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<id>[^/?#&]+))?'
_TESTS = [{ _TESTS = [{
'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1', 'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
@ -230,7 +220,10 @@ class BandcampAlbumIE(InfoExtractor):
'info_dict': { 'info_dict': {
'id': '1353101989', 'id': '1353101989',
'ext': 'mp3', 'ext': 'mp3',
'title': 'Intro', 'title': 'Blazo - Intro',
'timestamp': 1311756226,
'upload_date': '20110727',
'uploader': 'Blazo',
} }
}, },
{ {
@ -238,7 +231,10 @@ class BandcampAlbumIE(InfoExtractor):
'info_dict': { 'info_dict': {
'id': '38097443', 'id': '38097443',
'ext': 'mp3', 'ext': 'mp3',
'title': 'Kero One - Keep It Alive (Blazo remix)', 'title': 'Blazo - Kero One - Keep It Alive (Blazo remix)',
'timestamp': 1311757238,
'upload_date': '20110727',
'uploader': 'Blazo',
} }
}, },
], ],
@ -294,41 +290,31 @@ class BandcampAlbumIE(InfoExtractor):
else super(BandcampAlbumIE, cls).suitable(url)) else super(BandcampAlbumIE, cls).suitable(url))
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) uploader_id, album_id = re.match(self._VALID_URL, url).groups()
uploader_id = mobj.group('subdomain')
album_id = mobj.group('album_id')
playlist_id = album_id or uploader_id playlist_id = album_id or uploader_id
webpage = self._download_webpage(url, playlist_id) webpage = self._download_webpage(url, playlist_id)
track_elements = re.findall( tralbum = self._extract_data_attr(webpage, playlist_id)
r'(?s)<div[^>]*>(.*?<a[^>]+href="([^"]+?)"[^>]+itemprop="url"[^>]*>.*?)</div>', webpage) track_info = tralbum.get('trackinfo')
if not track_elements: if not track_info:
raise ExtractorError('The page doesn\'t contain any tracks') raise ExtractorError('The page doesn\'t contain any tracks')
# Only tracks with duration info have songs # Only tracks with duration info have songs
entries = [ entries = [
self.url_result( self.url_result(
compat_urlparse.urljoin(url, t_path), urljoin(url, t['title_link']), BandcampIE.ie_key(),
ie=BandcampIE.ie_key(), str_or_none(t.get('track_id') or t.get('id')), t.get('title'))
video_title=self._search_regex( for t in track_info
r'<span\b[^>]+\bitemprop=["\']name["\'][^>]*>([^<]+)', if t.get('duration')]
elem_content, 'track title', fatal=False))
for elem_content, t_path in track_elements
if self._html_search_meta('duration', elem_content, default=None)]
title = self._html_search_regex(
r'album_title\s*:\s*"((?:\\.|[^"\\])+?)"',
webpage, 'title', fatal=False)
if title:
title = title.replace(r'\"', '"')
return { return {
'_type': 'playlist', '_type': 'playlist',
'uploader_id': uploader_id, 'uploader_id': uploader_id,
'id': playlist_id, 'id': playlist_id,
'title': title, 'title': try_get(tralbum, lambda x: x['current']['title'], compat_str),
'entries': entries, 'entries': entries,
} }
class BandcampWeeklyIE(InfoExtractor): class BandcampWeeklyIE(BandcampIE):
IE_NAME = 'Bandcamp:weekly' IE_NAME = 'Bandcamp:weekly'
_VALID_URL = r'https?://(?:www\.)?bandcamp\.com/?\?(?:.*?&)?show=(?P<id>\d+)' _VALID_URL = r'https?://(?:www\.)?bandcamp\.com/?\?(?:.*?&)?show=(?P<id>\d+)'
_TESTS = [{ _TESTS = [{
@ -343,29 +329,23 @@ class BandcampWeeklyIE(InfoExtractor):
'release_date': '20170404', 'release_date': '20170404',
'series': 'Bandcamp Weekly', 'series': 'Bandcamp Weekly',
'episode': 'Magic Moments', 'episode': 'Magic Moments',
'episode_number': 208,
'episode_id': '224', 'episode_id': '224',
} },
'params': {
'format': 'opus-lo',
},
}, { }, {
'url': 'https://bandcamp.com/?blah/blah@&show=228', 'url': 'https://bandcamp.com/?blah/blah@&show=228',
'only_matching': True 'only_matching': True
}] }]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) show_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, show_id)
blob = self._parse_json( blob = self._extract_data_attr(webpage, show_id, 'blob')
self._search_regex(
r'data-blob=(["\'])(?P<blob>{.+?})\1', webpage,
'blob', group='blob'),
video_id, transform_source=unescapeHTML)
show = blob['bcw_show'] show = blob['bcw_data'][show_id]
# This is desired because any invalid show id redirects to `bandcamp.com`
# which happens to expose the latest Bandcamp Weekly episode.
show_id = int_or_none(show.get('show_id')) or int_or_none(video_id)
formats = [] formats = []
for format_id, format_url in show['audio_stream'].items(): for format_id, format_url in show['audio_stream'].items():
@ -390,20 +370,8 @@ class BandcampWeeklyIE(InfoExtractor):
if subtitle: if subtitle:
title += ' - %s' % subtitle title += ' - %s' % subtitle
episode_number = None
seq = blob.get('bcw_seq')
if seq and isinstance(seq, list):
try:
episode_number = next(
int_or_none(e.get('episode_number'))
for e in seq
if isinstance(e, dict) and int_or_none(e.get('id')) == show_id)
except StopIteration:
pass
return { return {
'id': video_id, 'id': show_id,
'title': title, 'title': title,
'description': show.get('desc') or show.get('short_desc'), 'description': show.get('desc') or show.get('short_desc'),
'duration': float_or_none(show.get('audio_duration')), 'duration': float_or_none(show.get('audio_duration')),
@ -411,7 +379,6 @@ class BandcampWeeklyIE(InfoExtractor):
'release_date': unified_strdate(show.get('published_date')), 'release_date': unified_strdate(show.get('published_date')),
'series': 'Bandcamp Weekly', 'series': 'Bandcamp Weekly',
'episode': show.get('subtitle'), 'episode': show.get('subtitle'),
'episode_number': episode_number, 'episode_id': show_id,
'episode_id': compat_str(video_id),
'formats': formats 'formats': formats
} }