KevinMidboe/spotify-downloader
1
0
Fork 0
mirror of https://github.com/KevinMidboe/spotify-downloader.git synced 2025-10-29 18:00:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix token issues and ability to download opus audio and encoding to AAC (#59)

Browse Source 
* Ability to download opus audio and encoding to AAC

Requires ffmpeg with libfdk_aac and mutagen.

* Change to oauth2 to get access token

* Add import spotipy.oauth2

* Final polish before merge
This commit is contained in:
Robert J
2017-06-01 22:28:23 +07:00
committed by Ritiek Malhotra
parent c4e3a9ac82
commit b5e3dbd2e1
2 changed files with 349 additions and 207 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
requirements.txt
View File

@@ -5,3 +5,7 @@ youtube_dl
pafy pafy
spotipy spotipy
eyed3 eyed3
mutagen
unicode-slugify
titlecase

552
spotdl.py
View File

@@ -5,6 +5,9 @@
from bs4 import BeautifulSoup from bs4 import BeautifulSoup
from shutil import copyfileobj from shutil import copyfileobj
from sys import path, version_info from sys import path, version_info
from slugify import slugify
from titlecase import titlecase
from mutagen.mp4 import MP4, MP4Cover
import spotipy import spotipy
import eyed3 import eyed3
import requests import requests
@@ -12,262 +15,397 @@ import pafy
import os import os
import argparse import argparse
import pathlib import pathlib
import spotipy.util as util import spotipy.oauth2 as oauth2
def getInputLink(links): def getInputLink(links):
while True: while True:
try: try:
the_chosen_one = int(raw_input('>> Choose your number: ')) the_chosen_one = int(raw_input('>> Choose your number: '))
if the_chosen_one >= 1 and the_chosen_one <= len(links): if the_chosen_one >= 1 and the_chosen_one <= len(links):
return links[the_chosen_one-1] return links[the_chosen_one - 1]
elif the_chosen_one == 0: elif the_chosen_one == 0:
return None return None
else: else:
print('Choose a valid number!') print('Choose a valid number!')
except ValueError: except ValueError:
print('Choose a valid number!') print('Choose a valid number!')
# Check if input song is Spotify URL or just a song name # Check if input song is Spotify URL or just a song name
def isSpotify(raw_song): def isSpotify(raw_song):
if (len(raw_song) == 22 and raw_song.replace(" ", "%20") == raw_song) or (raw_song.find('spotify') > -1): if (len(raw_song) == 22 and raw_song.replace(" ", "%20")
return True == raw_song) or (raw_song.find('spotify') > -1):
else: return True
return False else:
return False
# [Artist] - [Song Name] # [Artist] - [Song Name]
def generateSongName(raw_song): def generateSongName(raw_song):
if isSpotify(raw_song): if isSpotify(raw_song):
tags = generateMetaTags(raw_song) tags = generateMetaTags(raw_song)
raw_song = tags['artists'][0]['name'] + ' - ' + tags['name'] raw_song = tags['artists'][0]['name'] + ' - ' + tags['name']
return raw_song return raw_song
def generateMetaTags(raw_song): def generateMetaTags(raw_song):
try: try:
if isSpotify(raw_song): if isSpotify(raw_song):
return spotify.track(raw_song) return spotify.track(raw_song)
else: else:
return spotify.search(raw_song, limit=1)['tracks']['items'][0] return spotify.search(raw_song, limit=1)['tracks']['items'][0]
except: except BaseException:
return None return None
def generateSearchURL(song): def generateSearchURL(song):
URL = "https://www.youtube.com/results?sp=EgIQAQ%253D%253D&q=" + song.replace(" ", "%20") URL = "https://www.youtube.com/results?sp=EgIQAQ%253D%253D&q=" + \
return URL song.replace(" ", "%20")
return URL
def generateYouTubeURL(raw_song): def generateYouTubeURL(raw_song):
song = generateSongName(raw_song) song = generateSongName(raw_song)
searchURL = generateSearchURL(song) searchURL = generateSearchURL(song)
items = requests.get(searchURL).text items = requests.get(searchURL).text
items_parse = BeautifulSoup(items, "html.parser") items_parse = BeautifulSoup(items, "html.parser")
check = 1 check = 1
if args.manual: if args.manual:
links = [] links = []
print(song) print(song)
print('') print('')
print('0. Skip downloading this song') print('0. Skip downloading this song')
for x in items_parse.find_all('h3', {'class':'yt-lockup-title'}): for x in items_parse.find_all('h3', {'class': 'yt-lockup-title'}):
if not x.find('channel') == -1 or not x.find('googleads') == -1: if not x.find('channel') == -1 or not x.find('googleads') == -1:
print(str(check) + '. ' + x.get_text()) print(str(check) + '. ' + x.get_text())
links.append(x.find('a')['href']) links.append(x.find('a')['href'])
check += 1 check += 1
print('') print('')
result = getInputLink(links) result = getInputLink(links)
if result == None: if result is None:
return None return None
else: else:
result = items_parse.find_all(attrs={'class':'yt-uix-tile-link'})[0]['href'] result = items_parse.find_all(
while not result.find('channel') == -1 or not result.find('googleads') == -1: attrs={'class': 'yt-uix-tile-link'})[0]['href']
result = items_parse.find_all(attrs={'class':'yt-uix-tile-link'})[check]['href'] while not result.find('channel') == - \
check += 1 1 or not result.find('googleads') == -1:
full_link = "youtube.com" + result result = items_parse.find_all(
return full_link attrs={'class': 'yt-uix-tile-link'})[check]['href']
check += 1
full_link = "youtube.com" + result
return full_link
def goPafy(raw_song): def goPafy(raw_song):
trackURL = generateYouTubeURL(raw_song) trackURL = generateYouTubeURL(raw_song)
if trackURL == None: if trackURL is None:
return None return None
else: else:
return pafy.new(trackURL) return pafy.new(trackURL)
def getYouTubeTitle(content, number): def getYouTubeTitle(content, number):
title = content.title title = content.title
if number == None: if number is None:
return title return title
else: else:
return str(number) + '. ' + title return str(number) + '. ' + title
# Generate name for the song to be downloaded # Generate name for the song to be downloaded
def generateFileName(content): def generateFileName(content):
return fixEncoding((content.title).replace("\\", "_").replace("/", "_").replace(":", "_").replace("*", "_").replace("?", "_").replace('"', "_").replace("<", "_").replace(">", "_").replace("|", "_").replace(" ", "_")) return slugify(content.title)
def downloadSong(content): def downloadSong(content):
music_file = generateFileName(content) music_file = generateFileName(content)
link = content.getbestaudio(preftype="m4a") if args.input_ext == '.webm':
link.download(filepath="Music/" + music_file + ".m4a") link = content.getbestaudio(preftype='webm')
if link is not None:
link.download(filepath='Music/' + music_file + args.input_ext)
else:
link = content.getbestaudio(preftype="m4a")
if link is not None:
link.download(filepath="Music/" + music_file + ".m4a")
def convertToMP3(music_file): def convertToMP3(music_file):
if os.name == 'nt': if os.name == 'nt':
os.system('Scripts\\avconv.exe -loglevel 0 -i "' + 'Music/' + music_file + '.m4a" -ab 192k "' + 'Music/' + music_file + '.mp3"') os.system(
else: 'Scripts\\avconv.exe -loglevel 0 -i "' +
os.system('avconv -loglevel 0 -i "' + 'Music/' + music_file + '.m4a" -ab 192k "' + 'Music/' + music_file + '.mp3"') 'Music/' +
os.remove('Music/' + music_file + '.m4a') music_file +
'.m4a" -ab 192k "' +
'Music/' +
music_file +
'.mp3"')
else:
os.system(
'avconv -loglevel 0 -i "' +
'Music/' +
music_file +
'.m4a" -ab 192k "' +
'Music/' +
music_file +
'.mp3"')
os.remove('Music/' + music_file + '.m4a')
def convertToM4A(music_file):
# Here we prefer downloading .webm (Opus) audio and encode as m4a
# in format prefered by iTunes - AAC (256k)
# We are using ffmpeg with fdk_aac code and cutoff at 18kHz
# python3 spotdl.py -i '.webm' -o '.m4a'
if args.quiet:
ffmpeg_pre = 'ffmpeg -hide_banner -nostats -v panic -y '
else:
ffmpeg_pre = 'ffmpeg -y '
os.system(ffmpeg_pre +
'-i "Music/' + music_file + args.input_ext + '" ' +
'-cutoff 18000 -c:a libfdk_aac -b:a 256k -vn ' +
'"Music/_' + music_file + args.output_ext + '" ')
os.remove('Music/' + music_file + args.input_ext)
def checkExists(music_file, raw_song, islist): def checkExists(music_file, raw_song, islist):
if os.path.exists("Music/" + music_file + ".m4a.temp"): if os.path.exists("Music/" + music_file + ".m4a.temp"):
os.remove("Music/" + music_file + ".m4a.temp") os.remove("Music/" + music_file + ".m4a.temp")
if args.no_convert: if args.no_convert:
extension = '.m4a' extension = args.input_ext
else: else:
if os.path.exists("Music/" + music_file + ".m4a"): if os.path.exists('Music/' + music_file + args.input_ext):
os.remove("Music/" + music_file + ".m4a") os.remove('Music/' + music_file + args.input_ext)
extension = '.mp3' extension = args.output_ext
if os.path.isfile("Music/" + music_file + extension): if os.path.isfile("Music/" + music_file + extension):
if extension == '.mp3': if extension == '.mp3':
audiofile = eyed3.load("Music/" + music_file + extension) audiofile = eyed3.load("Music/" + music_file + extension)
if isSpotify(raw_song) and not audiofile.tag.title == (generateMetaTags(raw_song))['name']: if isSpotify(raw_song) and not audiofile.tag.title == (
os.remove("Music/" + music_file + extension) generateMetaTags(raw_song))['name']:
return False os.remove("Music/" + music_file + extension)
if islist: return False
return True if islist:
else: return True
prompt = raw_input('Song with same name has already been downloaded. Re-download? (y/n): ').lower() else:
if prompt == "y": prompt = raw_input(
os.remove("Music/" + music_file + extension) 'Song with same name has already been downloaded. Re-download? (y/n): ').lower()
return False if prompt == "y":
else: os.remove("Music/" + music_file + extension)
return True return False
else:
return True
# Remove song from list.txt once downloaded # Remove song from list.txt once downloaded
def trimSong(file): def trimSong(file):
with open(file, 'r') as fin: with open(file, 'r') as fin:
data = fin.read().splitlines(True) data = fin.read().splitlines(True)
with open(file, 'w') as fout: with open(file, 'w') as fout:
fout.writelines(data[1:]) fout.writelines(data[1:])
def fixSong(music_file, meta_tags): def fixSong(music_file, meta_tags):
audiofile = eyed3.load("Music/" + music_file + '.mp3') audiofile = eyed3.load("Music/" + music_file + '.mp3')
audiofile.tag.artist = meta_tags['artists'][0]['name'] audiofile.tag.artist = meta_tags['artists'][0]['name']
audiofile.tag.album_artist = meta_tags['artists'][0]['name'] audiofile.tag.album_artist = meta_tags['artists'][0]['name']
audiofile.tag.album = meta_tags['album']['name'] audiofile.tag.album = meta_tags['album']['name']
audiofile.tag.title = meta_tags['name'] audiofile.tag.title = meta_tags['name']
audiofile.tag.track_num = meta_tags['track_number'] audiofile.tag.track_num = meta_tags['track_number']
audiofile.tag.disc_num = meta_tags['disc_number'] audiofile.tag.disc_num = meta_tags['disc_number']
audiofile.tag.release_date = spotify.album(meta_tags['album']['id'])['release_date'] audiofile.tag.release_date = spotify.album(
albumart = (requests.get(meta_tags['album']['images'][0]['url'], stream=True)).raw meta_tags['album']['id'])['release_date']
with open('last_albumart.jpg', 'wb') as out_file: albumart = (
copyfileobj(albumart, out_file) requests.get(
albumart = open("last_albumart.jpg", "rb").read() meta_tags['album']['images'][0]['url'],
audiofile.tag.images.set(3,albumart,"image/jpeg") stream=True)).raw
audiofile.tag.save(version=(2,3,0)) with open('last_albumart.jpg', 'wb') as out_file:
copyfileobj(albumart, out_file)
albumart = open("last_albumart.jpg", "rb").read()
audiofile.tag.images.set(3, albumart, "image/jpeg")
audiofile.tag.save(version=(2, 3, 0))
def fixSongM4A(music_file, meta_tags):
# eyed serves only mp3 not aac so using mutagen
# Apple has specific tags - see mutagen docs -
# http://mutagen.readthedocs.io/en/latest/api/mp4.html
tags = {'album': '\xa9alb',
'artist': '\xa9ART',
'year': '\xa9day',
'title': '\xa9nam',
'comment': '\xa9cmt',
'group': '\xa9grp',
'writer': '\xa9wrt',
'genre': '\xa9gen',
'track': 'trkn',
'aart': 'aART',
'disk': 'disk',
'cpil': 'cpil',
'tempo': 'tmpo'}
audiofile = MP4('Music/_' + music_file + args.output_ext)
audiofile[tags['artist']] = meta_tags['artists'][0]['name']
audiofile[tags['album']] = meta_tags['album']['name']
audiofile[tags['title']] = meta_tags['name']
artist = spotify.artist(meta_tags['artists'][0]['id'])
audiofile[tags['genre']] = titlecase(artist['genres'][0])
album = spotify.album(meta_tags['album']['id'])
audiofile[tags['year']] = album['release_date']
audiofile[tags['track']] = [(meta_tags['track_number'], 0)]
albumart = (
requests.get(meta_tags['album']['images'][0]['url'], stream=True)).raw
with open('last_albumart.jpg', 'wb') as out_file:
copyfileobj(albumart, out_file)
with open("last_albumart.jpg", "rb") as f:
audiofile["covr"] = [
MP4Cover(
f.read(),
imageformat=MP4Cover.FORMAT_JPEG)]
audiofile.save()
# Logic behind preparing the song to download to finishing meta-tags # Logic behind preparing the song to download to finishing meta-tags
def grabSingle(raw_song, number=None): def grabSingle(raw_song, number=None):
if number: if number:
islist = True islist = True
else: else:
islist = False islist = False
content = goPafy(raw_song) content = goPafy(raw_song)
if content == None: if content is None:
return return
print(getYouTubeTitle(content, number)) print(getYouTubeTitle(content, number))
music_file = generateFileName(content) music_file = generateFileName(content)
if not checkExists(music_file, raw_song, islist=islist): if not checkExists(music_file, raw_song, islist=islist):
downloadSong(content) downloadSong(content)
print('') print('')
if not args.no_convert: if not args.no_convert:
print('Converting ' + music_file + '.m4a to mp3') print('Converting ' + music_file + '.m4a to mp3')
convertToMP3(music_file) if args.output_ext == '.m4a':
meta_tags = generateMetaTags(raw_song) convertToM4A(music_file)
if not meta_tags == None: meta_tags = generateMetaTags(raw_song)
print('Fixing meta-tags') if meta_tags is not None:
fixSong(music_file, meta_tags) print('Fixing meta-tags')
fixSongM4A(music_file, meta_tags)
else:
convertToMP3(music_file)
meta_tags = generateMetaTags(raw_song)
if meta_tags is not None:
print('Fixing meta-tags')
fixSong(music_file, meta_tags)
# Fix python2 encoding issues # Fix python2 encoding issues
def fixEncoding(query): def fixEncoding(query):
if version_info > (3,0): if version_info > (3, 0):
return query return query
else: else:
return query.encode('utf-8') return query.encode('utf-8')
def grabList(file): def grabList(file):
lines = open(file, 'r').read() lines = open(file, 'r').read()
lines = lines.splitlines() lines = lines.splitlines()
# Ignore blank lines in list.txt (if any) # Ignore blank lines in list.txt (if any)
try: try:
lines.remove('') lines.remove('')
except ValueError: except ValueError:
pass pass
print('Total songs in list = ' + str(len(lines)) + ' songs') print('Total songs in list = ' + str(len(lines)) + ' songs')
print('') print('')
# Count the number of song being downloaded # Count the number of song being downloaded
number = 1 number = 1
for raw_song in lines: for raw_song in lines:
try: try:
grabSingle(raw_song, number=number) grabSingle(raw_song, number=number)
trimSong(file) trimSong(file)
number += 1 number += 1
print('') print('')
except KeyboardInterrupt: except KeyboardInterrupt:
graceQuit() graceQuit()
except: except BaseException:
lines.append(raw_song) lines.append(raw_song)
trimSong(file) trimSong(file)
with open(file, 'a') as myfile: with open(file, 'a') as myfile:
myfile.write(raw_song) myfile.write(raw_song)
print('Failed to download song. Will retry after other songs.') print('Failed to download song. Will retry after other songs.')
def getArgs(argv=None):
parser = argparse.ArgumentParser(description='Find and convert songs \
from Spotify, Youtube etc.',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser = argparse.ArgumentParser()
parser.add_argument('-n', '--no-convert', default=False,
help='Skip the conversion process and meta-tags', action='store_true')
parser.add_argument('-m', '--manual', default=False,
help='Choose the song to download manually', action='store_true')
parser.add_argument('-l', '--list', default=False,
help='Download songs present in list.txt', action='store_true')
parser.add_argument('-q', '--quiet', default=False,
help='Spare us output of ffmpeg conversion', action='store_true')
parser.add_argument('-i', '--input_ext', default='.m4a',
help='Prefered input format .m4a or .webm (Opus)')
parser.add_argument('-o', '--output_ext', default='.mp3',
help='Prefered output extension .mp3 or .m4a (AAC)')
return parser.parse_args(argv)
def graceQuit(): def graceQuit():
print('') print('')
print('') print('')
print('Exitting..') print('Exitting..')
exit() exit()
# Python 3 compatibility
if version_info > (3,0):
raw_input = input
eyed3.log.setLevel("ERROR") if __name__ == '__main__':
os.chdir(path[0]) # Python 3 compatibility
if version_info > (3, 0):
raw_input = input
if not os.path.exists("Music"): eyed3.log.setLevel("ERROR")
os.makedirs("Music")
open('list.txt', 'a').close()
# Please respect this user token :) os.chdir(path[0])
token = util.prompt_for_user_token('Spotify-Downloader',
scope = 'playlist-read-private',
client_id='4fe3fecfe5334023a1472516cc99d805',
client_secret='0f02b7c483c04257984695007a4a8d5c',
redirect_uri='http://localhost:9876/callback')
spotify = spotipy.Spotify(auth=token) if not os.path.exists("Music"):
os.makedirs("Music")
open('list.txt', 'a').close()
# Set up arguments # Please respect this user token :)
parser = argparse.ArgumentParser() oauth2 = oauth2.SpotifyClientCredentials(client_id='4fe3fecfe5334023a1472516cc99d805',
parser.add_argument("-n", "--no-convert", help="skip the conversion process and meta-tags", action="store_true") client_secret='0f02b7c483c04257984695007a4a8d5c')
parser.add_argument("-m", "--manual", help="choose the song to download manually", action="store_true") token = oauth2.get_access_token()
parser.add_argument("-l", "--list", help="download songs present in list.txt", action="store_true")
args = parser.parse_args()
if args.no_convert: spotify = spotipy.Spotify(auth=token)
# Set up arguments
args = getArgs()
if args.no_convert:
print("-n, --no-convert skip the conversion process and meta-tags") print("-n, --no-convert skip the conversion process and meta-tags")
if args.manual: if args.manual:
print("-m, --manual choose the song to download manually") print("-m, --manual choose the song to download manually")
print('') print('')
if args.list: if args.list:
grabList(file='list.txt') grabList(file='list.txt')
exit() exit()
while True: while True:
for temp in os.listdir('Music/'): for temp in os.listdir('Music/'):
if temp.endswith('.m4a.temp'): if temp.endswith('.m4a.temp'):
os.remove('Music/' + temp) os.remove('Music/' + temp)
try: try:
print('Enter a Spotify URL or Song Name: ') print('Enter a Spotify URL or Song Name: ')
command = raw_input('>> ') command = raw_input('>> ')
print('') print('')
grabSingle(raw_song=command) grabSingle(raw_song=command)
print('') print('')
except KeyboardInterrupt: except KeyboardInterrupt:
graceQuit() graceQuit()