conversesamplelibrary.com downloader

README.md

converse-dl

This a hacky script to download a bunch of samples from https://www.conversesamplelibrary.com/.

Useful info: https://www.reddit.com/r/edmproduction/comments/2ux1bq/converse_has_released_their_new_rubber_tracks/cocwtga/

HELP

usage: converse-dl.py [-h] [-a ACCESS_TOKEN] [-j JSON] [-t FILE_TYPE] [-w]
                   [-f FILTER_BY]

optional arguments:
  -h, --help            show this help message and exit
  -a ACCESS_TOKEN, --access-token ACCESS_TOKEN
                        Access Token required to download WAV files
  -j JSON, --json JSON  JSON with samples
  -t FILE_TYPE, --file-type FILE_TYPE
                        File type: 'mp3' or 'wav'. Default: mp3
  -w, --request-wav     When dowloading wav files, it requires an extra
                        request which provides the actual wav file url, this
                        option forces that extra request.
  -f FILTER_BY, --filter-by FILTER_BY
                        Filters. format: 'artist:dreamers,type:one_shot'

converse-dl.py

#!/usr/bin/env python3

import sys
import json
import argparse
import requests


HOST_STORAGE = "https://d34x6xks9kc6p2.cloudfront.net/"
WAV_URL = "https://www.conversesamplelibrary.com/motian_auth/samples/{sample_id}/download?access_token={access_token}"


def download_mp3(record):
    return "{}{}".format(
        HOST_STORAGE,
        record.get('s3_key').replace('.wav', '.mp3')
    )

def download_wav(record, access_token):
    return WAV_URL.format(
        sample_id=record.get('_id'),
        access_token=access_token
    )

def parse_json(json_file):
    data = []
    with open(json_file, 'r') as j:
        data = json.load(j)

    return data


def simplify(thing):
    return "{}".format(thing).lower()

def matches(value, search):
    result = False
    search = simplify(search)
    if isinstance(value, list):
        if search in map(lambda s: simplify(s), value):
            result = True
    else:
        if simplify(value) == search:
            result = True

    return result

def get_results(json_file, filter_by):
    data = parse_json(json_file)
    results = []
    for record in data:
        if all( [matches(record.get(f), s) for f,s in filter_by] ):
            results += [record]

    return results

def start(args):
    filter_by = [(f,s) for f,s in [filt.split(':') for filt in args.filter_by.split(",")]]
    file_type = args.file_type
    if not args.access_token:
        file_type = "mp3"

    if args.json:
        results = get_results(args.json, filter_by)
        for r in results:
            if args.file_type.lower() == "wav":
                wav = download_wav(r, args.access_token)
                if args.request_wav:
                    req = requests.get(wav)
                    print (req.json().get('download_url'))
                else:
                    print (wav)

            if args.file_type.lower() == "mp3":
                print (download_mp3(r))


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument('-a', '--access-token',
                        type=str,
                        help="Access Token required to download WAV files")
    parser.add_argument('-j', '--json',
                        type=str,
                        help="JSON with samples")
    parser.add_argument('-t', '--file-type',
                        type=str,
                        help="File type: 'mp3' or 'wav'. Default: mp3",
                        default="mp3")
    parser.add_argument('-w', '--request-wav',
                        action="store_true",
                        help="When dowloading wav files, it requires an extra request which provides the actual wav file url, this option forces that extra request.",
                        default=False)
    parser.add_argument('-f', '--filter-by',
                        type=str,
                        help="Filters. format: 'artist:dreamers,type:one_shot'")

    args = parser.parse_args()
    start(args)