Gets a user's Discogs collection and places all videos from release pages into a YouTube playlist

discogs_collection_to_youtube_playlist.py

import requests
import json
import re
from more_itertools import unique_everseen
import time
from oauth2client import client, GOOGLE_TOKEN_URI, GOOGLE_REVOKE_URI
import httplib2

# INPUTS

# get your Discogs API token from https://www.discogs.com/settings/developers
username = "YOUR_DISCOGS_USERNAME"
token = "YOUR_DISCOGS_API_TOKEN"

# must be larger than your discogs collection
items_per_page = 1000

# choose a unique app name for the user agent
app_name = "YOUR_APP_NAME"

# playlist ID can be found in your playlist URL
playlist_id = "YOUR_PLAYLIST_ID"

# this info can be obtained from the client_secrets file from the Google API console
client_id = "YOUR_CLIENT_ID"
client_secret = "YOUR_CLIENT_SECRET"
refresh_token = "YOUR_REFRESH_TOKEN"

path = "https://api.discogs.com"
headers = {"User-Agent": "{app_name}/1.0".format(app_name=app_name)}
sleep = 2
collection_path = "/users/{username}/collection/folders/0/releases".format(username=username)


collection_url = "{path}{collection_path}?token={token}&per_page={per_page}".format(
    path=path,
    collection_path=collection_path,
    token=token,
    per_page=items_per_page)


releases = requests.get(collection_url).json()["releases"]

with open("full_collection.json", "w") as fp:
    json.dump(releases, fp)

release_ids = [release["id"] for release in releases]

full_collection_details = []
video_urls = []

for release_id in release_ids:
    print("Getting videos for {id}".format(id=release_id))
    release_url = "{path}/releases/{id}?token={token}".format(
        path=path,
        id=release_id,
        token=token)
    try:
        release_json = requests.get(release_url).json()
        release_videos = release_json["videos"]
        release_urls = [video["uri"] for video in release_videos]
        video_urls += release_urls
    except KeyError:
        None
    time.sleep(sleep)

# serialise full collection info
with open("full_collection_details.json", "w") as fp:
    json.dump(full_collection_details, fp)


def get_id(url):
    if "youtube" in url:
        if "attribution" in url:
            try:
                video_id = re.search("%3D(.*)\%26", url).group(1)
            except Exception:
                print(url)
                video_id = ""
        else:
            try:
                video_id = re.search("(\?|\&)(v|ci)=(.*)", url).group(3)
            except Exception:
                print(url)
                video_id = ""
    else:
        try:
            video_id = re.search("\.be\/(.*)", url).group(1)
        except Exception:
                print(url)
                video_id = ""
    return video_id


ids = [get_id(url) for url in video_urls]
ids = list(filter(lambda x: len(x) > 0, ids))
ids = list(unique_everseen(ids))

credentials = client.OAuth2Credentials(
    None, client_id, client_secret, refresh_token, None, GOOGLE_TOKEN_URI,
    None, revoke_uri=GOOGLE_REVOKE_URI)

credentials.refresh(httplib2.Http())
access_token = credentials.get_access_token()[0]

path = "https://www.googleapis.com/youtube/v3/playlistItems?part=snippet&access_token={access_token}".format(access_token=access_token)

for video_id in ids:
    payload = { 
        "snippet": {
            "playlistId": playlist_id,
            "resourceId": {
                "videoId": video_id,
                "kind": "youtube#video"
            }
        }
    }

    request = requests.post(
        path,
        data=json.dumps(payload),
        headers={'Content-Type': 'application/json'})