What's new in v2

Installation

# Latest beta
pip install "spotapi==2.0.0b1"

# Or directly from this branch
pip install "git+https://github.com/Aran404/SpotAPI@async-v2"

Quick start

Unauthenticated — public search

import asyncio
import spotapi

async def main() -> None:
    async for page in spotapi.v2.Public.search_tracks("Radiohead"):
        for track in page:
            duration_s = track.duration.total_milliseconds // 1000
            print(f"{track.name}  ({duration_s}s)  —  {track.uri}")

asyncio.run(main())

Search artists with pagination control

import asyncio
import spotapi

async def main() -> None:
    # Yields one page (up to 100 results) at a time
    async for page in spotapi.v2.Public.search_artists("Tame Impala"):
        for artist in page:
            verified = artist.on_platform_reputation_trait.verification.is_verified
            print(f"{artist.profile.name}  verified={verified}")

asyncio.run(main())

Search albums

import asyncio
import spotapi

async def main() -> None:
    async for page in spotapi.v2.Public.search_albums("OK Computer"):
        for album in page:
            artists = ", ".join(a.profile.name for a in album.artists.items)
            print(f"{album.name} ({album.date.year}) — {artists}")

asyncio.run(main())

Search podcasts

import asyncio
import spotapi

async def main() -> None:
    async for page in spotapi.v2.Public.search_podcasts("Lex Fridman"):
        for podcast in page:
            print(f"{podcast.name} by {podcast.publisher.name}")

asyncio.run(main())

Architecture overview

v2/
├── base.py                  # BaseClient — pathfinder query + pagination
├── client.py                # AsyncClient — thin HTTP facade + context manager
├── session.py               # BundleSession + AuthSession — tokens, TOTP, auto-refresh
├── public.py                # Public — typed search generators
│
├── connection/
│   ├── http.py              # HTTPClient — wreq wrapper, retry + backoff, browser emulation
│   ├── websocket.py         # WebSocketClient — heartbeat, event dispatch
│   └── types.py             # ResponseSuccess / ResponseFailure, backoff types
│
├── datastruct/
│   ├── object_dict.py       # ObjectDict — attribute-access dict for raw JSON
│   ├── pool.py              # Pool[T] — generic async object pool
│   └── event_handler.py     # EventDispatcher — async event system
│
├── specialized/
│   ├── totp.py              # TOTP generation — no CAPTCHA solver required
│   └── data_wrappers/
│       ├── _common.py       # Shared color/visual dataclasses
│       ├── track.py         # Track
│       ├── artist.py        # Artist
│       ├── album.py         # Album
│       ├── playlist.py      # Playlist
│       └── podcast.py       # Podcast
│
├── types/
│   ├── logger.py            # LoggerProtocol + 5 concrete loggers
│   └── exceptions.py        # HTTPError, WebsocketError, BaseClientError
│
└── utils/
    ├── cache.py             # timed_cache — TTL-based async/sync cache decorator
    ├── random.py            # Weighted random selection
    └── strings.py           # JS bundle parsing, hash extraction

Logging

v2 ships five ready-to-use loggers. All implement LoggerProtocol so you can drop in your own.

from spotapi.v2.types import LoggerColour, JsonLogger, NoopLogger, MultiLogger, Level

# Pretty coloured output to stderr (default)
logger = LoggerColour(min_level=Level.INFO)

# Structured JSON — good for log aggregation pipelines
json_logger = JsonLogger(min_level=Level.WARNING)

# Fan-out to multiple loggers at once
combined = MultiLogger(logger, json_logger)

# Silence everything
noop = NoopLogger()

# Bind extra context fields that appear on every subsequent record
scoped = logger.bind(request_id="abc123")
scoped.info("Token refreshed", token_version=2)

Connection pool & browser emulation

Every HTTPClient automatically selects a randomised browser profile (Chrome, Edge, Firefox, Opera) weighted towards modern versions, and a randomised OS (Windows 30×, macOS 6×, Linux 2×). A shared ClientPool reuses connections across requests.

from spotapi.v2.connection import HTTPClient, ClientPool

# Get a pooled client (shared across BaseClient instances)
http = await ClientPool.get()

# Or create a dedicated client with a proxy
async with HTTPClient(proxies=[wreq.Proxy.http("http://user:pass@host:port")]) as http:
    response = await http.get("https://open.spotify.com/")

Migration from v1

For a full side-by-side comparison of the v1 and v2 APIs, see MIGRATION.md.

# v1
from spotapi import Song
song = Song()
for batch in song.paginate_songs("weezer"):
    for item in batch:
        print(item["item"]["data"]["name"])

# v2
import spotapi
async for page in spotapi.v2.Public.search_tracks("weezer"):
    for track in page:
        print(track.name)

Contributing

All contributions are welcome. The async rewrite is still in progress and there are plenty of well-scoped tasks available — see CONTRIBUTING.md for the full guide and the open good first issues.

git clone https://github.com/Aran404/SpotAPI
cd SpotAPI
git checkout async-v2
pip install -e ".[dev]"

Legal notice

Disclaimer: This repository is provided for educational purposes only. The author accepts no responsibility for misuse. Accessing private Spotify endpoints may violate Spotify's Terms of Service. Use responsibly. See LEGAL_NOTICE.md for full details.

License

GPL-3.0 © Aran404