liquipydia 0.1.1

Python client library for the Liquipedia API (LPDB v3).

LIQUIPYDIA | Python client library for Liquipedia API (LPDB v3)

About

liquipydia aims to be a modern, typed Python wrapper for the Liquipedia Database (LPDB) REST API v3, covering all 16 data types (matches, tournaments, teams, players, transfers, and more).

Built with httpx and pydantic.

Status: Beta — see the Roadmap for progress.

Project Structure

liquipydia/
├── __init__.py       # Package exports, version
├── _client.py        # Core HTTP client (LiquipediaClient)
├── _models.py        # Pydantic models (one per LPDB data type)
├── _resources.py     # Resource classes (one per LPDB data type)
├── _exceptions.py    # Exception hierarchy
├── _response.py      # API response wrapper
└── py.typed          # PEP 561 type marker

API Access

Using this library requires an API key from Liquipedia. Access is not self-service — you must request it through their contact form.

Free access is available for educational, non-commercial open-source, and community projects. Paid plans (Basic, Premium, Enterprise) are available for commercial use.

Installation

# With uv (recommended)
uv add liquipydia

# With pip
pip install liquipydia

Or install from source:

# With uv
uv add git+https://github.com/Dyl-M/liquipydia.git

# With pip
pip install git+https://github.com/Dyl-M/liquipydia.git

Quick Start

from liquipydia import LiquipediaClient, Player, Match

with LiquipediaClient("my-app", api_key="your-api-key") as client:
    # Query players from a specific wiki
    response = client.players.list("dota2", conditions="[[name::Miracle-]]")
    for record in response.result:
        player = Player.model_validate(record)
        print(player.name, player.nationality, player.birthdate)

    # Automatic pagination across multiple pages
    for record in client.matches.paginate("counterstrike", page_size=100, max_results=500):
        match = Match.model_validate(record)
        print(match.match2id, match.date, match.winner)

    # Keyword filters — no need to write raw LPDB conditions.
    # Operator prefixes (`!`, `<`, `>`) pass through, so `earnings=">10000"` becomes `[[earnings::>10000]]`.
    response = client.players.list("rocketleague", pagename="Zen")

    # Match-specific parameters (stream data)
    response = client.matches.list("rocketleague", rawstreams=True, streamurls=True)

    # Team template lookup
    response = client.team_templates.get("dota2", "teamliquid")

Miracle- Jordan 1997-06-20
0042_R01-M001 2025-06-15 14:00:00 1
...

Available Resources

All 16 LPDB v3 data types are accessible as client attributes:

Attribute	Endpoint	Notes
client.broadcasters	/broadcasters
client.companies	/company
client.datapoints	/datapoint
client.external_media_links	/externalmedialink
client.matches	/match	Extra: rawstreams, streamurls
client.placements	/placement
client.players	/player
client.series	/series
client.squad_players	/squadplayer
client.standings_entries	/standingsentry
client.standings_tables	/standingstable
client.teams	/team
client.tournaments	/tournament
client.transfers	/transfer
client.team_templates	/teamtemplate	Uses get() instead of list()
client.team_template_list	/teamtemplatelist	Different params (pagination)

The API key can also be set via the LIQUIPEDIA_API_KEY environment variable.

Typed Models

Each data type has a corresponding Pydantic model for typed access and IDE autocompletion:

Broadcaster, Company, Datapoint, ExternalMediaLink, Match, Placement, Player, Series, SquadPlayer, StandingsEntry, StandingsTable, Team, TeamTemplate, TeamTemplateList, Tournament, Transfer

from liquipydia import Player

player = Player.model_validate(record)
print(player.name)  # str | None — with autocompletion
print(player.birthdate)  # date | None — null sentinels auto-converted
print(player.links)  # dict | None — empty API lists auto-converted

Miracle-
1997-06-20
{'twitter': 'https://twitter.com/Aborss', ...}

Models handle LPDB quirks automatically:

• Null date sentinels ("0000-01-01", "") → None
• Empty dict placeholders ([]) → None
• Unknown/future API fields are preserved (extra="allow")

Documentation

Full documentation (getting started, examples, API reference) is available at dyl-m.github.io/liquipydia.

Development

# Clone the repository
git clone https://github.com/Dyl-M/liquipydia.git
cd liquipydia

# Install dependencies (requires uv)
uv sync --group dev

# Run linting
uv run ruff check .
uv run ruff format --check .

# Run type checking
uv run mypy liquipydia

# Run tests
uv run pytest

License

Code is licensed under the MIT License.

Data License

Data returned by the Liquipedia API is subject to CC-BY-SA 3.0 as required by Liquipedia's API Terms of Use. If you redistribute or display data obtained through this library, you must comply with the CC-BY-SA 3.0 attribution requirements.

Contributing

See CONTRIBUTING.md for guidelines.

Security

See SECURITY.md for reporting vulnerabilities.