Skip to content

tests_overview.md

Latest commit

 

History

History
221 lines (174 loc) · 9.97 KB

tests_overview.md

File metadata and controls

221 lines (174 loc) · 9.97 KB
Raw

Tests Overview

Running Tests

make test                                                        # full suite
pytest -s tests/api/test_usdc.py                            # single file
pytest -s tests/api/test_usdc.py::test_mint_usdc            # single test
pytest -s -m integration                                         # live network only

All tests use pytest. pytest.ini streams logs at INFO level on every run — no -v flag needed.

Test layers:

                    ┌──────────────────────────┐
                    │   Integration (live net)  │  @pytest.mark.integration
                    │   test_polymarket_sync    │  hits Gamma API + Polygon RPC
                    │   test_conditional_token  │  not run by default
                    └──────────────────────────┘
              ┌──────────────────────────────────────────┐
              │        FastAPI / HTTP layer               │  make test
              │  [TestClient](https://fastapi.tiangolo.com/tutorial/testing/) + in-memory SQLite │
              │  test_usdc · test_positions · test_markets│
              │  test_lifecycle · test_resolution · etc.  │
              └──────────────────────────────────────────┘
        ┌──────────────────────────────────────────────────────┐
        │             py_clob_client utilities                  │  make test
        │  test_utilities.py — orderbook parsing, order_to_json │
        └──────────────────────────────────────────────────────┘

Layout

tests/
├── test_utilities.py           # py_clob_client utility helpers
├── fastapi/                    # AgentPit HTTP API (FastAPI TestClient)
│   ├── test_basic.py           # GET /
│   ├── test_create_user.py     # POST /create_user
│   ├── test_personality.py     # POST /create_personality
│   ├── test_create_agent.py    # POST /create_agent
│   ├── test_markets.py         # GET + POST /markets, GET /markets/{id}
│   ├── test_usdc.py            # mint, balance, transfer
│   ├── test_positions.py       # split_position, merge_positions
│   ├── test_resolution.py      # resolve + redeem_position
│   ├── test_lifecycle.py       # state machine transitions + cancel
│   ├── test_history.py         # transaction history
│   ├── test_portfolio.py       # portfolio summary
│   └── test_main.py            # backward-compat re-export of test_portfolio
└── polymarket/
    ├── test_polymarket_sync.py              # Gamma API sync (live)
    └── test_conditional_token_framework.py  # CTF reads (live Polygon RPC)

Infrastructure

FastAPI tests — TestClient + in-memory SQLite

from fastapi.testclient import TestClient
from agentpit.api import main

def test_something():
    with TestClient(main.app) as client:
        resp = client.post("/mint_usdc", json={"api_key": "k", "amount": 100})
        assert resp.status_code == 200

Each with TestClient(...) block gets a fresh in-memory SQLite DB (default when AGENTPIT_DB_PATH is unset).

test_portfolio.py uses an explicit fixture for finer isolation:

@pytest.fixture
def client():
    server = AgentPitServer(db_path=":memory:")
    with TestClient(server) as tc:
        yield tc
    server.shutdown()

Integration tests — live network required

Gated with @pytest.mark.integration. Not run by default.

File Calls
tests/polymarket/test_polymarket_sync.py https://gamma-api.polymarket.com
tests/polymarket/test_conditional_token_framework.py Polygon RPC via Tenderly 
pytest -s -m integration tests/polymarket/

FastAPI Test Details

test_basic.py

Test Checks
test_read_root_returns_version GET / → {"version": "1.0"}

test_create_user.py

Test Checks
test_create_user Returns user_id, UUID api_key, 0x-prefixed eth_address
test_create_user_duplicate Same user_id twice → 409
test_create_user_invalid_handle >15 chars, empty, spaces/special chars → 400
test_create_user_missing_field Missing user_id → 422
test_create_multiple_users_unique_keys 3 users → all api_key and eth_address values unique

test_personality.py

Tests POST /create_personality — the endpoint that registers an OpenClaw agent personality (the strategy spec OpenClaw uses to drive an agent's decisions: beliefs, methods, needs).

Test Checks
test_create_personality Returns all fields; spec contains beliefs, methods, needs
test_create_personality_missing_field Missing needs → 422
test_create_personality_empty_title Empty title → 400 via check_state

test_create_agent.py

Tests POST /create_agent — the endpoint that instantiates an OpenClaw agent by linking an agent_id to an existing personality. AgentPit persists the agent's state, history, and todo so OpenClaw can maintain continuity across sessions.

Test Checks
test_create_agent Returns agent_id, personality_id, empty state/history/todo
test_create_agent_duplicate Same agent_id twice → 409
test_create_agent_missing_personality Non-existent personality_id → 404

test_markets.py

Test Checks
test_create_and_get_market Create market → GET /markets/{id} returns all fields; GET /markets/9999 → 404
test_list_markets Empty → total=0; 5 created → total=5; limit=2 → 2 results; limit=2&offset=2 → correct page

test_usdc.py

Test Checks
test_mint_usdc First mint correct; second mint accumulates
test_get_usdc_balance 0 before mint; correct after; same eth_address on both calls
test_transfer_usdc Sender decreases, receiver increases; response fields correct
test_transfer_usdc_insufficient_balance Transfer > balance → 400

test_positions.py

Test Checks
test_split_and_merge_positions Split 100: USDC −100, Yes/No +100. Merge 50: USDC +50, tokens −50
test_split_position_insufficient_usdc Split with zero USDC → 400
test_merge_positions_insufficient_tokens Merge more than held → 400

test_resolution.py

Test Checks
test_resolve_market_and_redeem Two users split → resolve (Yes wins) → winner redeems 100 USDC, loser 0; all tokens burned

test_lifecycle.py

Test Checks
test_cancel_market Cancel from DRAFT → state = CANCELLED, refunds_processed = 0
test_market_lifecycle_happy_path DRAFT → ACTIVE → CLOSED → RESOLVED; each transition verified
test_cancel_market_with_positions User splits 50 sets, cancel → 50 USDC refunded, refunds_processed = 1
test_invalid_state_transitions Every illegal transition → 400 with descriptive message

test_history.py

Test Checks
test_get_transaction_history Empty on first call; split + merge → 2 transactions; resolve + redeem → 3rd with payout_usdc > 0

test_portfolio.py

Test Checks
test_get_portfolio_new_user usdc_balance = 0, positions = []
test_get_portfolio_with_usdc After minting 500: usdc_balance = 500, no positions
test_get_portfolio_after_split USDC decreases; positions populated with correct outcome_label, balance, market_id
(additional) Multi-market positions, zero-balance tokens excluded, post-merge update

Polymarket Integration Tests

test_polymarket_sync.py

test_sync_polymarket_markets_syncs_real_markets_to_db — sync inserts >5 markets; DB contents match returned objects; re-run is idempotent.

test_conditional_token_framework.py

test_get_onchain_resolution_status_real_resolved_market — known resolved condition (0xe3b423…) has slot_count = 2, resolved = True, winner_index = 1.

Coverage Map

Component Test Files
AgentPitServer tests/api/test_*.py
ERC20Simulator / ERC1155Simulator test_usdc.py, test_positions.py, test_resolution.py, test_lifecycle.py
Market state machine test_lifecycle.py, test_markets.py
Transaction history test_history.py
Portfolio test_portfolio.py
User / OpenClaw Agent / Personality CRUD test_create_user.py, test_personality.py, test_create_agent.py
Polymarket sync tests/polymarket/test_polymarket_sync.py
CTF on-chain reads tests/polymarket/test_conditional_token_framework.py

Not covered:

  • OrderService — only exercised by tests/onchain/test_trade_flow.py against a real Anvil chain; no isolated unit tests for the matching loop, balance pre-flight, or _settle_on_chain failure path
  • PredictionMarket.split/merge — server calls simulators directly, bypassing this class

See Also