7bb6fc417b
Some checks are pending
CI / lint-and-test (push) Waiting to run
* feat(kb): apply_proposal engine to land approved proposals into canonical Stage 2 of the KB apply pipeline (approve -> APPLY -> render -> surface). Turns an approved kb_stage.kb_proposals row into canonical public.* rows and flips the ledger to 'applied' in one verified transaction. - Connects as the narrow kb_apply role (never superuser): writes only strategies, strategy_nodes, claim_evidence, claim_edges + kb_proposals ledger. Enforces "agents propose, do not self-apply" at the DB boundary. - Per-type handlers: revise_strategy (versioned strategy + node replace), add_edge, attach_evidence (requires existing source_id; source minting is intentionally out of scope for kb_apply's grants). - Strict apply_payload contract (v1); freeform eval packets are normalized upstream, not applied directly. - --dry-run prints exact SQL; idempotent (refuses non-approved / already-applied); transactional with an in-txn DO-block invariant check that rolls back on failure. - Unit tests cover SQL builders, validation, dispatch, and status guards. * fix(kb): rowcount=1 apply guard + real applied_by FK stamp Closes the three draft-exit review items on the apply engine: - Ledger flip now runs in a DO block asserting exactly one 'approved' row moved to 'applied' (GET DIAGNOSTICS row_count). Closes the concurrent double-apply race — load_proposal (read) and the flip (write) are separate statements, so a row lock cannot span them; only one concurrent apply can match status='approved', so rowcount=1 is the authoritative guard. A loser RAISEs and the whole txn rolls back. - applied_by_agent_id is stamped as a real FK resolved from public.agents by handle, defaulting to the kb-apply service agent — no more NULL FK, no backfill needed. - scripts/kb_apply_prereqs.sql: one-time superuser bootstrap — inserts the kb-apply service-agent row (kb_apply never gets INSERT on agents), grants kb_apply SELECT on public.agents, and ensures the one-active-strategy unique index (idempotent; already present on prod). 18/18 unit tests pass. * fix(kb): hard-resolve applied_by handle, RAISE on NULL FK Resolve applied_by into a variable and assert NOT NULL before the ledger flip, instead of an inline subselect that silently stamps a NULL applied_by_agent_id on an unresolved handle. Since the FK is ON DELETE SET NULL, a bad handle (typo/unseeded agent) was a legal silent NULL -- the perpetually-NULL FK we eliminated. Unresolved handle now hard-fails -> rollback. Non-default --applied-by (operator, future drafters) is the path that goes through the lookup and could strand NULL.
198 lines
7 KiB
Python
198 lines
7 KiB
Python
"""Unit tests for scripts/apply_proposal.py.
|
|
|
|
Pure tests: exercise the SQL builders, payload validation, and dispatch without
|
|
a live database. Runnable under pytest or standalone (`python3 tests/test_apply_proposal.py`).
|
|
"""
|
|
|
|
from pathlib import Path
|
|
import sys
|
|
|
|
try:
|
|
import pytest
|
|
except ImportError: # standalone fallback so the file runs without pytest installed
|
|
import contextlib
|
|
import types
|
|
|
|
pytest = types.SimpleNamespace()
|
|
|
|
@contextlib.contextmanager
|
|
def _raises(exc):
|
|
try:
|
|
yield
|
|
except exc:
|
|
return
|
|
raise AssertionError(f"expected {exc.__name__} to be raised")
|
|
|
|
pytest.raises = _raises
|
|
|
|
REPO_ROOT = Path(__file__).resolve().parents[1]
|
|
sys.path.insert(0, str(REPO_ROOT / "scripts"))
|
|
|
|
import apply_proposal as ap # noqa: E402
|
|
|
|
|
|
# --- literals ------------------------------------------------------------- #
|
|
def test_sql_literal_escapes_single_quotes():
|
|
assert ap.sql_literal("O'Brien") == "'O''Brien'"
|
|
|
|
|
|
def test_sql_literal_none_is_null():
|
|
assert ap.sql_literal(None) == "null"
|
|
|
|
|
|
def test_sql_literal_bool_and_number():
|
|
assert ap.sql_literal(True) == "true"
|
|
assert ap.sql_literal(0.5) == "0.5"
|
|
|
|
|
|
# --- revise_strategy ------------------------------------------------------ #
|
|
def _revise_payload():
|
|
return {
|
|
"apply_payload": {
|
|
"agent_id": "11111111-1111-1111-1111-111111111111",
|
|
"strategy": {
|
|
"diagnosis": "Species-level inflection.",
|
|
"guiding_policy": "Build Teleo as a multi-agent public intelligence.",
|
|
"proximate_objectives": ["route before creating", "keep KB calibrated"],
|
|
},
|
|
"strategy_nodes": [
|
|
{"node_type": "diagnosis", "title": "Inflection", "body": "b1", "rank": 1},
|
|
{"node_type": "policy", "title": "Multi-agent PI", "body": "b2", "rank": 1},
|
|
],
|
|
}
|
|
}
|
|
|
|
|
|
def test_revise_strategy_sql_shape():
|
|
sql = ap.build_revise_strategy_sql(_revise_payload()["apply_payload"], "pid-1", "m3ta")
|
|
assert sql.startswith("begin;")
|
|
assert sql.rstrip().endswith("commit;")
|
|
# deactivate old, insert versioned active, retire nodes, insert new nodes
|
|
assert "update public.strategies" in sql
|
|
assert "active = false" in sql
|
|
assert "coalesce(max(version), 0) + 1" in sql
|
|
assert "update public.strategy_nodes" in sql
|
|
assert "insert into public.strategy_nodes" in sql
|
|
# ledger + invariant
|
|
assert "status = 'applied'" in sql
|
|
assert "exactly one active strategy" in sql
|
|
# proximate_objectives rendered as jsonb
|
|
assert "::jsonb" in sql
|
|
|
|
|
|
def test_revise_strategy_requires_agent_id():
|
|
payload = _revise_payload()["apply_payload"]
|
|
del payload["agent_id"]
|
|
with pytest.raises(ValueError):
|
|
ap.build_revise_strategy_sql(payload, "pid-1", None)
|
|
|
|
|
|
def test_revise_strategy_rejects_bad_node_type():
|
|
payload = _revise_payload()["apply_payload"]
|
|
payload["strategy_nodes"][0]["node_type"] = "manifesto"
|
|
with pytest.raises(ValueError):
|
|
ap.build_revise_strategy_sql(payload, "pid-1", None)
|
|
|
|
|
|
# --- add_edge ------------------------------------------------------------- #
|
|
def test_add_edge_sql_dedup_guard():
|
|
payload = {"from_claim": "aaaa", "to_claim": "bbbb", "edge_type": "supersedes", "weight": 0.9}
|
|
sql = ap.build_add_edge_sql(payload, "pid-2", None)
|
|
assert "insert into public.claim_edges" in sql
|
|
assert "not exists" in sql
|
|
assert "::edge_type" in sql
|
|
assert "status = 'applied'" in sql
|
|
|
|
|
|
def test_add_edge_rejects_self_loop():
|
|
payload = {"from_claim": "same", "to_claim": "same", "edge_type": "supports"}
|
|
with pytest.raises(ValueError):
|
|
ap.build_add_edge_sql(payload, "pid-2", None)
|
|
|
|
|
|
# --- attach_evidence ------------------------------------------------------ #
|
|
def test_attach_evidence_sql_with_source_id():
|
|
payload = {"evidence": [{"claim_id": "cccc", "source_id": "dddd", "role": "grounds", "weight": 0.78}]}
|
|
sql = ap.build_attach_evidence_sql(payload, "pid-3", None)
|
|
assert "insert into public.claim_evidence" in sql
|
|
assert "::evidence_role" in sql
|
|
assert "not exists" in sql
|
|
|
|
|
|
def test_attach_evidence_requires_source_id():
|
|
payload = {"evidence": [{"claim_id": "cccc"}]}
|
|
with pytest.raises(ValueError):
|
|
ap.build_attach_evidence_sql(payload, "pid-3", None)
|
|
|
|
|
|
# --- dispatch + guards ---------------------------------------------------- #
|
|
def test_build_apply_sql_requires_apply_payload():
|
|
proposal = {"id": "pid", "proposal_type": "add_edge", "payload": {"rationale": "x"}}
|
|
with pytest.raises(ValueError):
|
|
ap.build_apply_sql(proposal, None)
|
|
|
|
|
|
def test_build_apply_sql_rejects_unsupported_type():
|
|
proposal = {"id": "pid", "proposal_type": "reject_claim", "payload": {"apply_payload": {}}}
|
|
with pytest.raises(ValueError):
|
|
ap.build_apply_sql(proposal, None)
|
|
|
|
|
|
def test_assert_applyable_blocks_pending():
|
|
with pytest.raises(SystemExit):
|
|
ap.assert_applyable({"id": "pid", "status": "pending_review"})
|
|
|
|
|
|
def test_assert_applyable_blocks_already_applied():
|
|
with pytest.raises(SystemExit):
|
|
ap.assert_applyable({"id": "pid", "status": "applied"})
|
|
|
|
|
|
def test_assert_applyable_allows_approved():
|
|
ap.assert_applyable({"id": "pid", "status": "approved"}) # no raise
|
|
|
|
|
|
# --- ledger flip: concurrency guard + FK stamp --------------------------- #
|
|
def test_ledger_flip_asserts_rowcount_one():
|
|
# The flip must guard against a concurrent double-apply by asserting exactly
|
|
# one 'approved' row moved, not by re-reading status afterwards.
|
|
sql = ap.build_add_edge_sql({"from_claim": "a", "to_claim": "b", "edge_type": "supports"}, "pid", None)
|
|
assert "get diagnostics" in sql
|
|
assert "flipped <> 1" in sql
|
|
assert "and status = 'approved'" in sql
|
|
|
|
|
|
def test_ledger_flip_stamps_agent_fk():
|
|
sql = ap.build_add_edge_sql({"from_claim": "a", "to_claim": "b", "edge_type": "supports"}, "pid", "kb-apply")
|
|
# Hard resolve into a variable + NOT-NULL assert, never a silent inline
|
|
# subselect that would stamp NULL on an unresolved handle.
|
|
assert "select id into resolved_agent_id" in sql
|
|
assert "resolved_agent_id is null then" in sql
|
|
assert "applied_by_agent_id = resolved_agent_id" in sql
|
|
assert "applied_by_handle = 'kb-apply'" in sql
|
|
|
|
|
|
def test_build_apply_sql_defaults_applied_by_to_service_agent():
|
|
proposal = {
|
|
"id": "pid",
|
|
"proposal_type": "add_edge",
|
|
"payload": {"apply_payload": {"from_claim": "a", "to_claim": "b", "edge_type": "supports"}},
|
|
}
|
|
sql = ap.build_apply_sql(proposal, None)
|
|
assert f"applied_by_handle = '{ap.SERVICE_AGENT_HANDLE}'" in sql
|
|
|
|
|
|
if __name__ == "__main__":
|
|
import traceback
|
|
|
|
failures = 0
|
|
for name, fn in sorted(globals().items()):
|
|
if name.startswith("test_") and callable(fn):
|
|
try:
|
|
fn()
|
|
print(f"PASS {name}")
|
|
except Exception: # noqa: BLE001
|
|
failures += 1
|
|
print(f"FAIL {name}")
|
|
traceback.print_exc()
|
|
sys.exit(1 if failures else 0)
|