core/tests/test_public_showcase.py
Shay bfb54fb015 feat(demos): implement ADR-0099 — Public Showcase Demo
Single 30-second artifact composing four CORE invariants
(determinism, honest unknown, reviewed learning, multi-hop with
trace) by delegating to existing DemoCommand adapters. **No new
mechanism** — every claim is backed by an already-shipped,
separately-tested adapter. Closes the 8-ADR scale-up slate.

- new core/demos/learning_loop_adapter.py: LearningLoopDemo wraps
  ADR-0056 reviewed-teaching loop; _strip_volatile_paths drops
  transient temp-dir paths from raw before serialization so the
  adapter's report_sha256 is content-stable across runs
- new core/demos/showcase_adapters.py:
  - FabricationControlPublicDemo: re-runs ADR-0096 public split,
    produces 3 claims (refusal_recall_meets_threshold,
    fabrication_rate_below_threshold, trace_evidence_present)
  - MultiHopTraceDemo: runs 'Does light reveal truth?' with
    transitive_surface=True + composed_surface=True against
    cognition pack; surfaces a 3-hop walk light→truth→knowledge→
    evidence; produces 3 claims (grounded_answer, depth_two_or_more,
    walk_evidence_present)
- new core/demos/showcase.py: run_showcase() composes 4 scenes,
  emits showcase.json + per-scene artifacts; render_html() produces
  presentation-only static HTML with no JS injection vector;
  ShowcaseScene dataclass; MAX_RUNTIME_SECONDS=30 hard ceiling
  with DemoContractError if exceeded
- CLI: 'showcase' added to demo target choices; --output-dir flag
  added; cmd_demo dispatch branch writes showcase.json + showcase.html
- new evals/public_demo/ lane with 4 cases:
  - all_claims_supported (each scene + composite)
  - determinism_run_to_run_byte_equality (two runs identical after
    stripping volatile keys: total_runtime_ms, json_path,
    transient_corpus)
  - runtime_under_budget (≤30s)
  - pure_composition_no_new_mechanism (grep gate over showcase
    imports — must come from core/chat/generate/language_packs/
    teaching/evals or allowed stdlib only)
- lane is itself byte-identical across runs (sha256 5707db8efc6a..);
  runtime case omits exact runtime_ms (it varies near bucket
  boundaries) but still asserts ≤ budget
- 8 unit tests with module-scoped fixture (showcase runs once,
  ~13s total) covering payload shape, scene order, runtime budget,
  HTML render absence of <script>, and the pure-composition import
  gate independently of the lane
- ADR-0099 measured: total_runtime_ms ~12.8s, well under 30s budget
- smoke 67/67, cognition eval byte-identical 100/100/100/100;
  all 6 ADR-0092..0099 lanes byte-identical:
    reviewer_registry        681a2aab..
    miner_loop_closure       9f071733..
    domain_contract_validation f9c06cde..
    fabrication_control sum  01e1b6b7..
    demo_composition         27d83824..
    public_demo              5707db8e..
2026-05-21 19:44:48 -07:00

153 lines
4.9 KiB
Python

"""ADR-0099 Public Showcase Demo — unit tests."""
from __future__ import annotations
import tempfile
from collections.abc import Iterator
from pathlib import Path
from typing import Any
import pytest
from core.demos.showcase import (
MAX_RUNTIME_SECONDS,
SHOWCASE_VERSION,
render_html,
run_showcase,
)
REPO_ROOT = Path(__file__).resolve().parent.parent
@pytest.fixture(scope="module")
def showcase_payload() -> Iterator[dict[str, Any]]:
"""Run the showcase once and share its payload across all tests.
A full showcase run takes ~13s; running it per-test would balloon
the suite. Module-scoped fixture keeps all assertions on one
canonical artifact, which also better matches the production
invariant (one artifact, many claims).
"""
with tempfile.TemporaryDirectory(prefix="public_showcase_test_") as d:
yield run_showcase(output_dir=Path(d))
class TestShowcaseExecution:
def test_runs_and_returns_payload(self, showcase_payload: dict[str, Any]) -> None:
assert showcase_payload["showcase_version"] == SHOWCASE_VERSION
assert showcase_payload["claim_contract_version"] == 1
assert showcase_payload["max_runtime_seconds"] == MAX_RUNTIME_SECONDS
assert showcase_payload["all_claims_supported"] is True
def test_four_scenes_in_canonical_order(
self, showcase_payload: dict[str, Any]
) -> None:
assert [s["scene_id"] for s in showcase_payload["scenes"]] == [
"determinism",
"honest_unknown",
"reviewed_learning",
"multi_hop_trace",
]
def test_every_scene_has_at_least_one_supported_claim(
self, showcase_payload: dict[str, Any]
) -> None:
for scene in showcase_payload["scenes"]:
assert scene["claims"], f"scene {scene['scene_id']} has no claims"
for claim in scene["claims"]:
assert claim["supported"] is True
def test_runtime_within_budget(self, showcase_payload: dict[str, Any]) -> None:
runtime_ms = showcase_payload["total_runtime_ms"]
budget_ms = MAX_RUNTIME_SECONDS * 1000
assert runtime_ms <= budget_ms, (
f"showcase exceeded {budget_ms} ms budget; ran {runtime_ms} ms"
)
class TestHtmlRender:
def test_html_is_static_html(self, showcase_payload: dict[str, Any]) -> None:
html = render_html(showcase_payload)
assert html.startswith("<!doctype html>")
assert "</html>" in html
# No operator-supplied template path; no JS injection vector.
assert "<script" not in html.lower()
def test_html_renders_scene_ids(self, showcase_payload: dict[str, Any]) -> None:
html = render_html(showcase_payload)
for scene in showcase_payload["scenes"]:
assert scene["scene_id"] in html
class TestPureCompositionGate:
"""ADR-0099 invariant: ``public_showcase_pure_composition``.
Showcase imports must come from already-shipped packages
(``core/``, ``chat/``, ``generate/``, ``language_packs/``,
``teaching/``, ``evals/``) plus the stdlib. Any other import is
a new mechanism and must be blocked.
"""
ALLOWED_PREFIXES = (
"core.",
"chat.",
"generate.",
"language_packs.",
"teaching.",
"evals.",
)
ALLOWED_STDLIB = frozenset(
{
"__future__",
"subprocess",
"time",
"pathlib",
"typing",
"dataclasses",
"html",
"re",
"hashlib",
"json",
"os",
"sys",
}
)
SHOWCASE_SOURCES = (
REPO_ROOT / "core" / "demos" / "showcase.py",
REPO_ROOT / "core" / "demos" / "showcase_adapters.py",
REPO_ROOT / "core" / "demos" / "learning_loop_adapter.py",
)
def _imports_in(self, path: Path) -> list[str]:
import re
pattern = re.compile(r"^\s*(?:from\s+([\w\.]+)\s+import|import\s+([\w\.]+))")
mods: list[str] = []
for line in path.read_text(encoding="utf-8").splitlines():
match = pattern.match(line)
if match:
mods.append(match.group(1) or match.group(2))
return mods
def test_no_forbidden_imports(self) -> None:
forbidden: list[str] = []
for source in self.SHOWCASE_SOURCES:
for mod in self._imports_in(source):
if mod.startswith(self.ALLOWED_PREFIXES):
continue
if mod in self.ALLOWED_STDLIB:
continue
if mod.split(".", 1)[0] in self.ALLOWED_STDLIB:
continue
forbidden.append(f"{source.name}: {mod}")
assert forbidden == [], (
"ADR-0099 pure-composition violation: forbidden imports "
f"{forbidden}"
)
class TestRuntimeBudget:
def test_budget_is_thirty_seconds(self) -> None:
assert MAX_RUNTIME_SECONDS == 30