settled-reach/tooling/economy-db/test_atlas_idempotency.py

#!/usr/bin/env python3
"""
Idempotency tests for economy_import.atlas's names-only pool populators
(T-964 — Phase-4 test hardening).

`populate_atlas_city_names` and `populate_atlas_feature_names` both clear
their target table before reinserting (see each function's own docstring:
"there is no UNIQUE(body_id, name[, feature_type]) — without the clear a
re-run would accumulate duplicates"). This test proves that contract holds in
practice: two back-to-back runs against the same DB must yield an IDENTICAL
row count, not a doubled one, and Sol ('GJ 0') must remain permanently
exempt (D-223) on both runs.

Runs against a scratch COPY of the committed `server/data/systems.db` (real
schema + real `bodies`/FK data — not a hand-rolled in-memory schema, which
would drift from the real FK web this populator depends on). Wiki content
(`wiki/star-systems/*/bodies/*/markers.json`) is read directly from the repo
— read-only input, safe to reuse as-is; only the DB connection is scratch.
Deliberately calls the two populator functions directly rather than the full
`import_economics.py` CLI: the CLI's first step shells out to the Rust
`generate_brands` binary and overwrites `generated_brands.toml` on disk
(`brands.regenerate_brands`), a real side effect on shared repo content that
has nothing to do with atlas name-pool idempotency and would make this test
depend on a Rust build.

Stdlib only (unittest) — run directly or via `make test-tooling`:
  python3 tooling/economy-db/test_atlas_idempotency.py
"""

import shutil
import sqlite3
import sys
import tempfile
import unittest
from pathlib import Path

sys.path.insert(0, str(Path(__file__).resolve().parent))
sys.path.insert(0, str(Path(__file__).resolve().parent.parent))

from economy_import import atlas  # noqa: E402
from economy_import.paths import DB_PATH  # noqa: E402


class AtlasNamePoolIdempotencyTests(unittest.TestCase):
    """Two runs on a scratch DB copy must yield stable, non-doubled counts."""

    @classmethod
    def setUpClass(cls):
        if not DB_PATH.exists():
            raise unittest.SkipTest(f"committed systems.db not found at {DB_PATH}")

    def setUp(self):
        self._tmp = tempfile.TemporaryDirectory()
        scratch_path = Path(self._tmp.name) / "systems_scratch.db"
        shutil.copyfile(DB_PATH, scratch_path)
        self.conn = sqlite3.connect(str(scratch_path))
        self.conn.execute("PRAGMA foreign_keys=ON")

    def tearDown(self):
        self.conn.close()
        self._tmp.cleanup()

    def test_atlas_city_names_count_stable_across_two_runs(self):
        first = atlas.populate_atlas_city_names(self.conn, dry_run=False)
        first_count = self.conn.execute(
            "SELECT COUNT(*) FROM atlas_city_names"
        ).fetchone()[0]
        self.assertEqual(
            first, first_count, "populator return value must match the rows it wrote"
        )

        second = atlas.populate_atlas_city_names(self.conn, dry_run=False)
        second_count = self.conn.execute(
            "SELECT COUNT(*) FROM atlas_city_names"
        ).fetchone()[0]

        self.assertEqual(
            first_count,
            second_count,
            "a second run must yield an IDENTICAL atlas_city_names count, not "
            "a doubled one — the clear-before-reinsert contract must hold",
        )
        self.assertEqual(
            first, second, "the populator's own return value must also be stable"
        )
        self.assertGreater(
            first_count, 0, "fixture sanity: the committed wiki content must yield rows"
        )

    def test_atlas_feature_names_count_stable_across_two_runs(self):
        # T-1169: atlas_feature_names landed alongside atlas_city_names but
        # is a separate table/populator — must be checked independently, not
        # assumed to share the city-name populator's idempotency by proximity.
        first = atlas.populate_atlas_feature_names(self.conn, dry_run=False)
        first_count = self.conn.execute(
            "SELECT COUNT(*) FROM atlas_feature_names"
        ).fetchone()[0]
        self.assertEqual(first, first_count)

        second = atlas.populate_atlas_feature_names(self.conn, dry_run=False)
        second_count = self.conn.execute(
            "SELECT COUNT(*) FROM atlas_feature_names"
        ).fetchone()[0]

        self.assertEqual(
            first_count,
            second_count,
            "a second run must yield an IDENTICAL atlas_feature_names count, not "
            "a doubled one — the clear-before-reinsert contract must hold",
        )
        self.assertEqual(first, second)
        self.assertGreater(
            first_count, 0, "fixture sanity: the committed wiki content must yield rows"
        )

    def test_sol_system_gj0_gets_zero_rows_in_both_pools_across_two_runs(self):
        # D-223 permanent exemption: Sol ('GJ 0') keeps authored,
        # geometry-bearing markers.json via sol_import.py — never the names
        # pool. Must hold on the FIRST run (not just "never accumulates") and
        # remain zero on the second.
        for run in (1, 2):
            atlas.populate_atlas_city_names(self.conn, dry_run=False)
            atlas.populate_atlas_feature_names(self.conn, dry_run=False)

            sol_cities = self.conn.execute(
                """SELECT COUNT(*) FROM atlas_city_names
                   WHERE body_id IN (SELECT body_id FROM bodies WHERE system_id = 'GJ 0')"""
            ).fetchone()[0]
            sol_features = self.conn.execute(
                """SELECT COUNT(*) FROM atlas_feature_names
                   WHERE body_id IN (SELECT body_id FROM bodies WHERE system_id = 'GJ 0')"""
            ).fetchone()[0]

            self.assertEqual(
                sol_cities, 0, f"run {run}: Sol (GJ 0) must have zero atlas_city_names rows"
            )
            self.assertEqual(
                sol_features,
                0,
                f"run {run}: Sol (GJ 0) must have zero atlas_feature_names rows",
            )

        # Fixture sanity: GJ 0 must actually have bodies in this DB, or the
        # zero-rows assertions above would be vacuously true.
        gj0_body_count = self.conn.execute(
            "SELECT COUNT(*) FROM bodies WHERE system_id = 'GJ 0'"
        ).fetchone()[0]
        self.assertGreater(
            gj0_body_count,
            0,
            "fixture sanity: GJ 0 (Sol) must have bodies in the committed DB, or "
            "the exemption checks above are vacuous",
        )


if __name__ == "__main__":
    unittest.main(verbosity=2)