fix(preprocess): initialize fresh DWH writers safely

harness/tests/test_schema_introspect_guard.py

@@ -34,17 +34,62 @@ def _write_config(tmp_path):
     return cfg
 
 
-def test_introspect_cache_hit_skips_dwh(tmp_path):
-    # Le credenziali sono fasulle: se la guardia non scattasse PRIMA del branch
-    # transport, il comando tenterebbe la connessione e fallirebbe.
+def test_introspect_rejects_unbound_legacy_cache(tmp_path):
     catalog = _write_catalog(tmp_path)
-    before = catalog.read_bytes()
     cfg = _write_config(tmp_path)
     res = CliRunner().invoke(app, ["schema", "introspect", "-c", str(cfg)])
+    assert res.exit_code == 1
+    assert "legacy artifacts are unbound" in res.output
+    assert catalog.exists()
+
+
+def test_introspect_fresh_root_initializes_through_writer_job(tmp_path, monkeypatch):
+    import tht.cli.schema_cmd as module
+
+    cfg = _write_config(tmp_path)
+    physical = PhysicalSchema(
+        database="d", schema="s", introspected_at=datetime(2026, 1, 1),
+        tables={"dim_patient": TablePhysical(columns={"id": ColumnPhysical(type="bigint")})},
+    )
+
+    def refresh(_cfg, *, output_path=None, **_kwargs):
+        physical.to_yaml(output_path)
+        return physical
+
+    monkeypatch.setattr(module, "refresh_catalog", refresh)
+    res = CliRunner().invoke(app, ["schema", "introspect", "-c", str(cfg)])
     assert res.exit_code == 0, res.output
-    assert "OK (cache)" in res.output
+    assert (tmp_path / ".tht-dwh" / "OWNER.json").is_file()
     assert "1 tabelle" in res.output
-    assert catalog.read_bytes() == before
+
+
+def test_lsh_build_fresh_root_initializes_introspection_and_lsh(tmp_path, monkeypatch):
+    import tht.cli.lsh_cmd as lsh_module
+    import tht.cli.schema_cmd as schema_module
+    import tht.lshindex as lshindex_module
+
+    cfg = _write_config(tmp_path)
+    physical = PhysicalSchema(
+        database="d", schema="s", introspected_at=datetime(2026, 1, 1),
+        tables={"dim_patient": TablePhysical(columns={"id": ColumnPhysical(type="bigint")})},
+    )
+
+    def refresh(_cfg, *, output_path=None, **_kwargs):
+        physical.to_yaml(output_path)
+        return physical
+
+    def build(_cfg, *, physical_file, output_dir, **_kwargs):
+        assert physical_file.is_file()
+        for name in ("s_lsh.pkl", "s_minhashes.pkl", "s_meta.json"):
+            (output_dir / name).write_text("index")
+        return {}, [], [], {}
+
+    monkeypatch.setattr(schema_module, "refresh_catalog", refresh)
+    monkeypatch.setattr(lsh_module, "build_lsh_artifacts", build)
+    monkeypatch.setattr(lshindex_module, "load_index", lambda *_args, **_kwargs: (None, {}, None))
+    res = CliRunner().invoke(app, ["lsh", "build", "-c", str(cfg)])
+    assert res.exit_code == 0, res.output
+    assert (tmp_path / ".tht-dwh" / "OWNER.json").is_file()
 
 
 def test_introspect_refresh_bypasses_cache(tmp_path):