feat(evidence): contribute to semantic stages (#42)

harness/tests/test_search_pack.py

@@ -5,7 +5,9 @@ from types import SimpleNamespace
 from typer.testing import CliRunner
 
 from tht.cli import app
-from tht.config import load_config
+from tht.config import load_config, workspace_id_for_config
+from tht.evidence.corpus.models import CorpusManifest
+from tht.evidence.corpus.store import CorpusStore
 from tht.jobs.dwh_pipeline import DwhPreprocessPipeline, config_dwh_binding
 from tht.mschema.models import ColumnPhysical, PhysicalSchema, TablePhysical
 from tht.ports.vector import VectorReadUnavailable
@@ -75,6 +77,19 @@ def _workspace(tmp_path, with_session=None):
         ],
         lsh_filenames=("s_lsh.pkl", "s_minhashes.pkl", "s_meta.json"),
     ).run()
+    corpus = CorpusStore(tmp_path / "corpus")
+    generation = "gen:" + "a" * 32
+    active = corpus.stage(
+        CorpusManifest(
+            vector_generation=generation,
+            embedding_model="test",
+            embedding_dimensions=1,
+            metadata={"workspace_id": workspace_id_for_config(load_config(cfg), cfg)},
+        ),
+        {},
+        generation=generation,
+    )
+    corpus.publish(active)
     if with_session:
         sdir = tmp_path / "sessions" / with_session
         sdir.mkdir(parents=True)
@@ -111,7 +126,7 @@ def test_pack_single_embed_and_sections(tmp_path, monkeypatch):
     _patch(monkeypatch, emb, searcher)
     res = CliRunner().invoke(app, ["search", "pack", "quanti pazienti", "-c", str(cfg)])
     assert res.exit_code == 0, res.output
-    assert emb.calls == 1  # UN solo embedding per le tre ricerche
+    assert emb.calls == 2  # schema/memory share one; Evidence embeds its own deterministic text
     assert [call["kinds"] for call in searcher.calls] == [
         ["schema_table", "schema_column"],
         ["solved_question"],
@@ -167,3 +182,14 @@ def test_pack_degrades_gracefully(tmp_path, monkeypatch):
     data = json.loads(res.output[res.output.index("{"):])
     assert data["tables"] == [] and data["evidence"] == [] and data["solved"] == []
     assert any("retrieval non disponibile" in w for w in data["warnings"])
+
+
+def test_pack_refuses_to_treat_an_unavailable_evidence_corpus_as_no_matches(tmp_path, monkeypatch):
+    cfg = _workspace(tmp_path)
+    (tmp_path / "corpus" / "ACTIVE").unlink()
+    _patch(monkeypatch, _FakeEmbedder(), _FakeSearcher())
+
+    result = CliRunner().invoke(app, ["search", "pack", "q", "-c", str(cfg)])
+
+    assert result.exit_code == 1
+    assert "Evidence non disponibile" in result.output