feat: pristine harness JSON interfaces and require-existing semantic mode (P2)

harness/tests/test_qdrant_cli_commands.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import hashlib
 import json
 from datetime import UTC, datetime
 from pathlib import Path
@@ -9,6 +10,7 @@ from typer.testing import CliRunner
 
 from tht.cli import app
 from tht.memory import MemoryRecord, save_registry
+from tht.ports.vector import VectorStoreError
 
 
 class _FakeEmbedder:
@@ -33,6 +35,10 @@ class _FakeVectorStore:
         return 3
 
 
+def _sha_file(path: Path) -> str:
+    return "sha256:" + hashlib.sha256(path.read_bytes()).hexdigest()
+
+
 def _qdrant_runtime_config(tmp_path: Path) -> Path:
     cfg = tmp_path / "workspace.yaml"
     cfg.write_text(
@@ -76,9 +82,7 @@ tables:
         type: bigint
 """
     )
-    (tmp_path / "artifacts" / "mschema" / "annotations.yaml").write_text(
-        "tables: {}\n"
-    )
+    (tmp_path / "artifacts" / "mschema" / "annotations.yaml").write_text("tables: {}\n")
 
 
 def _memory_record() -> MemoryRecord:
@@ -111,6 +115,73 @@ def test_vector_index_schema_accepts_qdrant_only_runtime_config(tmp_path, monkey
     assert store.upserts
 
 
+def test_vector_index_schema_json_is_pristine_and_reports_artifacts(tmp_path, monkeypatch):
+    cfg = _qdrant_runtime_config(tmp_path)
+    _write_schema_artifacts(tmp_path)
+    store = _FakeVectorStore()
+
+    monkeypatch.setattr("tht.adapters.factory.build_vector_store", lambda cfg, require_write: store)
+    monkeypatch.setattr("tht.cli.vector_cmd.make_embedder", lambda _: _FakeEmbedder())
+
+    response = CliRunner().invoke(app, ["vector", "index-schema", "--json", "-c", str(cfg)])
+
+    assert response.exit_code == 0, response.output
+    assert response.stderr == ""
+    assert json.loads(response.stdout) == {
+        "artifactIdentities": [
+            {
+                "digest": _sha_file(tmp_path / "artifacts" / "mschema" / "annotations.yaml"),
+                "kind": "schema_annotations",
+            },
+            {
+                "digest": _sha_file(tmp_path / "artifacts" / "mschema" / "physical.yaml"),
+                "kind": "physical_schema",
+            },
+        ],
+        "code": "ok",
+        "collection": "psd-clinical",
+        "counts": {
+            "added": 2,
+            "columns": 1,
+            "deleted": 0,
+            "records": 2,
+            "tables": 1,
+            "unchanged": 0,
+            "updated": 0,
+        },
+        "operation": "index_schema",
+        "schemaVersion": 1,
+        "status": "succeeded",
+        "workspaceId": "psd-clinical",
+        "workspaceRevision": "a" * 40,
+    }
+
+
+def test_vector_index_schema_json_failure_is_pristine(tmp_path, monkeypatch):
+    cfg = _qdrant_runtime_config(tmp_path)
+    _write_schema_artifacts(tmp_path)
+
+    def boom(cfg, require_write):
+        raise VectorStoreError("semantic_index_incompatible")
+
+    monkeypatch.setattr("tht.adapters.factory.build_vector_store", boom)
+    monkeypatch.setattr("tht.cli.vector_cmd.make_embedder", lambda _: _FakeEmbedder())
+
+    response = CliRunner().invoke(app, ["vector", "index-schema", "--json", "-c", str(cfg)])
+
+    assert response.exit_code == 1
+    assert response.stderr == ""
+    assert json.loads(response.stdout) == {
+        "code": "semantic_index_incompatible",
+        "error": "semantic index incompatible",
+        "operation": "index_schema",
+        "schemaVersion": 1,
+        "status": "failed",
+        "workspaceId": "psd-clinical",
+        "workspaceRevision": "a" * 40,
+    }
+
+
 def test_memory_promote_accepts_qdrant_only_runtime_config(tmp_path, monkeypatch):
     cfg = _qdrant_runtime_config(tmp_path)
     store = _FakeVectorStore()