fix(investigation): persist large results and require graph links

Full-network pressure comparisons exceeded the worker limit before saved-result previews could run. Preserve complete calculation results and keep read-time summaries and pagination. Incremental graph publication accepted isolated nodes after flow-authoring rules were lost. Require new steps to connect explicitly while retaining historical snapshots and status-only patches. Regression tests reproduce both failures on the previous code.
2026-11-15 13:51:43 +08:00
parent d3a4433386
commit 4b47ae00d1
7 changed files with 173 additions and 5 deletions
@@ -1,5 +1,6 @@
 """Worker boundaries use real managed subprocesses, without external databases."""
 import asyncio
+import json
 from functools import wraps
 from types import SimpleNamespace
 from contextlib import asynccontextmanager
@@ -27,6 +28,66 @@ def run_async(function):
    return wrapper


+@run_async
+async def test_child_keeps_complete_result_above_eight_mib(monkeypatch, tmp_path):
+    from app.mcp import calculation_jobs
+    from app.services import calculation_worker
+
+    data = {"comparisons": "x" * (8 * 1024 * 1024), "map_payload": {"features": ["coverage"]}}
+
+    @asynccontextmanager
+    async def scope(row):
+        yield object()
+
+    async def calculate(scope, row):
+        return SimpleNamespace(model_dump=lambda **kwargs: data)
+
+    monkeypatch.setattr(calculation_worker, "execution_scope", scope)
+    monkeypatch.setattr(calculation_jobs, "execute_calculation_job", calculate)
+    input_path, output_path = tmp_path / "input.json", tmp_path / "result.json"
+    input_path.write_text("{}")
+    await calculation_worker.execute_child(str(input_path), str(output_path))
+    assert output_path.stat().st_size > 8 * 1024 * 1024
+    assert json.loads(output_path.read_text()) == {"status": "succeeded", "result": data}
+
+
+@run_async
+async def test_supervisor_persists_complete_result_above_eight_mib(monkeypatch, tmp_path):
+    monkeypatch.setenv("CALCULATION_WORKER_STATE_DIR", str(tmp_path))
+    original = asyncio.create_subprocess_exec
+
+    async def spawn(*args, **kwargs):
+        # A real child writes the same file protocol as the calculation process.
+        code = ('import json,pathlib,sys; pathlib.Path(sys.argv[1]).write_text('
+                'json.dumps({"status":"succeeded","result":{"comparisons":"x"*(8*1024*1024),'
+                '"map_payload":{"features":["coverage"]}}}))')
+        return await original(sys.executable, "-c", code, args[-1], start_new_session=True)
+
+    monkeypatch.setattr(asyncio, "create_subprocess_exec", spawn)
+    results = []
+
+    class Repository:
+        async def heartbeat(self, *args):
+            return "running"
+
+        async def finish(self, row, owner, **terminal):
+            results.append(terminal)
+            return True
+
+    @asynccontextmanager
+    async def repository(project):
+        yield Repository()
+
+    worker = CalculationWorker(WorkerSettings(heartbeat_seconds=.02), repository=repository)
+    await asyncio.wait_for(worker.run_job({"project_id": "project", "job_id": "large-result",
+                                          "timeout_seconds": 10}), 15)
+    assert len(results) == 1
+    assert results[0]["status"] == "succeeded"
+    assert results[0]["result"] == {
+        "comparisons": "x" * (8 * 1024 * 1024), "map_payload": {"features": ["coverage"]},
+    }
+
+
 def test_worker_settings_require_valid_concurrency(monkeypatch):
    settings = WorkerSettings.from_environment()
    assert settings.max_concurrency == 2
@@ -239,6 +239,85 @@ def test_graph_renderer_escapes_labels_and_keeps_explicit_parallel_and_failed_pa
    assert graph.node_refs[2].status == "reference"


+@pytest.mark.parametrize("edges", [[], [{"source": "check", "target": "check"}]])
+def test_new_multistep_graph_requires_relationships_instead_of_isolated_nodes(edges):
+    with pytest.raises(DomainError) as error:
+        update_investigation_graph(None, InvestigationGraphUpdate.model_validate({
+            "nodes": [{"node_id": "read", "label": "观测核验"},
+                      {"node_id": "check", "label": "异常判断"}],
+            "edges": edges,
+        }))
+    assert error.value.code == "GRAPH_RELATION_REQUIRED"
+    assert "check" in str(error.value)
+
+
+def test_new_branch_must_connect_to_the_saved_narrative_without_inventing_edges():
+    original = update_investigation_graph(None, InvestigationGraphUpdate.model_validate({
+        "nodes": [{"node_id": "root", "label": "调查起点"}],
+    }))
+    with pytest.raises(DomainError) as error:
+        update_investigation_graph(original, InvestigationGraphUpdate.model_validate({
+            "nodes": [{"node_id": "left", "label": "独立核查"},
+                      {"node_id": "right", "label": "核查发现"}],
+            "edges": [{"source": "left", "target": "right"}],
+        }))
+    assert error.value.code == "GRAPH_RELATION_REQUIRED"
+    assert original.edges == []
+    assert len(original.node_refs) == 1
+
+
+def test_connected_parallel_branches_and_join_render_top_down():
+    graph = update_investigation_graph(None, InvestigationGraphUpdate.model_validate({
+        "nodes": [{"node_id": key, "label": label} for key, label in (
+            ("start", "异常核验"), ("isolation", "隔离分析"),
+            ("impact", "爆管影响"), ("summary", "汇总判断"))],
+        "edges": [{"source": source, "target": target, "label": label}
+                  for source, target, label in (
+                      ("start", "isolation", "并行核查"), ("start", "impact", "并行核查"),
+                      ("isolation", "summary", "汇总依据"), ("impact", "summary", "汇总依据"))],
+    }))
+    assert graph.mermaid.startswith("flowchart TD\n")
+    assert len(graph.edges) == 4
+    assert 'start -->|"并行核查"| impact' in graph.mermaid
+    patched = update_investigation_graph(graph, InvestigationGraphUpdate.model_validate({
+        "nodes": [{"node_id": "impact", "status": "recorded"}],
+    }))
+    assert patched.edges == graph.edges
+
+
+def test_existing_disconnected_snapshot_remains_readable_and_patchable():
+    original = InvestigationDiagram.model_validate({
+        "title": "已保存的旧图", "mermaid": 'flowchart TD\n  read["观测"]\n  check["判断"]',
+        "edges": [], "node_refs": [{"node_id": "read", "label": "观测"},
+                                   {"node_id": "check", "label": "判断"}],
+    })
+    patched = update_investigation_graph(original, InvestigationGraphUpdate.model_validate({
+        "nodes": [{"node_id": "check", "status": "recorded"}],
+    }))
+    assert patched.mermaid == original.mermaid
+    assert patched.edges == original.edges == []
+    assert original.node_refs[1].status is None
+
+
+@pytest.mark.anyio
+async def test_publication_rejects_missing_links_and_accepts_a_corrected_graph():
+    task, repo, service = setup_service()
+    first = await publish(service, repo, task["task_id"], nodes=[
+        {"node_id": "read", "label": "观测核验"},
+    ])
+    original = first.task.latest_progress.progress.diagram.model_dump(mode="json")
+    with pytest.raises(DomainError) as error:
+        await publish(service, repo, task["task_id"], nodes=[
+            {"node_id": "check", "label": "异常判断"},
+        ])
+    assert error.value.code == "GRAPH_RELATION_REQUIRED"
+    assert (await service.detail(task["task_id"])).latest_progress.progress.diagram.model_dump(mode="json") == original
+    corrected = await publish(service, repo, task["task_id"], nodes=[
+        {"node_id": "check", "label": "异常判断"},
+    ], edges=[{"source": "read", "target": "check", "label": "核验依据"}])
+    assert 'read -->|"核验依据"| check' in corrected.task.latest_progress.progress.diagram.mermaid
+
+
 def test_first_structured_update_preserves_an_existing_mermaid_without_reinterpreting_it():
    original = InvestigationDiagram.model_validate({
        "title": "原调查路线", "mermaid": 'flowchart LR\nA{核查条件} -->|具备| B[原路径]\nA --> C[另一分支]',