chore: import existing codebase as baseline

tests/test_formula_export_reliability.py

@@ -0,0 +1,376 @@
+import stat
+from decimal import Decimal
+from io import BytesIO
+from pathlib import Path
+
+import pytest
+
+from finance.exporting import build_formula_workbook
+from finance.models import Fact, FormulaExport
+
+
+def custom_column():
+    return {
+        "name": "收入成本差",
+        "meaning": "实际收入减成本",
+        "scope": "按所选分组及本期、基期计算",
+        "missing": "结果不可用（#N/A）；零分母不可用",
+        "expression": {
+            "operator": "subtract",
+            "left": {"metric": "revenue"},
+            "right": {"metric": "cost"},
+        },
+    }
+
+
+pytestmark = pytest.mark.django_db
+
+
+def test_formula_workbook_fails_closed_when_row_or_group_limit_is_exceeded(workspace, monkeypatch):
+    import finance.exporting as exporting
+
+    ds, member, *_ = workspace
+    args = (
+        ds,
+        member,
+        {"start": "2026-08", "end": "2026-08", "comparison": "yoy", "group": "department"},
+        BytesIO(),
+    )
+    monkeypatch.setattr(exporting, "MAX_ROWS", 0)
+    with pytest.raises(ValueError, match="最多支持"):
+        build_formula_workbook(*args)
+
+    monkeypatch.setattr(exporting, "MAX_ROWS", 100_000)
+    monkeypatch.setattr(exporting, "MAX_GROUPS", 0)
+    with pytest.raises(ValueError, match="分组"):
+        build_formula_workbook(*args)
+
+
+def test_formula_workbook_uses_streaming_writer(workspace, monkeypatch):
+    import finance.exporting as exporting
+
+    ds, member, *_ = workspace
+    real_workbook = exporting.Workbook
+    options = {}
+
+    def workbook(*args, **kwargs):
+        options.update(kwargs)
+        return real_workbook(*args, **kwargs)
+
+    monkeypatch.setattr(exporting, "Workbook", workbook)
+    build_formula_workbook(
+        ds,
+        member,
+        {"start": "2026-08", "end": "2026-08", "comparison": "yoy", "group": "department"},
+        BytesIO(),
+    )
+    assert options == {"write_only": True}
+
+
+def test_formula_export_worker_publishes_complete_file_and_removes_temp_file(workspace):
+    from finance.exporting import process_formula_export
+
+    ds, _, client, *_ = workspace
+    context = {
+        "start": "2026-08",
+        "end": "2026-08",
+        "comparison": "yoy",
+        "group": "department",
+        "custom_columns": [custom_column()],
+    }
+    response = client.post(
+        f"/api/datasets/{ds.pk}/formula-exports", data=context, content_type="application/json"
+    )
+    job = FormulaExport.objects.get(pk=response.json()["id"])
+    stats = process_formula_export(job.pk)
+    job.refresh_from_db()
+    assert stats["rows"] == 5
+    assert stats["formula_version"] == "3"
+    assert job.status == "succeeded"
+    assert job.expires_at is not None
+    assert not job.path.endswith(".tmp")
+    assert Path(job.path).read_bytes()[:2] == b"PK"
+    assert stat.S_IMODE(Path(job.path).stat().st_mode) == 0o600
+    assert stat.S_IMODE(Path(job.path).parent.stat().st_mode) == 0o700
+    assert stat.S_IMODE(Path(job.path).parent.parent.stat().st_mode) == 0o700
+    assert not Path(job.path).with_suffix(".tmp").exists()
+    assert not list(Path(job.path).parent.glob(f"{job.pk}.*.tmp"))
+
+
+def test_formula_export_worker_records_failure_and_removes_partial_file(workspace, monkeypatch):
+    from finance.exporting import process_formula_export
+
+    ds, _, client, *_ = workspace
+    response = client.post(
+        f"/api/datasets/{ds.pk}/formula-exports",
+        data={"start": "2026-08", "end": "2026-08", "comparison": "yoy", "group": "department"},
+        content_type="application/json",
+    )
+    job = FormulaExport.objects.get(pk=response.json()["id"])
+
+    def fail(*args, **kwargs):
+        args[-1].write(b"partial")
+        raise ValueError("controlled export failure")
+
+    monkeypatch.setattr("finance.exporting.build_formula_workbook", fail)
+    with pytest.raises(ValueError, match="controlled export failure"):
+        process_formula_export(job.pk)
+    job.refresh_from_db()
+    assert job.status == "failed"
+    assert job.error == "controlled export failure"
+    response = client.get(f"/api/datasets/{ds.pk}/formula-exports/{job.pk}")
+    assert response.json()["error"] == "生成失败，请联系管理员并提供任务编号"
+    assert "controlled export failure" not in response.content.decode()
+    assert not job.path
+    temporary = (
+        Path(job.dataset.file.storage.location) / "formula-exports" / str(job.user_id) / f"{job.pk}.tmp"
+    )
+    assert not temporary.exists()
+
+
+def test_formula_export_worker_removes_stale_published_file_after_failed_retry(workspace, monkeypatch):
+    from finance.exporting import process_formula_export
+
+    ds, _, client, *_ = workspace
+    response = client.post(
+        f"/api/datasets/{ds.pk}/formula-exports",
+        data={"start": "2026-08", "end": "2026-08", "comparison": "yoy", "group": "department"},
+        content_type="application/json",
+    )
+    job = FormulaExport.objects.get(pk=response.json()["id"])
+    target = Path(job.dataset.file.storage.location) / "formula-exports" / str(job.user_id) / f"{job.pk}.xlsx"
+    target.parent.mkdir(parents=True)
+    target.write_bytes(b"stale workbook")
+
+    monkeypatch.setattr(
+        "finance.exporting.build_formula_workbook",
+        lambda *_args, **_kwargs: (_ for _ in ()).throw(ValueError("retry failed")),
+    )
+    with pytest.raises(ValueError, match="retry failed"):
+        process_formula_export(job.pk)
+
+    job.refresh_from_db()
+    assert job.status == "failed"
+    assert job.path == ""
+    assert job.expires_at is None
+    assert not target.exists()
+
+
+def test_formula_export_worker_treats_department_order_as_the_same_scope(workspace):
+    from finance.exporting import process_formula_export
+
+    ds, member, *_ = workspace
+    member.departments = ["外科", "儿科"]
+    member.save(update_fields=["departments"])
+    job = FormulaExport.objects.create(
+        dataset=ds,
+        user=ds.created_by,
+        context={"start": "2026-08", "end": "2026-08", "comparison": "yoy", "group": "department"},
+        departments=["外科", "儿科"],
+        status="processing",
+    )
+    member.departments = ["儿科", "外科"]
+    member.save(update_fields=["departments"])
+
+    process_formula_export(job.pk)
+    job.refresh_from_db()
+    assert job.status == "succeeded"
+
+
+def test_formula_export_worker_records_path_preparation_failure(workspace, settings):
+    from finance.exporting import process_formula_export
+
+    ds, _, client, *_ = workspace
+    response = client.post(
+        f"/api/datasets/{ds.pk}/formula-exports",
+        data={"start": "2026-08", "end": "2026-08", "comparison": "yoy", "group": "department"},
+        content_type="application/json",
+    )
+    job = FormulaExport.objects.get(pk=response.json()["id"])
+    job.status = "processing"
+    job.save(update_fields=["status"])
+    export_root = Path(settings.MEDIA_ROOT) / "formula-exports"
+    export_root.symlink_to(Path(settings.MEDIA_ROOT) / "elsewhere", target_is_directory=True)
+
+    with pytest.raises(ValueError, match="符号链接"):
+        process_formula_export(job.pk)
+    job.refresh_from_db()
+    assert job.status == "failed"
+
+
+def test_formula_workbook_rejects_unclassified_label_collision(workspace):
+    from decimal import Decimal
+
+    ds, member, *_ = workspace
+    Fact.objects.create(
+        dataset=ds,
+        period="2025-08",
+        dimensions={"department": None, "product": "医疗服务"},
+        revenue=Decimal("10"),
+        cost=Decimal("4"),
+        gross=Decimal("6"),
+        customer="missing-department",
+        source_row=20,
+    )
+    Fact.objects.create(
+        dataset=ds,
+        period="2025-08",
+        dimensions={"department": "（未分类）", "product": "医疗服务"},
+        revenue=Decimal("10"),
+        cost=Decimal("4"),
+        gross=Decimal("6"),
+        customer="literal-unclassified",
+        source_row=21,
+    )
+    with pytest.raises(ValueError, match="与系统缺失值标签冲突"):
+        build_formula_workbook(
+            ds,
+            member,
+            {
+                "start": "2026-08",
+                "end": "2026-08",
+                "comparison": "yoy",
+                "group": "department",
+            },
+            BytesIO(),
+        )
+
+
+def test_formula_criteria_escape_excel_wildcards_and_quotes():
+    from finance.exporting import _checked_formula, _criteria
+
+    criteria = _criteria(
+        {"subject": 'A*?~"科'},
+        "department",
+        "儿科?*",
+        ["2026-08", "2026-09"],
+    )
+    assert 'Data[subject],"=A~*~?~~""科"' in criteria
+    assert 'Data[department],"=儿科~?~*"' in criteria
+    assert 'Data[Period],">=2026-08"' in criteria
+    assert _criteria({}, "department", ">高值", ["2026-08"]).startswith('Data[department],"=>高值"')
+    with pytest.raises(ValueError, match="8,192"):
+        _checked_formula("=" + "A" * 8192)
+    with pytest.raises(ValueError, match="255"):
+        _criteria({}, "department", "科" * 255, ["2026-08"])
+
+
+def test_formula_workbook_rejects_case_insensitive_group_collision(workspace):
+    ds, member, *_ = workspace
+    for source_row, label in ((20, "Team"), (21, "team")):
+        Fact.objects.create(
+            dataset=ds,
+            period="2026-08",
+            dimensions={"department": label, "product": "医疗服务"},
+            revenue=Decimal(10),
+            cost=Decimal(4),
+            gross=Decimal(6),
+            customer=f"case-{source_row}",
+            source_row=source_row,
+        )
+
+    with pytest.raises(ValueError, match="大小写"):
+        build_formula_workbook(
+            ds,
+            member,
+            {"start": "2026-08", "end": "2026-08", "comparison": "yoy", "group": "department"},
+            BytesIO(),
+        )
+
+
+def test_formula_workbook_omits_default_sheet_and_writes_untrusted_labels_as_text(workspace):
+    from openpyxl import load_workbook
+
+    ds, member, *_ = workspace
+    dangerous = '=HYPERLINK("https://invalid")'
+    mapping = {**ds.mapping}
+    mapping["sheet"] = dangerous
+    mapping["columns"] = [dict(column) for column in mapping["columns"]]
+    mapping["columns"][1]["name"] = dangerous
+    mapping["columns"][3]["name"] = dangerous
+    mapping["self_operated_products"] = [dangerous]
+    ds.mapping = mapping
+    ds.save(update_fields=["mapping"])
+    ds.facts.update(dimensions={"department": dangerous, "product": "医疗服务"})
+
+    definition = custom_column()
+    definition["name"] = dangerous
+    output = BytesIO()
+    build_formula_workbook(
+        ds,
+        member,
+        {
+            "start": "2026-08",
+            "end": "2026-08",
+            "comparison": "yoy",
+            "group": "department",
+            "custom_columns": [definition],
+        },
+        output,
+    )
+
+    workbook = load_workbook(BytesIO(output.getvalue()), data_only=False)
+    assert "Sheet" not in workbook.sheetnames
+    cells = [
+        workbook["分析汇总"]["D3"],
+        workbook["原始证据"]["A2"],
+        workbook["原始证据"]["C1"],
+        workbook["参数与口径"]["B2"],
+        workbook["参数与口径"]["B4"],
+        workbook["对账检查"]["B2"],
+    ]
+    custom_reconcile = next(
+        cell for row in workbook["对账检查"].iter_rows() for cell in row if cell.value == dangerous
+    )
+    cells.append(custom_reconcile)
+    assert all(cell.value == dangerous and cell.data_type == "s" for cell in cells)
+
+
+def test_formula_reconcile_uses_literal_unclassified_group(workspace):
+    from openpyxl import load_workbook
+
+    ds, member, *_ = workspace
+    Fact.objects.create(
+        dataset=ds,
+        period="2026-08",
+        dimensions={"department": "（未分类）", "product": "医疗服务"},
+        revenue=Decimal(25),
+        cost=Decimal(10),
+        gross=Decimal(15),
+        customer="unclassified",
+        list_amount=Decimal(30),
+        source_row=20,
+        raw=["2026-08", "", "医疗服务", 25, 10, 15, "unclassified", 30],
+    )
+    output = BytesIO()
+    build_formula_workbook(
+        ds,
+        member,
+        {"start": "2026-08", "end": "2026-08", "comparison": "yoy", "group": "department"},
+        output,
+    )
+
+    workbook = load_workbook(BytesIO(output.getvalue()), data_only=False)
+    reference = next(
+        row[3].value
+        for row in workbook["对账检查"].iter_rows(min_row=2)
+        if row[0].value == "实际收入" and row[1].value == "（未分类）" and row[2].value == "本期"
+    )
+    assert Decimal(reference) == Decimal(25)
+
+
+def test_formula_workbook_rejects_dimension_keys_that_collide_with_reserved_headers(workspace):
+    ds, member, *_ = workspace
+    mapping = {**ds.mapping}
+    mapping["columns"] = [dict(column) for column in mapping["columns"]]
+    mapping["columns"][1]["key"] = "revenue"
+    ds.mapping = mapping
+    ds.save(update_fields=["mapping"])
+
+    with pytest.raises(ValueError, match="保留列名冲突"):
+        build_formula_workbook(
+            ds,
+            member,
+            {"start": "2026-08", "end": "2026-08", "comparison": "yoy", "group": "revenue"},
+            BytesIO(),
+        )