chore: 忽略 T+12.3 数据字典参考资料

- docs/**/*.rar、docs/T_12.3数据字典/ 与 __MACOSX/ 不入版本库

Co-Authored-By: Claude Code <noreply@anthropic.com>

.gitignore

@@ -64,3 +64,9 @@ CLAUDE.md
 backend/uploads/
 # 数据库备份导出目录（⑨ 备份脚本产物）
 db/backups/
+
+# 外部参考资料（2026-10-20）：T+12.3 数据字典压缩包与解压目录不入版本库
+docs/**/*.rar
+docs/T_12.3数据字典/
+# macOS 压缩包残留元数据
+__MACOSX/

backend/README.md

@@ -224,7 +224,10 @@ bcrypt 5.0 起对超过 72 字节的密码**直接抛 ValueError**（此前是
 ### 工作台口径（M4 ⑧，改动前先读）
 
 - **只读聚合**：总览只做 SQL 聚合（SUM / COUNT DISTINCT），不复用报表的部门汇总（后者加载整月台账，过重）；人数用 `COUNT(DISTINCT personnel_id)`——二次分摊拆行会让行数虚高。
-- **待办/预警实时派生**：查询即算，不建待办表/预警表。审批超时取 `approval_orders.create_time`（待审批状态）；回款逾期按「发货单 `ship_date` 超阈值且该批次无任何回款」判定；回款未指定批次 = `receipts.shipment_id IS NULL`。
+- **待办/预警实时派生**：查询即算，不建待办表/预警表。审批超时取 `approval_orders.create_time`（待审批状态）；回款逾期按「发货单 `ship_date` 超阈值且该批次无任何回款」判定，**已终止业务（status=3）排除**；回款未指定批次 = `receipts.shipment_id IS NULL`。
+- **待办按角色过滤**：`pending_statuses_for_roles`（approval_service）把当前用户角色映射到在途节点——财务只看待财务复核（status=1）、领导只看待领导审核（status=2）、admin 可看全部；多角色取并集。风险预警仍是全局经营视角，不按人裁剪。
+- **预警展示上限**：`items` 最多 20 条，`total` 是全部命中数（前端徽标用它，不能用 `items.length`）。
+- **趋势窗口**：回款趋势先按 `receipt_date` 范围过滤再按 `DATE_FORMAT` 分组，函数列不参与 WHERE，保证索引可用。
 - **同步降级**：⑥ 未建时 `sync_available=false`，前端显示「模块未启用」——不要返回 0 条冒充「无失败」。
 - **参数快照一次解析**：`warning.*` 三个阈值在每次请求只 `resolve_snapshot` 一次，多条规则复用。
 - **导出 SQL 的编码坑**：Windows 下 `alembic upgrade --sql > V*.sql` 默认按控制台 GBK 输出，`test_schema_sync` 按 UTF-8 读取会报 `UnicodeDecodeError`——必须 `PYTHONIOENCODING=utf-8 PYTHONUTF8=1` 前缀。
@@ -234,10 +237,12 @@ bcrypt 5.0 起对超过 72 字节的密码**直接抛 ValueError**（此前是
 - **数据源固定单笔生效行**：四张报表的明细一律 `performance_ledgers.is_active=1 AND calc_type=1`；经理聚合行（`calc_type=2`）**只在部门汇总作冻结对照**（`frozen_amount`）单读。聚合行 `receipt_id` 为 NULL、`payment_days` 为 NULL，混入明细会让合同编号/天数列失真，业务统计的绩效也会被整团队聚合额污染。
 - **调整单并入**：仅 `status=2`（生效），按 `(settlement_month, personnel_id)` 并入，**不要求该人当月有台账行**（只有调整单也要出零绩效成员行）；`direction=1` 补发记 +、`2` 扣减记 −（金额恒正）。待审批/驳回/撤回的金额绝不能进报表——`report_repository.list_month_adjustments` 与调整单管理页的 `list_adjustments` 是两个口径。调整单归属跟随该人普通行的实时 `manager_id`；只有经理个人户的经理跟随本人团队。
 - **回款金额按回款单去重**：二次分摊拆行使同一回款在台账出现多行（负责人 + 参与人），回款统计的金额侧必须 `DISTINCT receipt_id` 后再求和，否则按拆行数翻倍。个人明细页不放「回款金额」列（同一金额重复展示会让合计无法自洽），金额下沉到快照下钻。
-- **部门口径** = `personnel.manager_id` 实时上溯（无独立部门表）；离职人员不按 `status` 过滤，历史业绩留在原团队；无成员经理也出 0 值行（在职经理枚举 ∪ 数据中出现的经理）。团队基数与月结聚合行同源（`services/manager_share.py` 共享模块），**12.5% 与 5% 分列**——两项之和才是聚合行金额。
+- **部门口径** = `personnel.manager_id` 实时上溯（无独立部门表）；离职人员不按 `status` 过滤，历史业绩留在原团队；无成员经理也出 0 值行（在职经理枚举 ∪ 数据中出现的经理）。团队基数与月结聚合行同源（`services/manager_share.py` 共享模块）。
+- **全公司 5% 的发放口径**：月结只为有台账 bucket 的营销经理各生成一条聚合行（12.5%+5%），**每位经理各计一次 5%**；零成员团队无聚合行、不发 5%。报表的 `aggregate_amount` 优先取冻结聚合行、未生成月结时才用 `manager_team_share` 一次舍入预览；`team_subtotal`/`net_total` = 聚合额 + 调整净额。`team_share_total` 与 `company_share_total` 仅分列展示，两者之和与聚合额的差记入 `aggregate_rounding_diff`（分列各自 q2 会有 0.01～0.02 尾差）。
+- **对拍边界**：报表应发 = 月结 `total_payable` + 调整单净额（调整单不进月结 items）；筛选 `manager_id` 时只汇总展示团队，隐藏金额进 `other_*`，不能与全公司月结总额对拍。
 - **人员归属按行类型**：普通台账行跟随 `manager_id`；经理个人户行在 `manager.personal.included=1` 时归经理本人团队，且经理成员行**只取个人户金额**（其普通行业绩留在上级团队或未分配），保证团队基数 = 成员绩效合计。岗位判定范围必须覆盖所有成员引用的上级，否则其他经理的下属会被误判为「无经理」。按 `manager_id` 筛选时 `unassigned` 为空（语义是「仅看该经理团队」）。
 - **回款天数区间**是 8 档代码常量（`core/report_meta.py`，含「≤0 天先款后货」与「无天数」兜底桶），服务端下发元数据、前端禁止硬编码；空桶恒在且为 0。进度登记回款无台账行，不在回款统计口径内（页面注明）。
-- **导出**：openpyxl 普通 Workbook + `Decimal` 原值 + `number_format`（不写 Excel 公式）；超 `REPORT_EXPORT_MAX_ROWS`（20000）行返回 400；文本列防公式注入（`= + - @` 前缀加 `'`）；服务端文件名为 ASCII（响应头 latin-1 限制），中文名由前端 `link.download` 指定。部门汇总是**单 sheet 分块签字稿**（团队分成汇总 → 全公司 5% 单列 → 对账汇总 → 成员明细 → 未分配明细 → 签字栏）；全公司分成只出现一次，调整净额只在明细层求和，应发合计 = 团队成员应付 + 未分配应付 + 团队分成合计 + 全公司分成 + 调整净额合计。筛选页眉把人员/经理/客户/归属人/状态转为名称与中文标签；回款统计两个维度 sheet 共用主表筛选条件。
+- **导出**：openpyxl 普通 Workbook + `Decimal` 原值 + `number_format`（不写 Excel 公式）；超 `REPORT_EXPORT_MAX_ROWS`（20000）行返回 400；文本列防公式注入（`= + - @` 前缀加 `'`）；服务端文件名为 ASCII（响应头 latin-1 限制），中文名由前端 `link.download` 指定。部门汇总是**单 sheet 分块签字稿**（团队分成汇总 → 全公司 5% 单列 → 对账汇总 → 成员明细 → 未分配明细 → 签字栏）；第二块展示单经理 5%、计入次数与合计，对账块展示团队分成/全公司分成（分列）、聚合行舍入差、调整净额，应发合计 = 团队成员应付 + 未分配应付 + 各经理聚合行金额 + 调整净额合计。筛选页眉把人员/经理/客户/归属人/状态转为名称与中文标签；回款统计两个维度 sheet 共用主表筛选条件。
 - **导出审计**：`audit` 上下文覆盖取数与构建全过程，超限 400 和取数异常都会写失败操作日志（`success=0`）；成功摘要为静态报表标题。
 - **汇总 = 明细合计**：团队/公司基数由台账分桶结果直接得出，报表侧不另算一遍；未筛选时 `团队基数 + 未分配绩效 = 全公司基数`；集成测试逐项断言。
 - **岗位筛选选项**：`GET /api/v1/base/business-roles/options` 仅要求登录（含停用岗位），供财务/领导在报表中按岗位筛选；岗位管理接口仍仅管理员。

backend/app/api/v1/dashboard.py

@@ -30,10 +30,15 @@ _DashboardUser = Annotated[
 @router.get("", response_model=ApiResponse[DashboardOut], summary="工作台四区块")
 def get_dashboard(
     session: DbSession,
-    _: _DashboardUser,
+    current_user: _DashboardUser,
     month: Annotated[
         str | None, Query(max_length=7, description="统计月份 YYYY-MM，默认当月")
     ] = None,
 ) -> ApiResponse[DashboardOut]:
-    """一次返回经营总览/待办/风险预警/最近动态（待办与预警实时派生）。"""
-    return ok(dashboard_service.get_dashboard(session, month))
+    """一次返回经营总览/待办/风险预警/最近动态（待办与预警实时派生）。
+
+    待办按当前用户角色只返回其可操作节点（admin 可查看全部在途单）。
+    """
+    return ok(
+        dashboard_service.get_dashboard(session, month, current_user=current_user)
+    )

backend/app/api/v1/reports.py

@@ -14,10 +14,11 @@ from typing import Annotated, Literal
 
 from fastapi import APIRouter, Depends, Path, Query
 from fastapi.responses import StreamingResponse
-from pydantic import BaseModel
+from pydantic import BaseModel, ValidationError
 
 from app.api.deps import DbSession, RequestMetaDep, require_roles
 from app.core.constants import ROLE_ADMIN, ROLE_FINANCE, ROLE_LEADER
+from app.core.exceptions import BadRequestError
 from app.core.response import ApiResponse, ok
 from app.models import User
 from app.schemas.report import (
@@ -167,12 +168,24 @@ def export_report(
 
 
 def _build_export_query(code: str, query: ExportQuery):
-    """导出参数 → 各报表自己的查询模型（department-summary 月份必填由模型校验）。"""
+    """导出参数 → 各报表自己的查询模型（department-summary 月份必填由模型校验）。
+
+    手工构造模型抛出的是 `pydantic.ValidationError`，不是 FastAPI 的
+    `RequestValidationError`——不在这里转换就会落到全局 500。导出参数
+    非法与查询接口一样应当是 400。
+    """
     data = query.model_dump()
-    if code == "personal-detail":
-        return PersonalDetailQuery(**data)
-    if code == "department-summary":
-        return DepartmentSummaryQuery(**data)
-    if code == "business-stats":
-        return BusinessStatsQuery(**data)
-    return PaymentStatsQuery(**data)
+    model = {
+        "personal-detail": PersonalDetailQuery,
+        "department-summary": DepartmentSummaryQuery,
+        "business-stats": BusinessStatsQuery,
+        "payment-stats": PaymentStatsQuery,
+    }[code]
+    try:
+        return model(**data)
+    except ValidationError as exc:
+        first = exc.errors()[0]
+        location = ".".join(str(part) for part in first.get("loc", ()))
+        message = first.get("msg", "参数错误")
+        detail = f"{location} {message}" if location else message
+        raise BadRequestError(f"参数错误：{detail}") from exc

backend/app/core/report_meta.py

@@ -48,7 +48,10 @@ PRODUCT_CATEGORY_LABELS: dict[str, str] = {
     "mixed": "混合业务",
 }
 
-REPORT_TEMPLATE_VERSION = "V1.0"
+# 业务状态中文标签：导出列、筛选页眉与前端标签共用同一份映射
+BUSINESS_STATUS_LABELS: dict[int, str] = {1: "进行中", 2: "已完成", 3: "已终止"}
+
+REPORT_TEMPLATE_VERSION = "V1.2"
 # 导出超限即拒绝（需求「大数据量分批/超限提示」二选一取后者）：普通
 # Workbook 全量载入内存，2 万行约数十 MB，超过该量级必须引导按月导出
 REPORT_EXPORT_MAX_ROWS = 20000

backend/app/repositories/dashboard_repository.py

@@ -7,7 +7,7 @@
 
 from datetime import date, datetime
 
-from sqlalchemy import exists, func, select
+from sqlalchemy import exists, func, or_, select
 from sqlalchemy.orm import Session
 
 from app.models import (
@@ -66,12 +66,18 @@ def performance_trend(session: Session, months: list[str]) -> dict[str, object]:
     return dict(rows)
 
 
-def receipt_trend(session: Session, months: list[str]) -> dict[str, object]:
-    """近 N 个月回款合计（按到账月，含未核算回款——与台账口径不同，页面注明）。"""
+def receipt_trend(
+    session: Session, *, start_date: date, end_date: date
+) -> dict[str, object]:
+    """近 N 个月回款合计（按到账月，含未核算回款——与台账口径不同，页面注明）。
+
+    过滤走 ``receipt_date`` 原始列范围（可用索引）；``DATE_FORMAT`` 只用于
+    分组，不参与 WHERE，避免函数列导致全表扫描。
+    """
     month_expr = func.date_format(Receipt.receipt_date, "%Y-%m")
     rows = session.execute(
         select(month_expr, func.coalesce(func.sum(Receipt.amount), 0))
-        .where(month_expr.in_(months))
+        .where(Receipt.receipt_date >= start_date, Receipt.receipt_date < end_date)
         .group_by(month_expr)
     ).all()
     return dict(rows)
@@ -101,24 +107,33 @@ def get_month_settlement(session: Session, month: str) -> MonthlySettlement | No
 
 # ==== 待办事项 ====
 
-def list_pending_approvals(session: Session, *, limit: int) -> list[ApprovalOrder]:
-    """待处理审批单（按提交时间升序——最久的排前面）。"""
+def list_pending_approvals(
+    session: Session, *, statuses: tuple[int, ...], limit: int
+) -> list[ApprovalOrder]:
+    """待处理审批单（按提交时间升序——最久的排前面）。
+
+    ``statuses`` 由调用方按当前用户角色解析（待我审批），空元组返回空列表。
+    """
+    if not statuses:
+        return []
     return list(
         session.execute(
             select(ApprovalOrder)
-            .where(ApprovalOrder.status.in_(_PENDING_STATUSES))
+            .where(ApprovalOrder.status.in_(statuses))
             .order_by(ApprovalOrder.create_time.asc(), ApprovalOrder.id.asc())
             .limit(limit)
         ).scalars()
     )
 
 
-def count_pending_approvals(session: Session) -> int:
+def count_pending_approvals(session: Session, *, statuses: tuple[int, ...]) -> int:
+    if not statuses:
+        return 0
     return int(
         session.execute(
             select(func.count())
             .select_from(ApprovalOrder)
-            .where(ApprovalOrder.status.in_(_PENDING_STATUSES))
+            .where(ApprovalOrder.status.in_(statuses))
         ).scalar_one()
     )
 
@@ -152,7 +167,11 @@ def count_unassigned_receipts(session: Session) -> int:
 # ==== 风险预警 ====
 
 def list_overdue_shipments(session: Session, *, cutoff: date, limit: int) -> list:
-    """发货后超过阈值且该批次无任何回款的发货单（回款逾期规则）。"""
+    """发货后超过阈值且该批次无任何回款的发货单（回款逾期规则）。
+
+    已终止业务（status=3）不再催收，排除在预警之外；业务行缺失时仍保留，
+    避免历史脏数据让发货单从预警中静默消失。
+    """
     has_receipt = exists().where(Receipt.shipment_id == Shipment.id)
     return list(
         session.execute(
@@ -161,7 +180,11 @@ def list_overdue_shipments(session: Session, *, cutoff: date, limit: int) -> lis
                 Business.business_no.label("business_no"),
             )
             .outerjoin(Business, Business.id == Shipment.business_id)
-            .where(Shipment.ship_date <= cutoff, ~has_receipt)
+            .where(
+                Shipment.ship_date <= cutoff,
+                ~has_receipt,
+                or_(Business.status.is_(None), Business.status != 3),
+            )
             .order_by(Shipment.ship_date.asc())
             .limit(limit)
         ).all()
@@ -174,7 +197,12 @@ def count_overdue_shipments(session: Session, *, cutoff: date) -> int:
         session.execute(
             select(func.count())
             .select_from(Shipment)
-            .where(Shipment.ship_date <= cutoff, ~has_receipt)
+            .outerjoin(Business, Business.id == Shipment.business_id)
+            .where(
+                Shipment.ship_date <= cutoff,
+                ~has_receipt,
+                or_(Business.status.is_(None), Business.status != 3),
+            )
         ).scalar_one()
     )
 

backend/app/schemas/dashboard.py

@@ -98,6 +98,7 @@ class WarningBlock(BaseModel):
     receipt_overdue_days: int = 90
     approval_timeout_days: int = 3
     items: list[WarningItem] = Field(default_factory=list)
+    total: int = Field(default=0, description="全部规则命中总数（items 仅为展示上限内的子集）")
     receipt_overdue_total: int = 0
     approval_timeout_total: int = 0
     sync_available: bool = False

backend/app/schemas/report.py

@@ -7,10 +7,24 @@
 from datetime import date
 from decimal import Decimal
 
-from pydantic import BaseModel, ConfigDict, Field
+from pydantic import BaseModel, ConfigDict, Field, field_validator
 
+from app.core.month import validate_month
 from app.core.response import PageQuery
 
+
+def _required_month(value: str) -> str:
+    """必填月份：非法直接抛 ValueError，由 FastAPI 压成 400。"""
+    return validate_month(value)
+
+
+def _optional_month(value: str | None) -> str | None:
+    """可选月份：空串规范为 None（axios 未剔除时也不会误报格式错误）。"""
+    if value in (None, ""):
+        return None
+    return validate_month(value)
+
+
 # ==== 查询入参 ====
 
 
@@ -23,6 +37,11 @@ class PersonalDetailQuery(PageQuery):
     product_category: str | None = Field(default=None, max_length=32, description="产品类别筛选")
     role_code: str | None = Field(default=None, max_length=32, description="岗位筛选")
 
+    @field_validator("settlement_month")
+    @classmethod
+    def _v_settlement_month(cls, value: str | None) -> str | None:
+        return _optional_month(value)
+
 
 class DepartmentSummaryQuery(BaseModel):
     """部门汇总查询条件（月份必填——部门表按单月聚合，无跨月视角）。"""
@@ -30,6 +49,11 @@ class DepartmentSummaryQuery(BaseModel):
     settlement_month: str = Field(min_length=7, max_length=7, description="归属月 YYYY-MM")
     manager_id: int | None = Field(default=None, ge=1, description="仅看某经理团队")
 
+    @field_validator("settlement_month")
+    @classmethod
+    def _v_settlement_month(cls, value: str) -> str:
+        return _required_month(value)
+
 
 class BusinessStatsQuery(PageQuery):
     """业务统计查询条件；回款/绩效为该业务累计值，不随业务日期筛选而变。"""
@@ -51,6 +75,11 @@ class PaymentStatsQuery(BaseModel):
     customer_type: str | None = Field(default=None, max_length=32, description="客户类型筛选")
     product_category: str | None = Field(default=None, max_length=32, description="产品类别筛选")
 
+    @field_validator("settlement_month")
+    @classmethod
+    def _v_settlement_month(cls, value: str | None) -> str | None:
+        return _optional_month(value)
+
 
 class PaymentDetailQuery(PageQuery):
     """回款区间下钻（桶编码必填，入参由 payment-stats 页面展开行携带）。"""
@@ -62,6 +91,11 @@ class PaymentDetailQuery(PageQuery):
     customer_type: str | None = Field(default=None, max_length=32, description="客户类型筛选")
     product_category: str | None = Field(default=None, max_length=32, description="产品类别筛选")
 
+    @field_validator("settlement_month")
+    @classmethod
+    def _v_settlement_month(cls, value: str | None) -> str | None:
+        return _optional_month(value)
+
 
 # ==== 响应出参 ====
 
@@ -143,6 +177,13 @@ class DepartmentMemberRow(BaseModel):
     personnel_name: str
     role_code: str
     is_probation: int
+    member_kind: str = Field(
+        default="regular",
+        description="行来源：regular 普通台账 / manager_personal 经理个人户 / adjustment 仅调整单",
+    )
+    unassigned_reason: str | None = Field(
+        default=None, description="未纳入团队基数的原因（仅未分配块）"
+    )
     performance_sum: Decimal | None = None
     payable_sum: Decimal | None = None
     adjust_extra: Decimal = Field(default=Decimal("0"), description="调整单生效补发合计")
@@ -160,13 +201,26 @@ class DepartmentTeamRow(BaseModel):
     manager_status: int = Field(description="经理在职状态（1 在职 0 离职）")
     member_count: int
     team_base: Decimal = Field(description="团队基数（12.5% 的底，等于成员绩效合计）")
-    team_share_amount: Decimal = Field(description="12.5% 团队分成")
-    company_share_amount: Decimal = Field(description="5% 全公司分成（同一基数，各团队不互斥）")
+    team_share_amount: Decimal = Field(description="12.5% 团队分成（分列展示）")
+    company_share_amount: Decimal = Field(
+        description="单经理 5% 名义值；该团队无月结聚合行时为 0（不发放）"
+    )
+    has_aggregate: bool = Field(
+        default=False, description="本月是否有月结聚合行（决定是否发放 5%）"
+    )
+    aggregate_amount: Decimal = Field(
+        default=Decimal("0"),
+        description="月结聚合行应发（12.5%+5% 一次舍入）；无聚合行时 0",
+    )
+    team_subtotal: Decimal = Field(
+        default=Decimal("0"),
+        description="团队应发小计 = 聚合行金额 + 调整净额",
+    )
     company_base_ratio: Decimal = Field(description="团队基数占全公司基数比例（除零为 0）")
     adjust_extra: Decimal = Decimal("0")
     adjust_deduct: Decimal = Decimal("0")
     adjust_net: Decimal = Decimal("0")
-    net_total: Decimal = Field(description="团队净额 = 分成 + 调整净额")
+    net_total: Decimal = Field(description="团队净额 = 团队应发小计")
     frozen_amount: Decimal | None = Field(default=None, description="月结聚合行冻结额（对照）")
     aggregate_ledger_id: int | None = None
     members: list[DepartmentMemberRow] = Field(default_factory=list)
@@ -178,7 +232,50 @@ class DepartmentCompanyBlock(BaseModel):
     team_base_total: Decimal = Field(description="各团队基数之和（= 全部单笔绩效合计）")
     company_base: Decimal = Field(description="全公司基数（5% 的底）")
     company_share_pct: str = Field(description="分成比例参数原文")
-    company_share_amount: Decimal = Field(description="5% 全公司分成")
+    company_share_amount: Decimal = Field(description="单经理 5% 名义值")
+    company_share_count: int = Field(
+        default=0, description="计入应发的经理数（展示团队中有月结聚合行的数量）"
+    )
+    company_share_total: Decimal = Field(
+        default=Decimal("0"), description="5% 合计 = 单份 × 计入次数"
+    )
+
+
+class DepartmentReconciliation(BaseModel):
+    """部门汇总 · 对账块（导出/打印签字稿的合计口径，前端不自行求和）。
+
+    应发合计 = 团队成员应付 + 未分配应付 + 各经理聚合行金额 + 调整净额。
+    聚合行金额与月结 ``total_payable`` 同口径（12.5%+5% 一次舍入）；
+    ``team_share_total`` 与 ``company_share_total`` 仅作分列展示，两者
+    之和与聚合额的差记入 ``aggregate_rounding_diff``。
+    """
+
+    team_member_payable_total: Decimal = Field(description="展示团队成员的应付绩效合计")
+    unassigned_payable_total: Decimal = Field(description="展示未分配人员的应付绩效合计")
+    team_share_total: Decimal = Field(description="展示团队的 12.5% 分成合计（分列）")
+    company_share_total: Decimal = Field(
+        description="展示团队的全公司分成合计（每位有聚合行的经理各计一次，分列）"
+    )
+    aggregate_payable_total: Decimal = Field(
+        description="展示团队的经理聚合行应发合计（12.5%+5% 一次舍入）"
+    )
+    aggregate_rounding_diff: Decimal = Field(
+        default=Decimal("0"),
+        description="聚合额 − 团队分成合计 − 全公司分成合计（分列舍入差）",
+    )
+    adjust_net_total: Decimal = Field(description="明细层调整净额合计（成员 + 未分配）")
+    payable_total: Decimal = Field(description="应发合计 = 成员应付 + 未分配应付 + 聚合行 + 调整净额")
+    team_base_total: Decimal = Field(description="展示团队基数合计")
+    unassigned_performance_total: Decimal = Field(description="展示未分配人员绩效合计")
+    other_team_base: Decimal = Field(
+        default=Decimal("0"), description="未展示的基数（其他团队 + 全公司未分配）"
+    )
+    other_team_member_payable_total: Decimal = Field(default=Decimal("0"))
+    other_team_share_total: Decimal = Field(default=Decimal("0"))
+    other_team_company_share_total: Decimal = Field(default=Decimal("0"))
+    other_team_aggregate_total: Decimal = Field(default=Decimal("0"))
+    other_adjust_net_total: Decimal = Field(default=Decimal("0"))
+    company_base: Decimal = Field(description="全公司基数（5% 的底）")
 
 
 class DepartmentSummaryOut(BaseModel):
@@ -191,6 +288,9 @@ class DepartmentSummaryOut(BaseModel):
         default_factory=list, description="无经理人员的绩效（不参与分成，仅展示）"
     )
     company: DepartmentCompanyBlock | None = None
+    reconciliation: DepartmentReconciliation | None = Field(
+        default=None, description="签字稿对账块（应发公式各项）"
+    )
     totals: dict = Field(default_factory=dict, description="全表合计（成员/团队对账自证）")
 
 

backend/app/schemas/settlement.py

@@ -1,21 +1,17 @@
 """月度结算的请求/响应模型（M3 ④）。"""
 
-import re
 from datetime import datetime
 from decimal import Decimal
 
 from pydantic import BaseModel, ConfigDict, Field, field_validator
 
+from app.core.month import validate_month
 from app.core.response import PageQuery
 
-_MONTH_RE = re.compile(r"^\d{4}-\d{2}$")
-
 
 def _check_month(value: str) -> str:
-    """月份格式 YYYY-MM（settlement_month 全表唯一的维度）。"""
-    if not _MONTH_RE.match(value):
-        raise ValueError("月份格式须为 YYYY-MM")
-    return value
+    """月份格式 YYYY-MM 且月份 01-12（settlement_month 全表唯一的维度）。"""
+    return validate_month(value)
 
 
 class SettlementGenerateIn(BaseModel):
@@ -93,3 +89,8 @@ class AdjustmentCreateIn(BaseModel):
         description="金额（元，两位小数，恒正）",
     )
     reason: str = Field(min_length=1, max_length=255, description="扣补说明")
+
+    @field_validator("settlement_month")
+    @classmethod
+    def _v_settlement_month(cls, value: str) -> str:
+        return _check_month(value)

backend/app/services/approval_service.py

@@ -19,6 +19,7 @@
 """
 
 import logging
+from collections.abc import Iterable
 from datetime import date
 
 from sqlalchemy import func, select
@@ -29,6 +30,7 @@ from app.core.constants import (
     ACTION_REJECT,
     MODULE_APPROVAL,
     NO_PREFIX_APPROVAL,
+    ROLE_ADMIN,
     ROLE_FINANCE,
     ROLE_LEADER,
 )
@@ -84,6 +86,18 @@ _NODE_ROLE_BY_STATUS = {
 _ROLE_LABEL = {ROLE_FINANCE: "财务", ROLE_LEADER: "领导"}
 
 
+def pending_statuses_for_roles(role_codes: Iterable[str]) -> tuple[int, ...]:
+    """工作台「待我审批」：按当前用户角色映射在途审批状态。
+
+    admin 不代办审批，但作为管理视角可查看全部在途单（与审批管理列表
+    的可见性一致）；财务/领导只看到本节点可操作的条目。
+    """
+    codes = set(role_codes)
+    if ROLE_ADMIN in codes:
+        return _PENDING_STATUSES
+    return tuple(status for status, role in _NODE_ROLE_BY_STATUS.items() if role in codes)
+
+
 def is_enabled(session: Session, order_type: int) -> bool:
     """读当日参数快照判定审批开关（一次提交路径内只读一次，调用方勿重读）。"""
     code = ENABLE_CODE_BY_TYPE.get(order_type)

backend/app/services/dashboard_service.py

@@ -18,8 +18,9 @@ from sqlalchemy.orm import Session
 
 from app.core.exceptions import BadRequestError
 from app.core.month import validate_month
-from app.repositories import dashboard_repository, operation_log_repository
+from app.repositories import dashboard_repository, operation_log_repository, user_repository
 from app.repositories.query_utils import LogFilters
+from app.services import approval_service
 
 # 工作台只展示摘要，完整列表跳转对应管理页
 _TODO_LIMIT = 5
@@ -30,14 +31,21 @@ _TREND_MONTHS = 6
 _ORDER_TYPE_LABELS = {1: "二次分配方案", 2: "月度结算", 3: "调整单"}
 
 
-def get_dashboard(session: Session, month: str | None = None) -> dict:
-    """四区块一次性出数（前端一次请求渲染整页）。"""
+def get_dashboard(session: Session, month: str | None = None, *, current_user) -> dict:
+    """四区块一次性出数（前端一次请求渲染整页）。
+
+    待办按当前用户角色只展示其可操作节点（admin 可查看全部在途单）；
+    风险预警仍是全局经营视角，不按人裁剪。
+    """
     resolved = _normalize_month(month)
     warning_params = _warning_params(session)
+    pending_statuses = approval_service.pending_statuses_for_roles(
+        user_repository.get_role_codes(current_user)
+    )
     return {
         "month": resolved,
         "overview": _overview(session, resolved),
-        "todos": _todos(session),
+        "todos": _todos(session, pending_statuses),
         "warnings": _warnings(session, warning_params),
         "activities": _activities(session),
     }
@@ -61,6 +69,11 @@ def _shift_month(month: str, delta: int) -> str:
     return f"{index // 12:04d}-{index % 12 + 1:02d}"
 
 
+def _month_start(month: str) -> date:
+    """YYYY-MM → 该月首日（范围查询边界，避免函数列过滤）。"""
+    return date(int(month[:4]), int(month[5:7]), 1)
+
+
 def _trend_months(month: str) -> list[str]:
     """窗口为「当月及之前共 6 个月」，升序返回便于前端画图。"""
     return [_shift_month(month, offset) for offset in range(-(_TREND_MONTHS - 1), 1)]
@@ -72,7 +85,11 @@ def _overview(session: Session, month: str) -> dict:
     settlement = dashboard_repository.get_month_settlement(session, month)
     months = _trend_months(month)
     performance = dashboard_repository.performance_trend(session, months)
-    receipts = dashboard_repository.receipt_trend(session, months)
+    receipts = dashboard_repository.receipt_trend(
+        session,
+        start_date=_month_start(months[0]),
+        end_date=_month_start(_shift_month(months[-1], 1)),
+    )
     return {
         "month": month,
         "performance_total": _dec(sums["performance_total"]),
@@ -94,16 +111,27 @@ def _overview(session: Session, month: str) -> dict:
             }
             for item in months
         ],
-        "note": "绩效按台账归属月（单笔生效行，二次分摊拆行求和不失真）；回款按到账月（含未核算回款）；不含调整单",
+        "note": (
+            "绩效按台账归属月（单笔生效行，二次分摊拆行求和不失真）；回款按到账月"
+            "（含未核算回款）；不含经理 12.5%+5% 聚合分成与调整单"
+        ),
     }
 
 
-def _todos(session: Session) -> dict:
-    """待办事项（FR-034，澄清 #11 起步范围）。"""
-    orders = dashboard_repository.list_pending_approvals(session, limit=_TODO_LIMIT)
+def _todos(session: Session, pending_statuses: tuple[int, ...]) -> dict:
+    """待办事项（FR-034，澄清 #11 起步范围）。
+
+    ``pending_statuses`` 已按当前用户角色解析：财务只看待财务复核、
+    领导只看待领导审核；admin 查看全部在途单。
+    """
+    orders = dashboard_repository.list_pending_approvals(
+        session, statuses=pending_statuses, limit=_TODO_LIMIT
+    )
     receipts = dashboard_repository.list_unassigned_receipts(session, limit=_TODO_LIMIT)
     return {
-        "pending_approval_total": dashboard_repository.count_pending_approvals(session),
+        "pending_approval_total": dashboard_repository.count_pending_approvals(
+            session, statuses=pending_statuses
+        ),
         "pending_approvals": [
             {
                 "order_id": order.id,
@@ -154,6 +182,7 @@ def _warnings(session: Session, params: dict) -> dict:
             "receipt_overdue_days": params["receipt_overdue_days"],
             "approval_timeout_days": params["approval_timeout_days"],
             "items": [],
+            "total": 0,
             "receipt_overdue_total": 0,
             "approval_timeout_total": 0,
             "sync_available": False,
@@ -201,17 +230,21 @@ def _warnings(session: Session, params: dict) -> dict:
         }
         for order in orders
     ]
+    receipt_overdue_total = dashboard_repository.count_overdue_shipments(
+        session, cutoff=receipt_cutoff
+    )
+    approval_timeout_total = dashboard_repository.count_overdue_approvals(
+        session, cutoff=approval_cutoff
+    )
     return {
         "enabled": True,
         "receipt_overdue_days": params["receipt_overdue_days"],
         "approval_timeout_days": params["approval_timeout_days"],
         "items": items,
-        "receipt_overdue_total": dashboard_repository.count_overdue_shipments(
-            session, cutoff=receipt_cutoff
-        ),
-        "approval_timeout_total": dashboard_repository.count_overdue_approvals(
-            session, cutoff=approval_cutoff
-        ),
+        # items 有展示上限（_WARNING_LIMIT），徽标计数必须用全量命中数
+        "total": receipt_overdue_total + approval_timeout_total,
+        "receipt_overdue_total": receipt_overdue_total,
+        "approval_timeout_total": approval_timeout_total,
         "sync_available": False,
     }
 

backend/app/services/manager_share.py

@@ -14,6 +14,8 @@ from fractions import Fraction
 
 from sqlalchemy.orm import Session
 
+from app.core.exceptions import BadRequestError
+from app.core.month import parse_month
 from app.models import Personnel
 
 
@@ -38,7 +40,11 @@ def month_end(month: str) -> date:
     红线）。单笔台账行按回款日取参（裁决 #11）；聚合与月工资口径没有
     更细的业务日期，取该月最后可确定的生效时点。
     """
-    year, mon = int(month[:4]), int(month[5:7])
+    try:
+        year, mon = parse_month(month)
+    except ValueError as exc:
+        # schema 已拦住非法月份；内部调用绕过 schema 时也不能变成 500
+        raise BadRequestError(str(exc)) from exc
     return date(year, mon, monthrange(year, mon)[1])
 
 

backend/app/services/report_export.py

@@ -19,7 +19,11 @@ from openpyxl import Workbook
 from openpyxl.styles import Alignment, Border, Font, PatternFill, Side
 from openpyxl.utils import get_column_letter
 
-from app.core.report_meta import CUSTOMER_TYPE_LABELS, PRODUCT_CATEGORY_LABELS
+from app.core.report_meta import (
+    BUSINESS_STATUS_LABELS,
+    CUSTOMER_TYPE_LABELS,
+    PRODUCT_CATEGORY_LABELS,
+)
 
 _HEADER_FILL = PatternFill("solid", fgColor="DDE3F0")
 _THIN = Side(style="thin", color="B0B7C3")
@@ -31,7 +35,6 @@ _AMOUNT_FORMAT = "#,##0.00"
 _DATE_FORMAT = "YYYY-MM-DD"
 _FORMULA_PREFIXES = ("=", "+", "-", "@", "\t")
 
-BUSINESS_STATUS_LABELS = {1: "进行中", 2: "已完成", 3: "已终止"}
 
 
 @dataclass(frozen=True, slots=True)
@@ -44,6 +47,17 @@ class _Column:
     width: float
 
 
+@dataclass(frozen=True, slots=True)
+class _Block:
+    """签字稿中的一个区块：独立表头、数据行与可选合计行。"""
+
+    title: str
+    columns: tuple[_Column, ...]
+    rows: list[dict]
+    total: dict | None = None
+    empty_text: str = "（无）"
+
+
 _COLUMNS: dict[str, tuple[_Column, ...]] = {
     "personal-detail": (
         _Column("business_no", "合同编号", "text", 20),
@@ -101,11 +115,30 @@ def build_workbook(
     template_version: str,
     operator_name: str,
     company_name: str,
+    department: dict | None = None,
 ) -> bytes:
-    """构建并返回 xlsx 字节流（业务数据只读，不触碰数据库）。"""
+    """构建并返回 xlsx 字节流（业务数据只读，不触碰数据库）。
+
+    部门汇总是签字对账文件，走单 sheet 分块版式；其余报表仍是单表 +
+    合计行。`department` 为 `department_summary` 的完整出参。
+    """
     workbook = Workbook()
     sheet = workbook.active
     sheet.title = _sheet_name(code)
+    if code == "department-summary" and department is not None:
+        _write_department_document(
+            sheet,
+            department,
+            company_name=company_name,
+            title=title,
+            filters=filters,
+            template_version=template_version,
+            operator_name=operator_name,
+        )
+        buffer = BytesIO()
+        workbook.save(buffer)
+        return buffer.getvalue()
+
     columns = _COLUMNS[code]
     _write_header(sheet, columns, company_name, title, filters, template_version, operator_name)
     if rows:
@@ -129,6 +162,7 @@ def build_workbook(
                 extra,
                 dim_title,
                 company_name,
+                filters,
                 buckets.get(dim_key) or [],
                 template_version,
                 operator_name,
@@ -158,26 +192,15 @@ def _write_header(
     operator_name: str,
 ) -> None:
     """页眉四行 + 表头行；返回后 sheet.max_row 即数据行起点。"""
-    sheet.append([company_name])
-    _merge_row(sheet, len(columns))
-    sheet.cell(row=1, column=1).font = _COMPANY_FONT
-    sheet.cell(row=1, column=1).alignment = Alignment(horizontal="left", vertical="center")
-
-    sheet.append([title])
-    _merge_row(sheet, len(columns))
-    sheet.cell(row=2, column=1).font = _TITLE_FONT
-    sheet.cell(row=2, column=1).alignment = Alignment(horizontal="center", vertical="center")
-
-    sheet.append([f"筛选条件：{filters}"])
-    _merge_row(sheet, len(columns))
-    sheet.append(
-        [
-            f"导出时间：{datetime.now():%Y-%m-%d %H:%M:%S} ｜ 导出人：{operator_name} ｜ "
-            f"模板版本：{template_version}"
-        ]
+    _write_document_header(
+        sheet,
+        len(columns),
+        company_name,
+        title,
+        filters,
+        template_version,
+        operator_name,
     )
-    _merge_row(sheet, len(columns))
-    sheet.append([])
 
     sheet.append([column.title for column in columns])
     header_row = sheet.max_row
@@ -189,6 +212,38 @@ def _write_header(
     sheet.freeze_panes = sheet.cell(row=header_row + 1, column=1)
 
 
+def _write_document_header(
+    sheet,
+    column_count: int,
+    company_name: str,
+    title: str,
+    filters: str,
+    template_version: str,
+    operator_name: str,
+) -> None:
+    """公司名/标题/筛选条件/导出信息四行（分块文档与单表共用）。"""
+    sheet.append([company_name])
+    _merge_row(sheet, column_count)
+    sheet.cell(row=1, column=1).font = _COMPANY_FONT
+    sheet.cell(row=1, column=1).alignment = Alignment(horizontal="left", vertical="center")
+
+    sheet.append([title])
+    _merge_row(sheet, column_count)
+    sheet.cell(row=2, column=1).font = _TITLE_FONT
+    sheet.cell(row=2, column=1).alignment = Alignment(horizontal="center", vertical="center")
+
+    sheet.append([f"筛选条件：{filters}"])
+    _merge_row(sheet, column_count)
+    sheet.append(
+        [
+            f"导出时间：{datetime.now():%Y-%m-%d %H:%M:%S} ｜ 导出人：{operator_name} ｜ "
+            f"模板版本：{template_version}"
+        ]
+    )
+    _merge_row(sheet, column_count)
+    sheet.append([])
+
+
 def _merge_row(sheet, column_count: int) -> None:
     """合并当前行（公司名/标题/筛选行跨全表宽）。"""
     row = sheet.max_row
@@ -289,11 +344,12 @@ def _write_dimension_sheet(
     sheet,
     title: str,
     company_name: str,
+    filters: str,
     rows: list[dict],
     template_version: str,
     operator_name: str,
 ) -> None:
-    """维度对比 sheet（客户类型/产品类别）。"""
+    """维度对比 sheet（客户类型/产品类别），页眉筛选条件与主表一致。"""
     columns = (
         _Column("label", "维度", "text", 16),
         _Column("receipt_amount", "回款金额", "amount", 16),
@@ -301,9 +357,301 @@ def _write_dimension_sheet(
         _Column("payable_amount", "应付绩效", "amount", 14),
         _Column("ledger_count", "台账行数", "int", 10),
     )
-    _write_header(sheet, columns, company_name, title, "全部", template_version, operator_name)
+    _write_header(sheet, columns, company_name, title, filters, template_version, operator_name)
     for row in rows:
         _write_row(sheet, columns, row)
     _write_total(sheet, columns, rows)
     _write_signature(sheet, len(columns))
     _autofit(sheet, columns)
+
+
+# ==== 部门汇总签字稿（FR-030 / FR-032）====
+
+_MEMBER_KIND_LABELS = {
+    "regular": "普通绩效",
+    "manager_personal": "经理个人户",
+    "adjustment": "仅调整单",
+}
+
+# 各块列数不同，统一按列号取最大宽度（签字稿打印时不挤在一起）
+_DEPARTMENT_WIDTHS = (30, 16, 22, 13, 13, 15, 18, 14, 14, 13)
+
+
+def _write_department_document(
+    sheet,
+    data: dict,
+    *,
+    company_name: str,
+    title: str,
+    filters: str,
+    template_version: str,
+    operator_name: str,
+) -> None:
+    """部门汇总签字稿：团队分成 → 全公司 5% → 对账 → 成员 → 未分配 → 签字栏。
+
+    全公司 5% 每位有月结聚合行的经理各计一次，合计在第二块展示；
+    团队块的调整净额仅供成员汇总核对。应发合计由后端 `reconciliation`
+    按聚合行口径算好（含分列舍入差），本函数只渲染。
+    """
+    params = data.get("params") or {}
+    team_text = params.get("team_text", "")
+    company_text = params.get("company_text", "")
+    teams = data.get("teams") or []
+    unassigned = data.get("unassigned") or []
+    company = data.get("company") or {}
+    reconciliation = data.get("reconciliation") or {}
+    company_share_count = int(company.get("company_share_count") or 0)
+
+    _write_document_header(
+        sheet, len(_DEPARTMENT_WIDTHS), company_name, title, filters, template_version, operator_name
+    )
+
+    _write_block(
+        sheet,
+        _Block(
+            "一、团队分成汇总",
+            (
+                _Column("seq", "序号", "int", 6),
+                _Column("manager_name", "经理团队", "text", 16),
+                _Column("manager_status_text", "状态", "text", 8),
+                _Column("member_count", "团队人数", "int", 9),
+                _Column("team_base", "团队基数", "amount", 13),
+                _Column("team_share_amount", f"团队分成（{team_text}%）", "amount", 15),
+                _Column("adjust_net", "调整净额（仅核对）", "amount", 16),
+                _Column("team_subtotal", "团队应发小计（含聚合分成）", "amount", 18),
+                _Column("frozen_amount", "月结冻结额", "amount", 14),
+            ),
+            [
+                {
+                    "seq": index,
+                    "manager_name": team["manager_name"],
+                    "manager_status_text": "在职" if team["manager_status"] == 1 else "离职",
+                    "member_count": team["member_count"],
+                    "team_base": team["team_base"],
+                    "team_share_amount": team["team_share_amount"],
+                    "adjust_net": team["adjust_net"],
+                    "team_subtotal": team["team_subtotal"],
+                    "frozen_amount": team["frozen_amount"],
+                }
+                for index, team in enumerate(teams, start=1)
+            ],
+            {
+                "manager_name": "合计",
+                "member_count": sum(team["member_count"] for team in teams),
+                "team_base": _sum_amount(team["team_base"] for team in teams),
+                "team_share_amount": _sum_amount(
+                    team["team_share_amount"] for team in teams
+                ),
+                "team_subtotal": _sum_amount(team["team_subtotal"] for team in teams),
+                # 未生成月结时全部为 None，合计也留空（不能显示 0.00 误导）
+                "frozen_amount": (
+                    _sum_amount(
+                        team["frozen_amount"]
+                        for team in teams
+                        if team["frozen_amount"] is not None
+                    )
+                    if any(team["frozen_amount"] is not None for team in teams)
+                    else None
+                ),
+            },
+            "（无团队）",
+        ),
+    )
+
+    _write_block(
+        sheet,
+        _Block(
+            f"二、全公司分成（每位有聚合行的经理各计一次，共 {company_share_count} 次）",
+            (
+                _Column("company_base", "全公司基数", "amount", 16),
+                _Column("company_share_pct", "分成比例", "text", 12),
+                _Column("company_share_amount", "单经理分成", "amount", 13),
+                _Column("company_share_count", "计入次数", "int", 10),
+                _Column("company_share_total", "分成合计", "amount", 14),
+            ),
+            [
+                {
+                    "company_base": company.get("company_base"),
+                    "company_share_pct": f"{company_text}%",
+                    "company_share_amount": company.get("company_share_amount"),
+                    "company_share_count": company_share_count,
+                    "company_share_total": company.get("company_share_total"),
+                }
+            ],
+        ),
+    )
+
+    _write_block(
+        sheet,
+        _Block(
+            "三、对账汇总",
+            (
+                _Column("item", "项目", "text", 30),
+                _Column("amount", "金额", "amount", 16),
+                _Column("note", "说明", "text", 40),
+            ),
+            [
+                {
+                    "item": "团队成员应付合计",
+                    "amount": reconciliation.get("team_member_payable_total"),
+                },
+                {
+                    "item": "未分配人员应付合计",
+                    "amount": reconciliation.get("unassigned_payable_total"),
+                },
+                {"item": "团队分成合计（分列）", "amount": reconciliation.get("team_share_total")},
+                {
+                    "item": f"全公司分成合计（{company_share_count} 次，分列）",
+                    "amount": reconciliation.get("company_share_total"),
+                    "note": "每位有聚合行的经理各计一次",
+                },
+                {
+                    "item": "聚合行舍入差",
+                    "amount": reconciliation.get("aggregate_rounding_diff"),
+                    "note": "分列舍入与聚合行一次舍入的差",
+                },
+                {
+                    "item": "调整净额合计（明细层）",
+                    "amount": reconciliation.get("adjust_net_total"),
+                },
+                {
+                    "item": "应发合计",
+                    "amount": reconciliation.get("payable_total"),
+                    "note": "上列各项之和",
+                },
+                {"item": "团队基数合计", "amount": reconciliation.get("team_base_total")},
+                {
+                    "item": "未分配绩效合计",
+                    "amount": reconciliation.get("unassigned_performance_total"),
+                },
+                {
+                    "item": "其他未展示基数",
+                    "amount": reconciliation.get("other_team_base"),
+                    "note": "筛选时含其他团队与全公司未分配",
+                },
+                {"item": "全公司基数", "amount": reconciliation.get("company_base")},
+            ],
+        ),
+    )
+
+    member_columns = (
+        _Column("manager_name", "经理团队", "text", 15),
+        _Column("personnel_name", "成员", "text", 12),
+        _Column("role_code", "岗位", "text", 10),
+        _Column("member_kind_text", "来源", "text", 13),
+        _Column("performance_sum", "绩效金额", "amount", 13),
+        _Column("payable_sum", "应付绩效", "amount", 13),
+        _Column("adjust_extra", "调整补发", "amount", 12),
+        _Column("adjust_deduct", "调整扣减", "amount", 12),
+        _Column("adjust_net", "调整净额", "amount", 12),
+        _Column("net_total", "净合计", "amount", 13),
+    )
+    member_rows = [
+        {
+            "manager_name": team["manager_name"],
+            "personnel_name": member["personnel_name"],
+            "role_code": member["role_code"],
+            "member_kind_text": _MEMBER_KIND_LABELS.get(
+                member["member_kind"], member["member_kind"]
+            ),
+            "performance_sum": member["performance_sum"],
+            "payable_sum": member["payable_sum"],
+            "adjust_extra": member["adjust_extra"],
+            "adjust_deduct": member["adjust_deduct"],
+            "adjust_net": member["adjust_net"],
+            "net_total": member["net_total"],
+        }
+        for team in teams
+        for member in team["members"]
+    ]
+    _write_block(
+        sheet,
+        _Block("四、团队成员明细", member_columns, member_rows, empty_text="（无成员）"),
+        sum_amounts=True,
+    )
+
+    unassigned_columns = (
+        _Column("personnel_name", "人员", "text", 14),
+        _Column("role_code", "岗位", "text", 10),
+        _Column("unassigned_reason", "未纳入原因", "text", 24),
+        _Column("performance_sum", "绩效金额", "amount", 13),
+        _Column("payable_sum", "应付绩效", "amount", 13),
+        _Column("adjust_extra", "调整补发", "amount", 12),
+        _Column("adjust_deduct", "调整扣减", "amount", 12),
+        _Column("adjust_net", "调整净额", "amount", 12),
+        _Column("net_total", "净合计", "amount", 13),
+    )
+    unassigned_rows = [
+        {
+            "personnel_name": row["personnel_name"],
+            "role_code": row["role_code"],
+            "unassigned_reason": row.get("unassigned_reason") or "",
+            "performance_sum": row["performance_sum"],
+            "payable_sum": row["payable_sum"],
+            "adjust_extra": row["adjust_extra"],
+            "adjust_deduct": row["adjust_deduct"],
+            "adjust_net": row["adjust_net"],
+            "net_total": row["net_total"],
+        }
+        for row in unassigned
+    ]
+    _write_block(
+        sheet,
+        _Block("五、未分配人员明细", unassigned_columns, unassigned_rows),
+        sum_amounts=True,
+    )
+
+    _write_signature(sheet, len(_DEPARTMENT_WIDTHS))
+    for index, width in enumerate(_DEPARTMENT_WIDTHS, start=1):
+        sheet.column_dimensions[get_column_letter(index)].width = width
+
+
+def _write_block(sheet, block: _Block, *, sum_amounts: bool = False) -> None:
+    """写一个区块：标题、表头、数据/空提示，以及自定义或自动合计。"""
+    sheet.append([block.title])
+    title_row = sheet.max_row
+    _merge_row(sheet, len(block.columns))
+    sheet.cell(row=title_row, column=1).font = _HEADER_FONT
+
+    sheet.append([column.title for column in block.columns])
+    header_row = sheet.max_row
+    for cell in sheet[header_row]:
+        cell.font = _HEADER_FONT
+        cell.fill = _HEADER_FILL
+        cell.border = _BORDER
+        cell.alignment = Alignment(horizontal="center", vertical="center")
+
+    if block.rows:
+        for row in block.rows:
+            _write_row(sheet, block.columns, row)
+    else:
+        sheet.append([block.empty_text])
+        _merge_row(sheet, len(block.columns))
+
+    if block.total is not None:
+        _write_custom_total(sheet, block.columns, block.total)
+    elif sum_amounts:
+        _write_total(sheet, block.columns, block.rows)
+    sheet.append([])
+
+
+def _write_custom_total(sheet, columns: tuple[_Column, ...], total: dict) -> None:
+    """写后端预计算的合计行（不自动求和，避免重复计入共享基数）。"""
+    values = []
+    for column in columns:
+        raw = total.get(column.key)
+        values.append(_cell_value(column, raw) if raw is not None else None)
+    sheet.append(values)
+    for index, column in enumerate(columns, start=1):
+        cell = sheet.cell(row=sheet.max_row, column=index)
+        cell.font = _HEADER_FONT
+        cell.border = _BORDER
+        if column.kind == "amount":
+            cell.number_format = _AMOUNT_FORMAT
+
+
+def _sum_amount(values) -> Decimal:
+    """金额列求和后两位收口（None 跳过）。"""
+    return sum((value for value in values if value is not None), Decimal("0")).quantize(
+        Decimal("0.01"), rounding=ROUND_HALF_UP
+    )

backend/app/services/report_service.py

@@ -24,11 +24,14 @@ from app.core import report_meta
 from app.core.constants import ACTION_EXPORT, MODULE_REPORT
 from app.core.exceptions import BadRequestError
 from app.core.report_meta import (
+    BUSINESS_STATUS_LABELS,
+    CUSTOMER_TYPE_LABELS,
+    PRODUCT_CATEGORY_LABELS,
     REPORT_EXPORT_MAX_ROWS,
     REPORT_TEMPLATE_VERSION,
     DayBucket,
 )
-from app.models import BusinessRole, PerformanceLedger, Personnel, User
+from app.models import BusinessRole, Customer, PerformanceLedger, Personnel, User
 from app.models.settlement import ADJUSTMENT_DIRECTION_EXTRA
 from app.repositories import report_repository
 from app.services import audit_service, manager_share
@@ -124,12 +127,19 @@ def _q2(value: Decimal | None) -> Decimal | None:
 
 
 def department_summary(session, query) -> dict:
-    """按经理团队/月份汇总绩效工资（团队基数与月结聚合行同源）。"""
+    """按经理团队/月份汇总绩效工资（团队基数与月结聚合行同源）。
+
+    人员归属按「行类型」而不是「人」划分：普通台账行跟随 ``manager_id``，
+    经理个人户行按开关归经理本人——与 ``collect_team_buckets`` 完全同口径。
+    调整单是人的属性，优先跟随其普通行归属；只有经理个人户的经理跟随
+    本人团队。每个人/每类金额在团队与未分配之间只出现一次。
+    """
     month = query.settlement_month
     rows = report_repository.list_month_receipt_rows(session, month)
     share = manager_share.manager_share_params(session, month)
-    buckets = manager_share.collect_team_buckets(session, rows, share.include_manager_personal)
-    company_base = manager_share.company_base_of(rows, share.include_manager_personal)
+    include_mp = share.include_manager_personal
+    buckets = manager_share.collect_team_buckets(session, rows, include_mp)
+    company_base = manager_share.company_base_of(rows, include_mp)
 
     aggregates = {
         row.personnel_id: row
@@ -137,70 +147,235 @@ def department_summary(session, query) -> dict:
     }
     adjustments = _adjustments_by_person(report_repository.list_month_adjustments(session, month))
     member_map = _member_map(rows)
-    manager_ids = _manager_ids(session, buckets, query.manager_id)
-    # 岗位判定范围必须覆盖「无成员的经理」——只按 member_map 取会让零行团队被跳过
-    manager_of, manager_role = _team_relations(session, member_map, manager_ids)
+    # 先构建全部团队再按筛选展示：other_*（隐藏团队金额）需要完整集合，
+    # 只按 query.manager_id 枚举会让筛选态的 other_* 恒为 0
+    manager_ids = _manager_ids(session, buckets, None)
+    if query.manager_id is not None and query.manager_id not in manager_ids:
+        manager_ids = sorted(set(manager_ids) | {query.manager_id})
+    # 岗位判定范围必须覆盖「无成员的经理」与调整单人员——只按 member_map 取
+    # 会让零行团队被跳过，也会让其他经理的下属被误判为无经理
+    manager_of, manager_role, person_info = _team_relations(
+        session, member_map, manager_ids, adjustments
+    )
+    # 只有调整单、本月无台账行的人：补成零值成员后参与归属
+    _merge_adjustment_only_members(member_map, adjustments, person_info)
 
-    teams: list[dict] = []
-    for manager_id in manager_ids:
-        manager = session.get(Personnel, manager_id)
-        if manager is None or manager_role.get(manager_id) != "sales_manager":
-            continue  # 非营销经理的上级不展示团队行（与月结聚合同口径）
-        aggregate = aggregates.get(manager_id)
-        member_rows = _member_rows(
-            manager_id, member_map, manager_of, adjustments, share.include_manager_personal
-        )
-        team_extra = sum((m["adjust_extra"] for m in member_rows), Decimal("0"))
-        team_deduct = sum((m["adjust_deduct"] for m in member_rows), Decimal("0"))
-        team_base = _dec(buckets.get(manager_id, {}).get("performance", Decimal("0")))
-        # 12.5% 与 5% 分列（FR-030「全公司 5% 部分单列」）；两项之和才是月结
-        # 聚合行的金额（frozen_amount 对照），重复计入会双倍展示
-        team_share = _team_share(team_base, share.team_pct)
-        company_share = _company_share(company_base, share.company_pct)
-        teams.append(
-            {
-                "manager_id": manager_id,
-                "manager_name": manager.personnel_name,
-                "manager_status": manager.status,
-                "member_count": len(member_rows),
-                "team_base": _q2(team_base),
-                "team_share_amount": _q2(team_share),
-                "company_share_amount": _q2(company_share),
-                "company_base_ratio": _ratio(team_base, company_base),
-                "adjust_extra": _q2(team_extra),
-                "adjust_deduct": _q2(team_deduct),
-                "adjust_net": _q2(team_extra - team_deduct),
-                "net_total": _q2(team_share + company_share + team_extra - team_deduct),
-                "frozen_amount": aggregate.performance_amount if aggregate else None,
-                "aggregate_ledger_id": aggregate.id if aggregate else None,
-                "members": member_rows,
-            }
-        )
+    context = {
+        "aggregates": aggregates,
+        "buckets": buckets,
+        "member_map": member_map,
+        "manager_of": manager_of,
+        "manager_role": manager_role,
+        "adjustments": adjustments,
+        "include_mp": include_mp,
+        "share": share,
+        "company_base": company_base,
+    }
+    teams_all = _build_teams(session, manager_ids, context)
+    if query.manager_id is None:
+        # 离职经理本月无台账、但其下属只有调整单时，团队仍必须展示，
+        # 否则调整金额会从团队与未分配同时消失
+        extra_ids = {
+            target
+            for pid in adjustments
+            if (
+                target := _adjustment_target(
+                    pid, member_map, manager_of, manager_role, include_mp
+                )
+            )
+            is not None
+        } - set(manager_ids)
+        if extra_ids:
+            teams_all.extend(_build_teams(session, sorted(extra_ids), context))
 
-    unassigned = _unassigned_rows(member_map, manager_of, manager_role, adjustments)
+    teams = [
+        team
+        for team in teams_all
+        if query.manager_id is None or team["manager_id"] == query.manager_id
+    ]
+    unassigned_all = _unassigned_rows(
+        member_map, manager_of, manager_role, adjustments, include_mp
+    )
+    # 「仅看某经理团队」时未分配桶不随展示：全公司无经理人员与该团队无关
+    unassigned = [] if query.manager_id is not None else unassigned_all
+    company_share_amount = _company_share(company_base, share.company_pct)
+    company_share_count = sum(1 for team in teams if team["has_aggregate"])
+    reconciliation = _reconciliation(teams, teams_all, unassigned, company_base)
     return {
         "settlement_month": month,
         "params": {
             "team_text": share.team_text,
             "company_text": share.company_text,
-            "manager_personal_included": 1 if share.include_manager_personal else 0,
+            "manager_personal_included": 1 if include_mp else 0,
             "params_as_of": str(share.param_date),
         },
         "teams": teams,
         "unassigned": unassigned,
         "company": {
-            "team_base_total": _q2(sum((t["team_base"] for t in teams), Decimal("0"))),
+            "team_base_total": reconciliation["team_base_total"],
             "company_base": _q2(company_base),
             "company_share_pct": share.company_text,
-            "company_share_amount": _q2(_company_share(company_base, share.company_pct)),
+            # 单经理名义值；结算总额见 company_share_total（每位有聚合行的经理各计一次）
+            "company_share_amount": _q2(company_share_amount),
+            "company_share_count": company_share_count,
+            "company_share_total": _q2(company_share_amount * company_share_count),
         },
+        "reconciliation": reconciliation,
         "totals": {
-            "performance_total": _q2(sum((t["team_base"] for t in teams), Decimal("0"))),
-            "payable_total": _q2(sum((t["net_total"] for t in teams), Decimal("0"))),
+            "performance_total": reconciliation["team_base_total"],
+            "payable_total": reconciliation["payable_total"],
         },
     }
 
 
+def _build_teams(session, manager_ids: list[int], context: dict) -> list[dict]:
+    """按经理枚举生成团队行（含零成员团队）。"""
+    teams = []
+    for manager_id in manager_ids:
+        team = _build_team(session, manager_id, context)
+        if team is not None:
+            teams.append(team)
+    return teams
+
+
+def _build_team(session, manager_id: int, context: dict) -> dict | None:
+    """单个经理团队行；非营销经理的上级不展示（与月结聚合同口径）。"""
+    manager = session.get(Personnel, manager_id)
+    if manager is None or context["manager_role"].get(manager_id) != "sales_manager":
+        return None
+    aggregate = context["aggregates"].get(manager_id)
+    member_rows = _member_rows(
+        manager_id,
+        context["member_map"],
+        context["manager_of"],
+        context["manager_role"],
+        context["adjustments"],
+        context["include_mp"],
+    )
+    team_extra = sum((m["adjust_extra"] for m in member_rows), Decimal("0"))
+    team_deduct = sum((m["adjust_deduct"] for m in member_rows), Decimal("0"))
+    team_base = _dec(context["buckets"].get(manager_id, {}).get("performance", Decimal("0")))
+    # 月结只为有台账 bucket 的营销经理生成聚合行；零成员团队不发 5%
+    has_aggregate = manager_id in context["buckets"] or aggregate is not None
+    team_share = _team_share(team_base, context["share"].team_pct)
+    # 聚合额必须与月结「一次舍入」口径一致：优先取冻结行；未生成月结时
+    # 才用 12.5%+5% 合并公式预览。分列各自 q2 后相加会差 0.01～0.02。
+    if aggregate is not None:
+        aggregate_amount = _dec(aggregate.performance_amount or Decimal("0"))
+    elif manager_id in context["buckets"]:
+        aggregate_amount = _q2(
+            manager_team_share(
+                team_base,
+                context["company_base"],
+                team_pct=context["share"].team_pct,
+                company_pct=context["share"].company_pct,
+            )
+        )
+    else:
+        aggregate_amount = Decimal("0")
+    company_share = (
+        _company_share(context["company_base"], context["share"].company_pct)
+        if has_aggregate
+        else Decimal("0")
+    )
+    team_subtotal = _q2(aggregate_amount + team_extra - team_deduct)
+    return {
+        "manager_id": manager_id,
+        "manager_name": manager.personnel_name,
+        "manager_status": manager.status,
+        "member_count": len(member_rows),
+        "team_base": _q2(team_base),
+        "team_share_amount": _q2(team_share),
+        # 单经理 5% 名义值；无聚合行时为 0（不会发放）
+        "company_share_amount": _q2(company_share),
+        "has_aggregate": has_aggregate,
+        # 与月结聚合行同口径（12.5% + 5% 一次舍入）；无聚合行时为 0
+        "aggregate_amount": _q2(aggregate_amount),
+        # 团队应发小计 = 聚合行金额 + 调整净额；不再另加一次 5%
+        "team_subtotal": team_subtotal,
+        "company_base_ratio": _ratio(team_base, context["company_base"]),
+        "adjust_extra": _q2(team_extra),
+        "adjust_deduct": _q2(team_deduct),
+        "adjust_net": _q2(team_extra - team_deduct),
+        "net_total": team_subtotal,
+        "frozen_amount": aggregate.performance_amount if aggregate else None,
+        "aggregate_ledger_id": aggregate.id if aggregate else None,
+        "members": member_rows,
+    }
+
+
+def _reconciliation(
+    teams: list[dict],
+    teams_all: list[dict],
+    unassigned: list[dict],
+    company_base: Decimal,
+) -> dict:
+    """展示口径的对账块。
+
+    应发 = 团队成员应付 + 未分配应付 + 各经理聚合行金额 + 调整净额。
+    聚合行金额（``aggregate_amount``）含 12.5%+5%，且与月结「一次舍入」
+    完全一致；``team_share_total``/``company_share_total`` 只作分列展示，
+    两者之和与聚合额的差记入 ``aggregate_rounding_diff``（分列各自 q2
+    会产生 0.01～0.02 的尾差，直接相加会与月结总额不符）。
+
+    筛选态下 ``other_*`` 记录被隐藏团队的金额；它们不进入应发。
+    """
+    displayed_ids = {team["manager_id"] for team in teams}
+    other_teams = [team for team in teams_all if team["manager_id"] not in displayed_ids]
+
+    team_member_payable = sum(
+        (m["payable_sum"] for team in teams for m in team["members"]), Decimal("0")
+    )
+    unassigned_payable = sum((row["payable_sum"] for row in unassigned), Decimal("0"))
+    team_share_total = sum((team["team_share_amount"] for team in teams), Decimal("0"))
+    company_share_total = sum((team["company_share_amount"] for team in teams), Decimal("0"))
+    aggregate_payable_total = sum((team["aggregate_amount"] for team in teams), Decimal("0"))
+    aggregate_rounding_diff = aggregate_payable_total - team_share_total - company_share_total
+    adjust_net_total = sum(
+        (m["adjust_net"] for team in teams for m in team["members"]), Decimal("0")
+    ) + sum((row["adjust_net"] for row in unassigned), Decimal("0"))
+    team_base_total = sum((team["team_base"] for team in teams), Decimal("0"))
+    unassigned_performance = sum(
+        (row["performance_sum"] for row in unassigned), Decimal("0")
+    )
+    payable_total = (
+        team_member_payable + unassigned_payable + aggregate_payable_total + adjust_net_total
+    )
+    return {
+        "team_member_payable_total": _q2(team_member_payable),
+        "unassigned_payable_total": _q2(unassigned_payable),
+        "team_share_total": _q2(team_share_total),
+        # 各位有聚合行的经理 5% 之和（单份 × 计入次数）
+        "company_share_total": _q2(company_share_total),
+        # 与月结聚合行同口径的应发合计（12.5% + 5% 一次舍入）
+        "aggregate_payable_total": _q2(aggregate_payable_total),
+        # 分列展示与聚合额之间的舍入差；导出签字稿单列一行
+        "aggregate_rounding_diff": _q2(aggregate_rounding_diff),
+        "adjust_net_total": _q2(adjust_net_total),
+        "payable_total": _q2(payable_total),
+        "team_base_total": _q2(team_base_total),
+        "unassigned_performance_total": _q2(unassigned_performance),
+        # 筛选时包含其他团队基数与全公司未分配；未筛选时应为 0
+        "other_team_base": _q2(company_base - team_base_total - unassigned_performance),
+        "other_team_member_payable_total": _q2(
+            sum((m["payable_sum"] for team in other_teams for m in team["members"]), Decimal("0"))
+        ),
+        "other_team_share_total": _q2(
+            sum((team["team_share_amount"] for team in other_teams), Decimal("0"))
+        ),
+        "other_team_company_share_total": _q2(
+            sum((team["company_share_amount"] for team in other_teams), Decimal("0"))
+        ),
+        "other_team_aggregate_total": _q2(
+            sum((team["aggregate_amount"] for team in other_teams), Decimal("0"))
+        ),
+        "other_adjust_net_total": _q2(
+            sum((team["adjust_net"] for team in other_teams), Decimal("0"))
+        ),
+        "company_base": _q2(company_base),
+    }
+
+
 def _team_share(team_base: Decimal, team_pct: Fraction) -> Decimal:
     """12.5% 团队分成：全公司部分传 0，仅取 team_pct 项。"""
     return manager_team_share(
@@ -223,11 +398,19 @@ def _ratio(part: Decimal, whole: Decimal) -> Decimal:
 
 
 def _adjustments_by_person(orders) -> dict[int, dict]:
-    """生效调整单按人汇总（补发为正、扣减为负；金额本身恒正）。"""
+    """生效调整单按人汇总（补发为正、扣减为负；金额本身恒正）。
+
+    不要求该人本月有台账行——只有调整单的人也必须进入报表（计划 §一.3）。
+    """
     by_person: dict[int, dict] = {}
     for order in orders:
         acc = by_person.setdefault(
-            order.personnel_id, {"extra": Decimal("0"), "deduct": Decimal("0")}
+            order.personnel_id,
+            {
+                "extra": Decimal("0"),
+                "deduct": Decimal("0"),
+                "personnel_name": order.personnel_name,
+            },
         )
         if order.direction == ADJUSTMENT_DIRECTION_EXTRA:
             acc["extra"] += order.amount
@@ -237,7 +420,11 @@ def _adjustments_by_person(orders) -> dict[int, dict]:
 
 
 def _member_map(rows) -> dict[int, dict]:
-    """台账行按人求和（成员明细的唯一数据源），经理个人户金额单独记账。"""
+    """台账行按人求和（成员明细的唯一数据源）。
+
+    普通行与经理个人户行分开记账：团队基数只含后者，成员行也只能取
+    后者，否则经理本人的普通业绩会同时出现在成员行与未分配中。
+    """
     members: dict[int, dict] = {}
     for row in rows:
         acc = members.setdefault(
@@ -248,9 +435,13 @@ def _member_map(rows) -> dict[int, dict]:
                 "is_probation": row.is_probation,
                 "performance": Decimal("0"),
                 "payable": Decimal("0"),
+                "regular_performance": Decimal("0"),
+                "regular_payable": Decimal("0"),
                 "manager_personal_performance": Decimal("0"),
                 "manager_personal_payable": Decimal("0"),
-                "ledger_ids": [],
+                "regular_ledger_ids": [],
+                "manager_personal_ledger_ids": [],
+                "has_regular_rows": False,
                 "has_manager_personal": False,
             },
         )
@@ -258,34 +449,90 @@ def _member_map(rows) -> dict[int, dict]:
         payable = row.payable_amount or Decimal("0")
         acc["performance"] += performance
         acc["payable"] += payable
-        acc["ledger_ids"].append(row.id)
         if row.customer_type == "manager_personal":
             acc["has_manager_personal"] = True
             acc["manager_personal_performance"] += performance
             acc["manager_personal_payable"] += payable
+            acc["manager_personal_ledger_ids"].append(row.id)
+        else:
+            acc["has_regular_rows"] = True
+            acc["regular_performance"] += performance
+            acc["regular_payable"] += payable
+            acc["regular_ledger_ids"].append(row.id)
     return members
 
 
 def _team_relations(
-    session, member_map: dict[int, dict], manager_ids: list[int]
-) -> tuple[dict[int, int | None], dict[int, str]]:
-    """人员 → 直属经理与人员 → 岗位编码（各一次查询，避免逐人 get 的 N+1）。
+    session,
+    member_map: dict[int, dict],
+    manager_ids: list[int],
+    adjustments: dict[int, dict],
+) -> tuple[dict[int, int | None], dict[int, str], dict[int, dict]]:
+    """人员 → 直属经理、岗位编码与展示信息（批量查询，避免 N+1）。
 
-    查询范围 = 成员 ∪ 经理枚举：无成员的经理也要有岗位判定，否则「零行团队」
-    会被岗位过滤静默跳过（FR-030 要求无团队成员出 0 行）。
+    查询范围 = 成员 ∪ 经理枚举 ∪ 调整单人员；随后再补查这些人引用的
+    上级。缺少上级的岗位信息时，其他经理的下属会被误判为「无经理」，
+    零成员团队也会被岗位过滤静默跳过（FR-030 要求出 0 行）。
     经理判定与 ``manager_share.collect_team_buckets`` 的实时上溯同口径：
     离职人员的 manager_id 保留，历史业绩因此留在原团队。
     """
-    pids = list(set(member_map) | set(manager_ids))
-    if not pids:
-        return {}, {}
-    people = list(session.execute(select(Personnel).where(Personnel.id.in_(pids))).scalars())
-    manager_of = {p.id: p.manager_id for p in people}
+    seed = set(member_map) | set(manager_ids) | set(adjustments)
+    if not seed:
+        return {}, {}, {}
+    people = list(session.execute(select(Personnel).where(Personnel.id.in_(seed))).scalars())
+    referenced = {p.manager_id for p in people if p.manager_id is not None}
+    missing = referenced - seed
+    if missing:
+        people.extend(
+            session.execute(select(Personnel).where(Personnel.id.in_(missing))).scalars()
+        )
     role_by_id = {
         r.id: r.role_code for r in session.execute(select(BusinessRole)).scalars()
     }
+    manager_of = {p.id: p.manager_id for p in people}
     manager_role = {p.id: role_by_id.get(p.business_role_id, "") for p in people}
-    return manager_of, manager_role
+    person_info = {
+        p.id: {
+            "personnel_name": p.personnel_name,
+            "role_code": role_by_id.get(p.business_role_id, ""),
+            "is_probation": p.is_probation,
+        }
+        for p in people
+    }
+    return manager_of, manager_role, person_info
+
+
+def _merge_adjustment_only_members(
+    member_map: dict[int, dict], adjustments: dict[int, dict], person_info: dict[int, dict]
+) -> None:
+    """把「只有调整单、本月无台账行」的人补成零值成员。
+
+    补进 member_map 后，``_member_rows`` / ``_unassigned_rows`` 的归属
+    逻辑无需再为调整单单独开一条遍历分支；绩效为 0 也保证团队基数不变。
+    """
+    for personnel_id, adjustment in adjustments.items():
+        if personnel_id in member_map:
+            continue
+        info = person_info.get(personnel_id) or {
+            "personnel_name": adjustment["personnel_name"],
+            "role_code": "",
+            "is_probation": 0,
+        }
+        member_map[personnel_id] = {
+            "personnel_name": info["personnel_name"],
+            "role_code": info["role_code"],
+            "is_probation": info["is_probation"],
+            "performance": Decimal("0"),
+            "payable": Decimal("0"),
+            "regular_performance": Decimal("0"),
+            "regular_payable": Decimal("0"),
+            "manager_personal_performance": Decimal("0"),
+            "manager_personal_payable": Decimal("0"),
+            "regular_ledger_ids": [],
+            "manager_personal_ledger_ids": [],
+            "has_regular_rows": False,
+            "has_manager_personal": False,
+        }
 
 
 def _manager_ids(session, buckets: dict, only: int | None) -> list[int]:
@@ -304,37 +551,143 @@ def _manager_ids(session, buckets: dict, only: int | None) -> list[int]:
     return sorted(found)
 
 
+def _regular_team_of(
+    pid: int, manager_of: dict[int, int | None], manager_role: dict[int, str]
+) -> int | None:
+    """普通台账行的团队归属：直属上级是营销经理时归其团队。"""
+    manager_id = manager_of.get(pid)
+    if manager_id is not None and manager_role.get(manager_id) == "sales_manager":
+        return manager_id
+    return None
+
+
+def _adjustment_target(
+    pid: int,
+    member_map: dict[int, dict],
+    manager_of: dict[int, int | None],
+    manager_role: dict[int, str],
+    include_manager_personal: bool,
+) -> int | None:
+    """调整单归属：优先跟随普通行；只有经理个人户的经理跟随本人团队。
+
+    调整单没有行类型，只能按人的实时归属落位。每人只允许一行携带调整，
+    因此这里返回唯一的团队键（None = 未分配）。
+    """
+    member = member_map.get(pid)
+    if member is not None and member["has_regular_rows"]:
+        return _regular_team_of(pid, manager_of, manager_role)
+    if (
+        member is not None
+        and include_manager_personal
+        and member["has_manager_personal"]
+        and manager_role.get(pid) == "sales_manager"
+    ):
+        return pid
+    return _regular_team_of(pid, manager_of, manager_role)
+
+
 def _member_rows(
     manager_id: int,
     member_map: dict[int, dict],
     manager_of: dict[int, int | None],
+    manager_role: dict[int, str],
     adjustments: dict[int, dict],
     include_manager_personal: bool,
 ) -> list[dict]:
-    """团队内成员行：业务员按 manager_id 归属；经理个人户按开关归属本人。"""
-    pids = [pid for pid, mid in manager_of.items() if mid == manager_id and pid != manager_id]
-    # 经理本人（经理个人户行）：#12 开关开启时计入本团队基数
-    if include_manager_personal and manager_id in member_map:
-        pids.append(manager_id)
-    return [_member_row(pid, member_map[pid], adjustments) for pid in pids]
+    """团队内成员行：普通行按 manager_id 归属；经理个人户按开关归属本人。
+
+    经理本人只以「经理个人户金额」出现在自己团队，普通行业绩留在上级
+    团队或未分配——同一笔业绩不会在团队与未分配中重复展示。
+    """
+    rows = []
+    for pid, member in member_map.items():
+        target = _adjustment_target(
+            pid, member_map, manager_of, manager_role, include_manager_personal
+        )
+        regular_team = _regular_team_of(pid, manager_of, manager_role)
+        if pid != manager_id and member["has_regular_rows"] and regular_team == manager_id:
+            rows.append(
+                _member_row(
+                    pid,
+                    member,
+                    adjustments,
+                    "regular",
+                    performance=member["regular_performance"],
+                    payable=member["regular_payable"],
+                    ledger_ids=member["regular_ledger_ids"],
+                    include_adjust=target == manager_id,
+                )
+            )
+        personal_in_team = (
+            pid == manager_id
+            and include_manager_personal
+            and member["has_manager_personal"]
+        )
+        if personal_in_team:
+            rows.append(
+                _member_row(
+                    pid,
+                    member,
+                    adjustments,
+                    "manager_personal",
+                    performance=member["manager_personal_performance"],
+                    payable=member["manager_personal_payable"],
+                    ledger_ids=member["manager_personal_ledger_ids"],
+                    include_adjust=target == manager_id,
+                )
+            )
+        carried = (
+            member["has_regular_rows"] and regular_team == manager_id and target == manager_id
+        ) or (personal_in_team and target == manager_id)
+        # 调整单没有被普通行/个人户行承载时单独出一行（例如只有个人户、
+        # 但上级是营销经理的人员：调整随实时 manager_id 归上级团队）
+        if pid in adjustments and target == manager_id and not carried:
+            rows.append(
+                _member_row(
+                    pid,
+                    member,
+                    adjustments,
+                    "adjustment",
+                    performance=Decimal("0"),
+                    payable=Decimal("0"),
+                    ledger_ids=[],
+                    include_adjust=True,
+                )
+            )
+    return rows
 
 
-def _member_row(pid: int, member: dict, adjustments: dict[int, dict]) -> dict:
-    """单条成员行（绩效/应付求和 + 调整净额 + 台账 id 列表）。"""
-    adj = adjustments.get(pid, {"extra": Decimal("0"), "deduct": Decimal("0")})
-    net = member["payable"] + adj["extra"] - adj["deduct"]
+def _member_row(
+    pid: int,
+    member: dict,
+    adjustments: dict[int, dict],
+    member_kind: str,
+    *,
+    performance: Decimal,
+    payable: Decimal,
+    ledger_ids: list[int],
+    include_adjust: bool,
+) -> dict:
+    """单条成员行（按行类型取金额，调整净额最多计入一次）。"""
+    adj = (
+        adjustments.get(pid, {"extra": Decimal("0"), "deduct": Decimal("0")})
+        if include_adjust
+        else {"extra": Decimal("0"), "deduct": Decimal("0")}
+    )
+    net = payable + adj["extra"] - adj["deduct"]
     return {
         "personnel_id": pid,
         "personnel_name": member["personnel_name"],
         "role_code": member["role_code"],
         "is_probation": member["is_probation"],
-        "performance_sum": _q2(member["performance"]),
-        "payable_sum": _q2(member["payable"]),
+        "member_kind": member_kind,
+        "performance_sum": _q2(performance),
+        "payable_sum": _q2(payable),
         "adjust_extra": _q2(adj["extra"]),
         "adjust_deduct": _q2(adj["deduct"]),
         "adjust_net": _q2(adj["extra"] - adj["deduct"]),
         "net_total": _q2(net),
-        "ledger_ids": member["ledger_ids"],
+        "ledger_ids": list(ledger_ids),
     }
 
 
@@ -343,24 +696,68 @@ def _unassigned_rows(
     manager_of: dict[int, int | None],
     manager_role: dict[int, str],
     adjustments: dict[int, dict],
+    include_manager_personal: bool,
 ) -> list[dict]:
     """未进入团队基数的成员：无经理业务员 + 非营销经理上级的下属。
 
-    经理个人户业绩默认不计入任何基数（#12 开关关闭时），也不在此展示——
-    其金额仍可在个人明细报表中查询，避免与「团队基数 + 未分配」口径混淆。
+    经理个人户开关关闭时不展示个人户金额（与全公司基数排除口径一致）；
+    开关开启但本人不是营销经理时放入未分配，避免金额静默消失。
+    调整单在这里只出现一次：普通行/个人户行未承载时才单列。
     """
     rows = []
     for pid, member in member_map.items():
-        mid = manager_of.get(pid)
-        if mid is not None and manager_role.get(mid) == "sales_manager":
-            continue
-        regular = member["performance"] - member["manager_personal_performance"]
-        if regular == 0 and member["manager_personal_performance"] != 0:
-            continue
-        row = _member_row(pid, member, adjustments)
-        row["performance_sum"] = _q2(regular)
-        row["payable_sum"] = _q2(member["payable"] - member["manager_personal_payable"])
-        rows.append(row)
+        target = _adjustment_target(
+            pid, member_map, manager_of, manager_role, include_manager_personal
+        )
+        regular_team = _regular_team_of(pid, manager_of, manager_role)
+        regular_unassigned = member["has_regular_rows"] and regular_team is None
+        personal_unassigned = (
+            include_manager_personal
+            and member["has_manager_personal"]
+            and manager_role.get(pid) != "sales_manager"
+        )
+        if regular_unassigned:
+            row = _member_row(
+                pid,
+                member,
+                adjustments,
+                "regular",
+                performance=member["regular_performance"],
+                payable=member["regular_payable"],
+                ledger_ids=member["regular_ledger_ids"],
+                include_adjust=target is None,
+            )
+            row["unassigned_reason"] = "无经理或上级非营销经理"
+            rows.append(row)
+        if personal_unassigned:
+            row = _member_row(
+                pid,
+                member,
+                adjustments,
+                "manager_personal",
+                performance=member["manager_personal_performance"],
+                payable=member["manager_personal_payable"],
+                ledger_ids=member["manager_personal_ledger_ids"],
+                include_adjust=target is None and not member["has_regular_rows"],
+            )
+            row["unassigned_reason"] = "经理个人户（本人非营销经理）"
+            rows.append(row)
+        carried = (regular_unassigned and target is None) or (
+            personal_unassigned and target is None and not member["has_regular_rows"]
+        )
+        if pid in adjustments and target is None and not carried:
+            row = _member_row(
+                pid,
+                member,
+                adjustments,
+                "adjustment",
+                performance=Decimal("0"),
+                payable=Decimal("0"),
+                ledger_ids=[],
+                include_adjust=True,
+            )
+            row["unassigned_reason"] = "仅有调整单且无团队归属"
+            rows.append(row)
     return rows
 
 
@@ -547,26 +944,41 @@ EXPORT_FILENAMES = {
     "payment-stats": "payment_stats",
 }
 
+# 标题是导出审计摘要的一部分：进入 audit 前就要确定，不能在取数后才拼
+EXPORT_TITLES = {
+    "personal-detail": "个人绩效工资明细表",
+    "department-summary": "部门绩效工资汇总表",
+    "business-stats": "业务统计表",
+    "payment-stats": "回款情况统计表",
+}
+
 
 def export_report(session, code: str, query, *, operator: User, meta: RequestMeta | None = None):
-    """导出 Excel：上限校验 → 取数 → 构建 → 审计留痕（独立 commit）。
+    """导出 Excel：取数 → 构建 → 审计留痕（独立 commit）。
+
+    取数与上限校验必须在审计上下文内：超限 400 和取数异常同样要留下
+    失败记录（M4 计划 T12「导出 400/失败审计」）。`AuditRecord` 是 frozen
+    的，摘要只写静态标题，超限行数由异常文案带入 error_message。
 
     返回 (bytes, 文件名)；文件名由 API 层拼日期（ASCII，响应头 latin-1）。
     """
     from app.services import report_export
 
-    rows, context = _export_payload(session, code, query)
+    title = EXPORT_TITLES.get(code)
+    if title is None:
+        raise BadRequestError(f"未知的报表编码：{code}")
     with audit_service.audit(
         session,
         operator=operator,
         record=AuditRecord(
             module=MODULE_REPORT,
             action=ACTION_EXPORT,
-            summary=f"导出{context['title']}（{context['row_count']} 行）",
+            summary=f"导出{title}",
             request_data=query.model_dump(exclude_none=True),
         ),
         meta=meta,
     ):
+        rows, context = _export_payload(session, code, query)
         content = report_export.build_workbook(
             code=code,
             title=context["title"],
@@ -574,6 +986,7 @@ def export_report(session, code: str, query, *, operator: User, meta: RequestMet
             rows=rows,
             summary=context.get("summary"),
             buckets=context.get("buckets"),
+            department=context.get("department"),
             template_version=REPORT_TEMPLATE_VERSION,
             operator_name=getattr(operator, "username", ""),
             company_name=context["company_name"],
@@ -592,9 +1005,9 @@ def _export_payload(session, code: str, query) -> tuple[list[dict], dict]:
         role_names = _role_names(session)
         sums = report_repository.sum_personal_detail(session, query)
         return [_person_row(r, role_names) for r in rows], {
-            "title": "个人绩效工资明细表",
+            "title": EXPORT_TITLES["personal-detail"],
             "row_count": total,
-            "filters": _filter_text(query),
+            "filters": _filter_text(session, query),
             "summary": {
                 "performance_total": _q2(_dec(sums["performance_total"])),
                 "payable_total": _q2(_dec(sums["payable_total"])),
@@ -603,14 +1016,17 @@ def _export_payload(session, code: str, query) -> tuple[list[dict], dict]:
         }
     if code == "department-summary":
         data = department_summary(session, query)
-        member_total = sum(len(t["members"]) for t in data["teams"])
-        _check_export_limit(member_total)
-        rows = _department_export_rows(data)
-        return rows, {
-            "title": "部门绩效工资汇总表",
-            "row_count": member_total,
-            "filters": f"统计月份 {query.settlement_month}",
-            "summary": data["company"],
+        row_count = (
+            len(data["teams"])
+            + sum(len(team["members"]) for team in data["teams"])
+            + len(data["unassigned"])
+        )
+        _check_export_limit(row_count)
+        return [], {
+            "title": EXPORT_TITLES["department-summary"],
+            "row_count": row_count,
+            "filters": _filter_text(session, query),
+            "department": data,
             "company_name": _company_name(session),
         }
     if code == "business-stats":
@@ -621,9 +1037,9 @@ def _export_payload(session, code: str, query) -> tuple[list[dict], dict]:
         )
         sums = report_repository.sum_business_stats(session, query)
         return [_business_row(r) for r in rows], {
-            "title": "业务统计表",
+            "title": EXPORT_TITLES["business-stats"],
             "row_count": total,
-            "filters": _filter_text(query),
+            "filters": _filter_text(session, query),
             "summary": {
                 "ship_total": _q2(_dec(sums["ship_total"])),
                 "received_total": _q2(_dec(sums["received_total"])),
@@ -635,9 +1051,9 @@ def _export_payload(session, code: str, query) -> tuple[list[dict], dict]:
         data = payment_stats(session, query)
         rows = data["buckets"]
         return rows, {
-            "title": "回款情况统计表",
+            "title": EXPORT_TITLES["payment-stats"],
             "row_count": len(rows),
-            "filters": _filter_text(query),
+            "filters": _filter_text(session, query),
             "summary": data["totals"],
             "buckets": data,
             "company_name": _company_name(session),
@@ -654,43 +1070,79 @@ def _check_export_limit(row_count: int) -> None:
         )
 
 
-def _department_export_rows(data: dict) -> list[dict]:
-    """部门汇总导出展平：每个成员一行，团队列冗余（便于签字稿核对）。"""
-    rows = []
-    for team in data["teams"]:
-        for member in team["members"]:
-            rows.append(
-                {
-                    "manager_name": team["manager_name"],
-                    "member_name": member["personnel_name"],
-                    "role_code": member["role_code"],
-                    "performance_sum": member["performance_sum"],
-                    "payable_sum": member["payable_sum"],
-                    "adjust_net": member["adjust_net"],
-                    "net_total": member["net_total"],
-                }
-            )
-    return rows
+def _filter_text(session, query) -> str:
+    """筛选条件摘要（导出页眉展示；未设置项不展示；ID 转名称）。
 
-
-def _filter_text(query) -> str:
-    """筛选条件摘要（导出页眉展示；未设置项不展示）。"""
+    页眉是签字稿追溯筛选范围的唯一凭据：漏掉人员/经理/客户/归属人/
+    状态会让复核者把筛选结果当成全量数据。
+    """
+    personnel_ids = {
+        value
+        for field in ("personnel_id", "manager_id", "owner_personnel_id")
+        if (value := getattr(query, field, None)) not in (None, "")
+    }
+    personnel_names = _personnel_names(session, personnel_ids)
     parts = []
     for field, label in (
         ("settlement_month", "统计月份"),
+        ("personnel_id", "人员"),
+        ("manager_id", "经理"),
+        ("customer_id", "客户"),
+        ("owner_personnel_id", "归属人"),
         ("customer_type", "客户类型"),
         ("product_category", "产品类别"),
         ("role_code", "岗位"),
+        ("status", "状态"),
         ("keyword", "关键字"),
         ("date_from", "日期起"),
         ("date_to", "日期止"),
     ):
         value = getattr(query, field, None)
-        if value:
+        if value in (None, ""):
+            continue
+        if field in ("personnel_id", "manager_id", "owner_personnel_id"):
+            name = personnel_names.get(value)
+            parts.append(f"{label} {name}(#{value})" if name else f"{label} #{value}")
+        elif field == "customer_id":
+            name = _customer_name(session, value)
+            parts.append(f"{label} {name}(#{value})" if name else f"{label} #{value}")
+        elif field == "customer_type":
+            parts.append(f"{label} {CUSTOMER_TYPE_LABELS.get(value, value)}")
+        elif field == "product_category":
+            parts.append(f"{label} {PRODUCT_CATEGORY_LABELS.get(value, value)}")
+        elif field == "role_code":
+            parts.append(f"{label} {_role_name_by_code(session, value) or value}")
+        elif field == "status":
+            parts.append(f"{label} {BUSINESS_STATUS_LABELS.get(value, value)}")
+        else:
             parts.append(f"{label} {value}")
     return " ｜ ".join(parts) or "全部"
 
 
+def _personnel_names(session, ids: set[int]) -> dict[int, str]:
+    """批量取人员姓名；查不到时由调用方回退展示 #id。"""
+    if not ids:
+        return {}
+    rows = session.execute(
+        select(Personnel.id, Personnel.personnel_name).where(Personnel.id.in_(ids))
+    ).all()
+    return dict(rows)
+
+
+def _customer_name(session, customer_id: int) -> str | None:
+    """客户名称（导出页眉 ID→名称）。"""
+    return session.execute(
+        select(Customer.company_name).where(Customer.id == customer_id)
+    ).scalar_one_or_none()
+
+
+def _role_name_by_code(session, role_code: str) -> str | None:
+    """岗位名称（导出页眉展示中文，而不是内部编码）。"""
+    return session.execute(
+        select(BusinessRole.role_name).where(BusinessRole.role_code == role_code)
+    ).scalar_one_or_none()
+
+
 def _company_name(session) -> str:
     """导出页眉公司名（未配置时 system_setting_service 返回默认「衍水集团」）。"""
     from app.services import system_setting_service

backend/tests/README.md

@@ -140,8 +140,8 @@ uv run pytest tests/integration -q   # 仅集成
 
 | 文件 | 守什么 | 用例数 |
 | --- | --- | --- |
-| `tests/integration/test_reports.py` | 个人明细与台账逐笔一致 + **pageSize=1 时 summary 仍为全筛选合计**；部门汇总团队基数=成员绩效合计、分成按参数、无成员经理出 0 值行；**无台账经理挂上级不 500、仅有调整单人员并入团队/未分配、经理个人户开关两侧口径、按经理筛选不误判其他团队下属、对账恒等式**；导出页眉 ID→名称/中文标签、回款维度 sheet 透传筛选、**导出超限与取数异常的失败审计**、非法月份查询/导出均 400；回款统计 8 档桶恒在、空桶为 0、**二次分摊拆行时回款金额只计一次**；业务统计排除经理聚合行；普通用户 403 | 24 |
-| `tests/unit/test_report_helpers.py` | 区间分桶边界与 8 档编码唯一；标签覆盖全部枚举；Excel 模板页眉/签字栏/空数据/金额格式/合计/公式注入/维度 sheet/状态中文；**部门汇总签字稿五块顺序、零成员团队与未分配行、全公司分成只出现一次、应发合计取对账口径、比例标题参数化** | 16 |
+| `tests/integration/test_reports.py` | 个人明细与台账逐笔一致 + **pageSize=1 时 summary 仍为全筛选合计**；部门汇总团队基数=成员绩效合计、分成按参数、无成员经理出 0 值行；**无台账经理挂上级不 500、仅有调整单人员并入团队/未分配、经理个人户开关两侧口径、按经理筛选不误判其他团队下属、对账恒等式**；**应发与月结 `total_payable` 对拍（两位经理各计一次 5%）、有调整单时等于月结+调整净额、筛选态只汇总展示团队**；导出页眉 ID→名称/中文标签、回款维度 sheet 透传筛选、**导出超限与取数异常的失败审计**、非法月份查询/导出均 400；回款统计 8 档桶恒在、空桶为 0、**二次分摊拆行时回款金额只计一次**；业务统计排除经理聚合行；普通用户 403 | 27 |
+| `tests/unit/test_report_helpers.py` | 区间分桶边界与 8 档编码唯一；标签覆盖全部枚举；Excel 模板页眉/签字栏/空数据/金额格式/合计/公式注入/维度 sheet/状态中文；**部门汇总签字稿五块顺序、零成员团队与未分配行、5% 按计入次数累计、应发合计取对账口径、聚合行舍入差单列、比例标题参数化** | 17 |
 | `tests/unit/test_month_validation.py` | 非法月份（`2026-13`/`2026-00`/`0000-01` 等）在报表、月结、调整单 schema 入口即拒绝；可选月份空串规范为 None | 17 |
 
 运行：`uv run pytest tests/integration/test_reports.py -q` → **24 passed**（2026-10-19 修复回归）；`uv run pytest tests/unit/test_report_helpers.py tests/unit/test_month_validation.py -q` → **33 passed**；全量 **847 passed**（2026-10-19）。
@@ -152,10 +152,10 @@ uv run pytest tests/integration -q   # 仅集成
 
 | 文件 | 守什么 | 用例数 |
 | --- | --- | --- |
-| `tests/integration/test_dashboard.py` | 权限 403 且文案「无权访问工作台」；总览绩效合计与台账一致 + 非法月份 400；待办（月结提交→待审批命中、回款未指定批次命中、**同步未建显式降级**）；风险预警（审批超时命中、**参数关闭后不产生条目**）；最近动态来源操作日志 | 5 |
+| `tests/integration/test_dashboard.py` | 权限 403 且文案「无权访问工作台」；总览绩效合计与台账一致 + 非法月份 400；待办（月结提交→待审批命中、回款未指定批次命中、**同步未建显式降级**）；风险预警（审批超时命中、**参数关闭后不产生条目**）；最近动态来源操作日志；**待办按角色过滤（财务/领导/多角色/admin）+ 真实流转；回款逾期命中与阈值参数生效；终止业务排除；`warnings.total` 不受展示上限截断；回款趋势跨月归属** | 11 |
 | `tests/unit/test_param_registry.py`（改） | 注册表 19→**22 项**（V10 预警规则三项）；编码集合与迁移种子一致 | 改 |
 
-运行：`uv run pytest tests/integration/test_dashboard.py -q` → **5 passed**（2026-10-19）。
+运行：`uv run pytest tests/integration/test_dashboard.py -q` → **11 passed**（2026-10-19 code-review 修复回归）。
 
 **实施中发现并修复**：`alembic upgrade --sql` 在 Windows 控制台按 GBK 输出，重定向生成的 `V10__*.sql` 被 `test_schema_sync` 按 UTF-8 读取时抛 `UnicodeDecodeError`——导出命令须加 `PYTHONIOENCODING=utf-8 PYTHONUTF8=1`（已写入 backend/README）。
 
@@ -181,6 +181,8 @@ uv run pytest tests/integration -q   # 仅集成
 | 2026-10-19 | 拆行收口批次 | 全量 788 | ✅ 788 passed（9:20） | **2026-10-19 业务答复落实完成**：#14 首日取整 10‰（公式段/锚点不变，用例反转）；#12 单笔侧开关消费（1=业务员递减计提）；**二次分摊拆行**（参与人份额+负责人余额守恒，uk 三列版）；权限分层（params 查看放开/写仅 admin）；V8 迁移（审批开关默认开/撤试用期参数/增基数开关）。定点：calc 20 / settlements 13（含 #12 开关两档金标准 6.84→24.34）/ businesses 30（保存侧+3）/ formula 35。e2e calc_settlement_flow 交接并行会话 |
 | 2026-10-19 | **报表统计修复回归** | 全量 847 | ✅ 847 passed（10:13） | **M4 报表 code-review 15 项修复后全量回归**（后端 8 + 前端 7）：归属三集合/调整单全量并入/经理个人户拆分/对账出参、分块签字稿导出、导出审计覆盖取数、筛选页眉 ID→名称、月份校验 400、岗位只读选项。报表集成 24 / 报表单元 16 / 月份校验 17。type-check + build 绿；e2e `report_flow.py` 扩至 12 步，待 `E2E_ADMIN_PASSWORD` 实跑 |
 | 2026-10-19 | 全部 | 852 | ✅ 852 passed（9:17） | **M4 ⑧ 工作台完成**（四区块实时聚合 + ECharts 5.6 前端 + V10 预警参数三项；新增 5 例；参数注册表 19→22；同步模块显式降级）。期间修复：alembic `--sql` 导出 GBK 导致 schema 同步误报 |
+| 2026-10-19 | **code-review 修复回归** | 全量 862 | ✅ 862 passed（9:36） | **M4 工作台 + 报表 3 必须 + 8 建议修复**：普通用户落地页/守卫（报表集成 27、工作台集成 11、报表单元 17，新增 10 例）；部门汇总 5% 按计入次数累计并与月结 `total_payable` 对拍、聚合行舍入差；工作台待办按角色过滤、逾期命中与阈值生效、终止业务排除、`warnings.total` 截断、趋势跨月归属。ruff 绿；type-check + build 绿 |
+| 2026-10-20 | e2e 全链路 | — | ✅ login 8 / system 13 / role_home 4 / report 12 / calc_settlement 8 | 重启后端加载新代码后实跑：普通用户落地页与受限页回退、报表四页/导出/打印、核算与月结链路全过。`report_flow` 回款统计改为展开目标桶后按本次业务编号定位明细，不依赖桶累计值，可重复运行 |
 | 2026-10-19 | 单元全量 + 报表定点 | 484 | ✅ 484 passed（45.94s）；报表单元 11 例 0.70s | **M4 报表模块收尾**（前端四页/打印视图/导出与单元测试补齐）。ruff 绿；前端 `pnpm type-check` 绿。报表集成 7 例记录见上节 |
 | 2026-10-19 | 全部 | 806 | ✅ 806 passed（7:59） | **M4 ⑤ 报表统计全量回归**（报表新增 18 例：集成 7 + 单元 11；另含并行会话并入的 M3 前端修复与字体合规）。ruff 绿。e2e：`calc_settlement_flow` 8 步由并行会话跑通；`report_flow.py` 待 `E2E_ADMIN_PASSWORD` 运行 |
 
@@ -195,6 +197,9 @@ uv run pytest tests/integration -q   # 仅集成
 | 2026-10-19 | 导出取数与超限校验在审计上下文之外 | 超限 400 与取数异常**不留失败日志**，与 M4 计划 T12「导出 400/失败审计」不符 | `_export_payload` 移入 `with audit`；静态标题映射（AuditRecord frozen） |
 | 2026-10-19 | 月份只校验长度 | `2026-13` 传到 `calendar.monthrange` 抛 ValueError → **500**；导出手工构造模型抛裸 ValidationError 同样 500 | 新增 `core/month.py`（格式+01-12+年份）；四类 query、月结、调整单接入；导出 ValidationError 转 400 |
 | 2026-10-19 | 前端报表五处状态缺陷 | 打印超限判定恒假（静默截断）、部门下钻 query 未读取、回款多桶共用一个 drill 状态（面板串行）、summary 在请求序号守卫前写入（合计串号）、e2e 全页行数选择器必失败 | 分别修正判定/读取 query/按桶独立状态+数组参数/`onData` 回调/限定主桶表；补 e2e 多桶展开、下钻、截断、部门打印用例 |
+| 2026-10-19 | 工作台权限收紧后守卫仍回退 `/dashboard` | 普通用户登录/直连受限页被反复重定向到同一受限页 → **无法进入系统**（导航被判定无限重定向） | `router/access.ts` 统一准入与落地页解析（工作台 → 业务数据 → 回款）；登录回跳经 `canAccessPath` 校验；菜单/面包屑复用同一判定 |
+| 2026-10-19 | 部门汇总应发只计一次全公司 5% | 月结每位有聚合行的经理各计一次 5%，签字稿少计 N-1 份；**导出/打印金额小于月结应发** | 团队行增 `has_aggregate/aggregate_amount`（优先冻结聚合行）；对账改按聚合额求和，新增「聚合行舍入差」吸收分列 q2 尾差；补「与 `total_payable` 对拍」用例 |
+| 2026-10-19 | 工作台待办未按角色过滤 | 财务看得到「待领导审核」、领导看得到「待财务复核」，与 FR-034「仅看到与自己职责相关」不符 | `pending_statuses_for_roles` 按角色映射在途节点；repository 参数化 statuses；admin 保留全部在途视角 |
 | 2026-10-09 | 日志脱敏在 `logging.Filter` 阶段改写 `record.msg` | 抹掉 `%s` 占位符 → logging 内部 `msg % args` 抛 TypeError → **整条日志丢失**（比不脱敏更危险） | 改为 `SensitiveFormatter` 在格式化**之后**脱敏；补回归用例 |
 | 2026-10-09 | `PasswordHash.recommended()` 在 pwdlib 0.3.x 推荐 Argon2 而非 BCrypt | 密码算法与规范不符，且随库版本变化——同一批密码可能混用两种算法 | 改为显式 `PasswordHash((BcryptHasher(),))` 钉死算法 |
 | 2026-10-09 | 模型未设置表级 `comment`，`update_time` 用 `server_onupdate` | 违反规范红线；`server_onupdate` 只是 ORM 标记，**不生成 `ON UPDATE CURRENT_TIMESTAMP` 子句**，更新时间永不刷新 | 补表注释；改用 `text("CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP")` |

backend/tests/integration/test_business_roles.py

@@ -147,9 +147,18 @@ class TestBuiltinProtection:
 
 
 class TestPermission:
-    """权限：普通用户全部 403（当前默认管理员专用）。"""
+    """权限：管理接口普通用户 403；只读选项登录即可（报表筛选依赖）。"""
 
     def test_普通用户访问岗位列表_403(self, client: TestClient, make_user, auth_headers) -> None:
         user = make_user([ROLE_USER])
         headers = auth_headers(user)
         assert client.get(_ROLES, headers=headers).status_code == 403
+
+    def test_普通用户可读取岗位选项(self, client: TestClient, make_user, auth_headers) -> None:
+        """财务/领导做报表岗位筛选时不能因管理员接口 403 而失去筛选能力。"""
+        user = make_user([ROLE_USER])
+        headers = auth_headers(user)
+        resp = client.get("/api/v1/base/business-roles/options", headers=headers)
+        assert resp.status_code == 200, resp.text
+        codes = {row["role_code"] for row in resp.json()["data"]}
+        assert {"salesperson", "sales_manager", "info_officer"} <= codes

backend/tests/integration/test_dashboard.py

@@ -191,7 +191,7 @@ def test_预警_审批超时命中且参数关闭后不产生(client: TestClient
     client.post(f"{_SETTLEMENTS}/{settlement_id}/submit", headers=admin["headers"])
 
     # 把待审批单提交时间改到 4 天前（阈值默认 3 天）
-    from sqlalchemy import select, update
+    from sqlalchemy import update
 
     from app.core.db import get_session_factory
     from app.models import ApprovalOrder
@@ -230,3 +230,224 @@ def test_动态_来源操作日志(client: TestClient, admin: dict):
     data = _dashboard(client, admin["headers"])
     assert data["activities"], "工作台动态为空：写操作应产生操作日志"
     assert {"module", "action", "summary", "create_time"} <= set(data["activities"][0])
+
+
+# ==== 角色待办 / 预警阈值 / 趋势（2026-10-19 code-review 修复回归）====
+
+
+@pytest.fixture
+def finance(make_user, auth_headers) -> dict:
+    user = make_user(["finance"])
+    user["headers"] = auth_headers(user)
+    return user
+
+
+@pytest.fixture
+def leader(make_user, auth_headers) -> dict:
+    user = make_user(["leader"])
+    user["headers"] = auth_headers(user)
+    return user
+
+
+def _insert_approval(status: int) -> None:
+    """直插在途审批单（避免依赖两次流转，专注工作台过滤口径）。"""
+    from app.core.db import get_session_factory
+    from app.models import ApprovalOrder
+
+    factory = get_session_factory()
+    with factory() as session:
+        session.add(
+            ApprovalOrder(
+                order_no=f"AP{uuid.uuid4().hex[:12]}",
+                order_type=2,
+                ref_id=0,
+                refer_no="t_工作台",
+                status=status,
+                submitter_name="t_admin",
+            )
+        )
+        session.commit()
+
+
+def _set_param(client: TestClient, headers: dict, code: str, value: str) -> None:
+    resp = client.post(
+        f"{_PARAMS}/changes",
+        json={
+            "change_type": 1,
+            "remark": f"t_工作台：{code}",
+            "items": [{"param_code": code, "dimension_value": "", "value": value}],
+        },
+        headers=headers,
+    )
+    assert resp.status_code == 200, resp.text
+
+
+def _shipment_scene(client: TestClient, headers: dict, *, days_ago: int) -> tuple[dict, dict]:
+    """造一笔无回款的发货（用于回款逾期预警）。"""
+    from tests.integration.factories import business_item, business_payload, customer
+
+    person = _person(client, headers)
+    cust = customer(client, headers)
+    resp = client.post(
+        _BUSINESSES,
+        json=business_payload(
+            cust["id"],
+            owner_personnel_id=person["id"],
+            items=[business_item(cost_amount="3000.00")],
+        ),
+        headers=headers,
+    )
+    assert resp.status_code == 200, resp.text
+    business = resp.json()["data"]
+    ship_date = (date.today() - timedelta(days=days_ago)).isoformat()
+    resp = client.post(
+        _SHIPMENTS,
+        json={
+            "business_id": business["id"],
+            "ship_date": ship_date,
+            "items": [{"product_name": "t_工作台逾期", "quantity": "1", "sale_price": "1.00"}],
+        },
+        headers=headers,
+    )
+    assert resp.status_code == 200, resp.text
+    return business, resp.json()["data"]
+
+
+def test_待办_按角色仅显示可操作节点(
+    client: TestClient, admin: dict, finance: dict, leader: dict, make_user, auth_headers
+):
+    """财务只看到待财务复核、领导只看到待领导审核；admin 与多角色看并集。"""
+    _insert_approval(1)
+    _insert_approval(2)
+
+    assert _dashboard(client, finance["headers"])["todos"]["pending_approval_total"] == 1
+    assert _dashboard(client, leader["headers"])["todos"]["pending_approval_total"] == 1
+    assert _dashboard(client, admin["headers"])["todos"]["pending_approval_total"] == 2
+
+    both = make_user(["finance", "leader"])
+    both["headers"] = auth_headers(both)
+    assert _dashboard(client, both["headers"])["todos"]["pending_approval_total"] == 2
+
+    finance_orders = {
+        item["order_no"] for item in _dashboard(client, finance["headers"])["todos"]["pending_approvals"]
+    }
+    leader_orders = {
+        item["order_no"] for item in _dashboard(client, leader["headers"])["todos"]["pending_approvals"]
+    }
+    assert finance_orders.isdisjoint(leader_orders)
+
+
+def test_待办_财务通过后领导出现待办(
+    client: TestClient, admin: dict, finance: dict, leader: dict
+):
+    """真实流转：财务节点通过后，条目从财务待办消失并进入领导待办。"""
+    _chain(client, admin["headers"])
+    resp = client.post(
+        f"{_SETTLEMENTS}/generate", json={"month": _month()}, headers=admin["headers"]
+    )
+    assert resp.status_code == 200, resp.text
+    settlement_id = resp.json()["data"]["id"]
+    resp = client.post(
+        f"{_SETTLEMENTS}/{settlement_id}/submit", headers=admin["headers"]
+    )
+    assert resp.status_code == 200, resp.text
+
+    order = _dashboard(client, admin["headers"])["todos"]["pending_approvals"][0]
+    assert _dashboard(client, finance["headers"])["todos"]["pending_approval_total"] == 1
+    assert _dashboard(client, leader["headers"])["todos"]["pending_approval_total"] == 0
+
+    resp = client.post(
+        f"/api/v1/system/approvals/{order['order_id']}/action",
+        json={"action": 1},
+        headers=finance["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    assert _dashboard(client, finance["headers"])["todos"]["pending_approval_total"] == 0
+    assert _dashboard(client, leader["headers"])["todos"]["pending_approval_total"] == 1
+
+
+def test_预警_回款逾期命中且阈值参数生效(client: TestClient, admin: dict):
+    """发货超阈值且无任何回款 → 命中；调大阈值后不再命中。"""
+    _, shipment = _shipment_scene(client, admin["headers"], days_ago=120)
+
+    data = _dashboard(client, admin["headers"])
+    assert any(
+        item["code"] == "receipt_overdue" and item["object_id"] == shipment["id"]
+        for item in data["warnings"]["items"]
+    ), "120 天未回款的发货批次未命中预警"
+
+    _set_param(client, admin["headers"], "warning.receipt.overdue.days", "200")
+    data = _dashboard(client, admin["headers"])
+    assert not any(
+        item["code"] == "receipt_overdue" and item["object_id"] == shipment["id"]
+        for item in data["warnings"]["items"]
+    ), "阈值调大后仍命中，说明参数未生效"
+
+
+def test_预警_终止业务不再产生回款逾期(client: TestClient, admin: dict):
+    """已终止业务不再催收，其未回款发货单不得留在预警列表。"""
+    business, shipment = _shipment_scene(client, admin["headers"], days_ago=120)
+    resp = client.put(
+        f"{_BUSINESSES}/{business['id']}/status",
+        json={"status": 3, "reason": "t_工作台终止"},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+
+    data = _dashboard(client, admin["headers"])
+    assert not any(item["object_id"] == shipment["id"] for item in data["warnings"]["items"])
+    assert data["warnings"]["receipt_overdue_total"] == 0
+
+
+def test_预警_总数不受展示上限截断(client: TestClient, admin: dict, monkeypatch):
+    """items 只展示前 N 条，徽标 total 必须仍是全量命中数。"""
+    from app.services import dashboard_service
+
+    _shipment_scene(client, admin["headers"], days_ago=120)
+    _shipment_scene(client, admin["headers"], days_ago=120)
+    monkeypatch.setattr(dashboard_service, "_WARNING_LIMIT", 1)
+
+    data = _dashboard(client, admin["headers"])
+    assert data["warnings"]["total"] == 2
+    assert len(data["warnings"]["items"]) == 1
+    assert data["warnings"]["total"] == data["warnings"]["receipt_overdue_total"]
+
+
+def test_总览_回款趋势按到账月跨月归属(client: TestClient, admin: dict):
+    """趋势窗口按 receipt_date 范围过滤，月初/月末不得串月。"""
+    from tests.integration.factories import business_item, business_payload, customer
+
+    person = _person(client, admin["headers"])
+    cust = customer(client, admin["headers"])
+    resp = client.post(
+        _BUSINESSES,
+        json=business_payload(
+            cust["id"],
+            owner_personnel_id=person["id"],
+            items=[business_item(cost_amount="3000.00")],
+        ),
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    business = resp.json()["data"]
+
+    first_of_month = date.today().replace(day=1)
+    previous_last = first_of_month - timedelta(days=1)
+    for receipt_date in (previous_last, first_of_month):
+        resp = client.post(
+            _RECEIPTS,
+            json={
+                "business_id": business["id"],
+                "receipt_date": receipt_date.isoformat(),
+                "amount": "1000.00",
+            },
+            headers=admin["headers"],
+        )
+        assert resp.status_code == 200, resp.text
+
+    trend = {
+        item["month"]: Decimal(str(item["receipt_total"] or 0))
+        for item in _dashboard(client, admin["headers"])["overview"]["trend"]
+    }
+    assert trend[previous_last.strftime("%Y-%m")] == Decimal("1000.00")
+    assert trend[first_of_month.strftime("%Y-%m")] == Decimal("1000.00")

backend/tests/integration/test_reports.py

@@ -306,3 +306,498 @@ def test_权限_普通用户403文案(admin, normal, client: TestClient):
     resp = client.get(f"{_REPORTS}/personal-detail", headers=normal["headers"])
     assert resp.status_code == 403
     assert resp.json()["message"] == "无权访问报表统计"
+
+
+# ==== 审查回归：归属、调整单、经理个人户、导出审计 ====
+
+
+def _set_param(client: TestClient, headers: dict, code: str, value: str) -> None:
+    """立即生效的参数变更（用例不依赖种子默认值，结束显式恢复）。"""
+    resp = client.post(
+        "/api/v1/system/params/changes",
+        json={
+            "change_type": 1,
+            "remark": f"用例设置 {code}={value}",
+            "items": [{"param_code": code, "dimension_value": "", "value": value}],
+        },
+        headers=headers,
+    )
+    assert resp.status_code == 200, resp.text
+
+
+def _insert_adjustment(person: dict, month: str, amount: str, direction: int = 1) -> None:
+    """直插生效调整单（status=2）：绕开审批链，聚焦报表归属口径。"""
+    from app.core.db import get_session_factory
+    from app.models import AdjustmentOrder
+
+    factory = get_session_factory()
+    with factory() as session:
+        session.add(
+            AdjustmentOrder(
+                order_no=f"TZ{uuid.uuid4().hex[:12]}",
+                settlement_month=month,
+                personnel_id=person["id"],
+                personnel_name=person["personnel_name"],
+                direction=direction,
+                amount=Decimal(amount),
+                reason="t_报表用例",
+                status=2,
+                submitter_name="t_admin",
+            )
+        )
+        session.commit()
+
+
+def _insert_manager_personal_ledger(manager: dict, month: str, amount: str) -> None:
+    """直插一条经理个人户台账行（模拟单笔侧已计提的结果）。"""
+    from datetime import date
+
+    from app.core.db import get_session_factory
+    from app.models import PerformanceLedger
+
+    factory = get_session_factory()
+    with factory() as session:
+        session.add(
+            PerformanceLedger(
+                receipt_id=None,
+                calc_version=1,
+                calc_type=1,
+                business_id=0,
+                business_no="t_个人户",
+                personnel_id=manager["id"],
+                personnel_name=manager["personnel_name"],
+                role_code="sales_manager",
+                customer_type="manager_personal",
+                product_category="mixed",
+                receipt_date=date.today(),
+                rate_text="0",
+                performance_amount=Decimal(amount),
+                barter_applied=0,
+                payable_amount=Decimal(amount),
+                settlement_month=month,
+                is_probation=0,
+            )
+        )
+        session.commit()
+
+
+def _latest_export_log():
+    from sqlalchemy import select
+
+    from app.core.db import get_session_factory
+    from app.models import OperationLog
+
+    factory = get_session_factory()
+    with factory() as session:
+        return session.execute(
+            select(OperationLog)
+            .where(OperationLog.module == "报表统计", OperationLog.action == "导出")
+            .order_by(OperationLog.id.desc())
+        ).scalars().first()
+
+
+def test_部门汇总_无台账经理挂上级不报500(admin, client: TestClient):
+    """经理 A 无台账且上级是经理 B：不能因 member_map 直接索引而 KeyError。"""
+    manager_b = _person(client, admin["headers"], role="sales_manager")
+    manager_a = _person(
+        client, admin["headers"], role="sales_manager", manager_id=manager_b["id"]
+    )
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month()},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    data = resp.json()["data"]
+    team_a = next(t for t in data["teams"] if t["manager_id"] == manager_a["id"])
+    assert team_a["member_count"] == 0
+    team_b = next(t for t in data["teams"] if t["manager_id"] == manager_b["id"])
+    assert all(m["personnel_id"] != manager_a["id"] for m in team_b["members"])
+
+
+def test_部门汇总_仅有调整单人员并入对应团队(admin, client: TestClient):
+    """只有调整单、当月无台账行的人也要出现在团队明细中。"""
+    manager = _person(client, admin["headers"], role="sales_manager")
+    salesperson = _person(client, admin["headers"], manager_id=manager["id"])
+    _insert_adjustment(salesperson, _month(), "500.00")
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month()},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    data = resp.json()["data"]
+    team = next(t for t in data["teams"] if t["manager_id"] == manager["id"])
+    member = next(m for m in team["members"] if m["personnel_id"] == salesperson["id"])
+    assert member["member_kind"] == "adjustment"
+    assert Decimal(str(member["performance_sum"])) == Decimal("0")
+    assert Decimal(str(member["adjust_net"])) == Decimal("500.00")
+    assert Decimal(str(team["adjust_net"])) == Decimal("500.00")
+    assert Decimal(str(data["reconciliation"]["adjust_net_total"])) == Decimal("500.00")
+
+
+def test_部门汇总_仅调整单无经理人员进入未分配(admin, client: TestClient):
+    """无经理人员的调整单不能被静默丢弃。"""
+    salesperson = _person(client, admin["headers"])
+    _insert_adjustment(salesperson, _month(), "300.00")
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month()},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    data = resp.json()["data"]
+    row = next(u for u in data["unassigned"] if u["personnel_id"] == salesperson["id"])
+    assert row["member_kind"] == "adjustment"
+    assert Decimal(str(row["adjust_net"])) == Decimal("300.00")
+
+
+def test_部门汇总_经理个人户计入时基数等于成员合计(admin, client: TestClient):
+    """开关 1：经理成员行只取个人户金额，团队基数 == 成员绩效合计。"""
+    manager = _person(client, admin["headers"], role="sales_manager")
+    _insert_manager_personal_ledger(manager, _month(), "100.00")
+    _set_param(client, admin["headers"], "manager.personal.included", "1")
+    try:
+        resp = client.get(
+            f"{_REPORTS}/department-summary",
+            params={"settlement_month": _month()},
+            headers=admin["headers"],
+        )
+        assert resp.status_code == 200, resp.text
+        team = next(
+            t for t in resp.json()["data"]["teams"] if t["manager_id"] == manager["id"]
+        )
+        assert Decimal(str(team["team_base"])) == Decimal("100.00")
+        assert sum(
+            (Decimal(str(m["performance_sum"])) for m in team["members"]), Decimal("0")
+        ) == Decimal("100.00")
+        assert all(m["member_kind"] == "manager_personal" for m in team["members"])
+    finally:
+        _set_param(client, admin["headers"], "manager.personal.included", "0")
+
+
+def test_部门汇总_经理个人户关闭时不展示个人户金额(admin, client: TestClient):
+    """开关 0：个人户金额既不计入团队基数，也不进未分配（与全公司口径一致）。"""
+    manager = _person(client, admin["headers"], role="sales_manager")
+    _insert_manager_personal_ledger(manager, _month(), "100.00")
+    _set_param(client, admin["headers"], "manager.personal.included", "0")
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month()},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    data = resp.json()["data"]
+    team = next(t for t in data["teams"] if t["manager_id"] == manager["id"])
+    assert team["member_count"] == 0
+    assert all(u["personnel_id"] != manager["id"] for u in data["unassigned"])
+
+
+def test_部门汇总_按经理筛选不误判其他团队下属(admin, client: TestClient):
+    """筛选只影响展示的团队；其他经理的下属不能进入未分配。"""
+    manager_1 = _person(client, admin["headers"], role="sales_manager")
+    manager_2 = _person(client, admin["headers"], role="sales_manager")
+    salesperson_1 = _person(client, admin["headers"], manager_id=manager_1["id"])
+    salesperson_2 = _person(client, admin["headers"], manager_id=manager_2["id"])
+    _full_chain(client, admin["headers"], owner_id=salesperson_1["id"])
+    _full_chain(client, admin["headers"], owner_id=salesperson_2["id"])
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month(), "manager_id": manager_1["id"]},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    data = resp.json()["data"]
+    assert [t["manager_id"] for t in data["teams"]] == [manager_1["id"]]
+    assert data["unassigned"] == []
+    assert all(
+        m["personnel_id"] != salesperson_2["id"]
+        for team in data["teams"]
+        for m in team["members"]
+    )
+    assert Decimal(str(data["reconciliation"]["other_team_base"])) > 0
+
+
+def test_部门汇总_对账恒等式(admin, client: TestClient):
+    """未筛选：团队基数 + 未分配绩效 = 全公司基数；应发公式成立。"""
+    manager = _person(client, admin["headers"], role="sales_manager")
+    salesperson = _person(client, admin["headers"], manager_id=manager["id"])
+    unassigned = _person(client, admin["headers"])
+    _full_chain(client, admin["headers"], owner_id=salesperson["id"])
+    _full_chain(client, admin["headers"], owner_id=unassigned["id"])
+    _insert_adjustment(salesperson, _month(), "200.00")
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month()},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    reconciliation = resp.json()["data"]["reconciliation"]
+    assert Decimal(str(reconciliation["team_base_total"])) + Decimal(
+        str(reconciliation["unassigned_performance_total"])
+    ) == Decimal(str(reconciliation["company_base"]))
+    assert Decimal(str(reconciliation["other_team_base"])) == Decimal("0")
+    assert Decimal(str(reconciliation["payable_total"])) == (
+        Decimal(str(reconciliation["team_member_payable_total"]))
+        + Decimal(str(reconciliation["unassigned_payable_total"]))
+        + Decimal(str(reconciliation["aggregate_payable_total"]))
+        + Decimal(str(reconciliation["adjust_net_total"]))
+    )
+    # 分列合计 + 舍入差 = 聚合额（月结一次舍入口径）
+    assert Decimal(str(reconciliation["aggregate_rounding_diff"])) == (
+        Decimal(str(reconciliation["aggregate_payable_total"]))
+        - Decimal(str(reconciliation["team_share_total"]))
+        - Decimal(str(reconciliation["company_share_total"]))
+    )
+
+
+def _two_manager_scene(
+    client: TestClient, headers: dict
+) -> tuple[dict, dict, dict, dict, dict]:
+    """两位经理各有下属与业务 + 一位零成员经理（无聚合行）。"""
+    manager_1 = _person(client, headers, role="sales_manager")
+    manager_2 = _person(client, headers, role="sales_manager")
+    empty_manager = _person(client, headers, role="sales_manager")
+    salesperson_1 = _person(client, headers, manager_id=manager_1["id"])
+    salesperson_2 = _person(client, headers, manager_id=manager_2["id"])
+    _full_chain(client, headers, owner_id=salesperson_1["id"])
+    _full_chain(client, headers, owner_id=salesperson_2["id"])
+    return manager_1, manager_2, empty_manager, salesperson_1, salesperson_2
+
+
+def test_部门汇总_应发合计与月结对拍_两位经理(admin, client: TestClient):
+    """月结每位有聚合行的经理各计一次 5%；签字稿应发必须等于 total_payable。"""
+    _two_manager_scene(client, admin["headers"])
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month()},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    before = resp.json()["data"]
+    assert before["company"]["company_share_count"] == 2
+    assert Decimal(str(before["company"]["company_share_total"])) == (
+        Decimal(str(before["company"]["company_share_amount"])) * 2
+    )
+
+    resp = client.post(
+        f"{_SETTLEMENTS}/generate", json={"month": _month()}, headers=admin["headers"]
+    )
+    assert resp.status_code == 200, resp.text
+    settlement = resp.json()["data"]
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month()},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    data = resp.json()["data"]
+    teams = {team["manager_id"]: team for team in data["teams"]}
+    frozen_total = sum(
+        (
+            Decimal(str(team["frozen_amount"]))
+            for team in teams.values()
+            if team["frozen_amount"] is not None
+        ),
+        Decimal("0"),
+    )
+    reconciliation = data["reconciliation"]
+    assert Decimal(str(reconciliation["aggregate_payable_total"])) == frozen_total
+    assert Decimal(str(reconciliation["payable_total"])) == Decimal(
+        str(settlement["total_payable"])
+    )
+
+
+def test_部门汇总_有调整单时应发等于月结加调整(admin, client: TestClient):
+    """调整单不进月结 items；报表应发 = 月结总额 + 调整净额。"""
+    _, _, _, salesperson, _ = _two_manager_scene(client, admin["headers"])
+    _insert_adjustment(salesperson, _month(), "500.00")
+
+    resp = client.post(
+        f"{_SETTLEMENTS}/generate", json={"month": _month()}, headers=admin["headers"]
+    )
+    assert resp.status_code == 200, resp.text
+    settlement = resp.json()["data"]
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month()},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    reconciliation = resp.json()["data"]["reconciliation"]
+    assert Decimal(str(reconciliation["adjust_net_total"])) == Decimal("500.00")
+    assert Decimal(str(reconciliation["payable_total"])) == (
+        Decimal(str(settlement["total_payable"])) + Decimal("500.00")
+    )
+
+
+def test_部门汇总_按经理筛选时5百分比只计该团队(admin, client: TestClient):
+    """筛选态只汇总展示团队；隐藏团队的金额进 other_*，不参与应发。"""
+    manager_1, _, _, _, _ = _two_manager_scene(client, admin["headers"])
+
+    resp = client.get(
+        f"{_REPORTS}/department-summary",
+        params={"settlement_month": _month(), "manager_id": manager_1["id"]},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    data = resp.json()["data"]
+    assert data["company"]["company_share_count"] == 1
+    reconciliation = data["reconciliation"]
+    assert Decimal(str(reconciliation["other_team_aggregate_total"])) > 0
+    assert Decimal(str(reconciliation["other_team_company_share_total"])) > 0
+
+
+def test_导出_筛选页眉ID转名称与中文标签(admin, client: TestClient):
+    """页眉是签字稿追溯筛选范围的唯一凭据，ID 必须转名称。"""
+    from io import BytesIO
+
+    from openpyxl import load_workbook
+
+    from app.core.db import get_session_factory
+    from app.models import BusinessRole
+
+    person = _person(client, admin["headers"])
+    _full_chain(client, admin["headers"], owner_id=person["id"])
+    factory = get_session_factory()
+    with factory() as session:
+        role_name = (
+            session.query(BusinessRole.role_name)
+            .filter(BusinessRole.role_code == "salesperson")
+            .scalar()
+        )
+
+    resp = client.get(
+        f"{_REPORTS}/personal-detail/export",
+        params={
+            "settlement_month": _month(),
+            "personnel_id": person["id"],
+            "role_code": "salesperson",
+            "customer_type": "personal",
+            "product_category": "municipal_water",
+        },
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    workbook = load_workbook(BytesIO(resp.content))
+    text = "\n".join(
+        str(cell.value)
+        for row in workbook.active.iter_rows()
+        for cell in row
+        if cell.value is not None
+    )
+    assert person["personnel_name"] in text
+    assert f"人员 {person['personnel_name']}(#{person['id']})" in text
+    assert "个人户" in text
+    assert "市政水利" in text
+    assert role_name in text
+
+
+def test_导出_回款维度sheet透传筛选条件(admin, client: TestClient):
+    """两个维度 sheet 的页眉不能固定写「全部」。"""
+    from io import BytesIO
+
+    from openpyxl import load_workbook
+
+    resp = client.get(
+        f"{_REPORTS}/payment-stats/export",
+        params={"customer_type": "personal"},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 200, resp.text
+    workbook = load_workbook(BytesIO(resp.content))
+    for sheet_name in ("客户类型对比", "产品类别对比"):
+        text = "\n".join(
+            str(cell.value)
+            for row in workbook[sheet_name].iter_rows()
+            for cell in row
+            if cell.value is not None
+        )
+        assert "客户类型 个人户" in text
+        assert "筛选条件：全部" not in text
+
+
+def test_导出超限_失败审计留痕(admin, client: TestClient, monkeypatch):
+    """超限 400 发生在审计上下文内，必须留下失败记录。"""
+    from app.services import report_service
+
+    person = _person(client, admin["headers"])
+    _full_chain(client, admin["headers"], owner_id=person["id"])
+    monkeypatch.setattr(report_service, "REPORT_EXPORT_MAX_ROWS", 0)
+
+    resp = client.get(
+        f"{_REPORTS}/personal-detail/export",
+        params={"settlement_month": _month()},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 400, resp.text
+    log = _latest_export_log()
+    assert log is not None
+    assert log.success == 0
+    assert "超过上限" in (log.error_message or "")
+
+
+def test_导出取数异常_失败审计留痕(admin, client: TestClient, monkeypatch):
+    """取数阶段的异常同样要留下失败记录。
+
+    TestClient 默认重抛服务端异常，这里直接调 service 层，聚焦审计落库。
+    """
+    from app.core.db import get_session_factory
+    from app.models import User
+    from app.schemas.report import PersonalDetailQuery
+    from app.services import report_service
+
+    def _boom(*args, **kwargs):
+        raise RuntimeError("t_取数异常")
+
+    monkeypatch.setattr(report_service, "_export_payload", _boom)
+    factory = get_session_factory()
+    with factory() as session:
+        operator = session.get(User, admin["id"])
+        with pytest.raises(RuntimeError, match="t_取数异常"):
+            report_service.export_report(
+                session,
+                "personal-detail",
+                PersonalDetailQuery(settlement_month=_month()),
+                operator=operator,
+            )
+    log = _latest_export_log()
+    assert log is not None
+    assert log.success == 0
+    # 未预期异常按安全口径收敛为通用文案，摘要仍能定位到报表
+    assert "导出个人绩效工资明细表" in (log.summary or "")
+
+
+@pytest.mark.parametrize(
+    "path", ["/personal-detail", "/department-summary", "/payment-stats"]
+)
+def test_非法月份_查询返回400(admin, client: TestClient, path: str):
+    """非法月份必须是参数错误 400，不能落 500。"""
+    resp = client.get(
+        f"{_REPORTS}{path}",
+        params={"settlement_month": "2026-13"},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 400, resp.text
+
+
+@pytest.mark.parametrize(
+    "code", ["personal-detail", "department-summary", "payment-stats"]
+)
+def test_非法月份_导出返回400(admin, client: TestClient, code: str):
+    """导出手工构造查询模型同样要转 400。"""
+    resp = client.get(
+        f"{_REPORTS}/{code}/export",
+        params={"settlement_month": "2026-13"},
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 400, resp.text

backend/tests/integration/test_settlements.py

@@ -749,3 +749,24 @@ class TestLockInterception:
         resp = client.post(f"{_CALC}/receipts/{unsettled['id']}/calc", headers=admin["headers"])
         assert resp.status_code == 400, resp.text
         assert "锁定" in resp.json()["message"]
+
+
+def test_非法月份_月结生成与调整单均返回400(client: TestClient, admin: dict) -> None:
+    """月份范围必须在 schema 拦住：`2026-13` 不能传到 monthrange 变成 500。"""
+    resp = client.post(
+        f"{_SETTLEMENTS}/generate", json={"month": "2026-13"}, headers=admin["headers"]
+    )
+    assert resp.status_code == 400, resp.text
+
+    resp = client.post(
+        f"{_SETTLEMENTS}/adjustments",
+        json={
+            "settlement_month": "2026-13",
+            "personnel_id": 1,
+            "direction": 1,
+            "amount": "100.00",
+            "reason": "t_非法月份",
+        },
+        headers=admin["headers"],
+    )
+    assert resp.status_code == 400, resp.text

backend/tests/unit/test_report_helpers.py

@@ -209,3 +209,214 @@ class TestExcel模板:
         ]
         text = _sheet_text(_build("business-stats", rows))
         assert "已完成" in text
+
+
+def _member(
+    personnel_id: int,
+    name: str,
+    *,
+    kind: str = "regular",
+    performance: str = "0.00",
+    payable: str = "0.00",
+    adjust: str = "0.00",
+    reason: str | None = None,
+) -> dict:
+    return {
+        "personnel_id": personnel_id,
+        "personnel_name": name,
+        "role_code": "salesperson",
+        "is_probation": 0,
+        "member_kind": kind,
+        "unassigned_reason": reason,
+        "performance_sum": Decimal(performance),
+        "payable_sum": Decimal(payable),
+        "adjust_extra": Decimal(adjust),
+        "adjust_deduct": Decimal("0"),
+        "adjust_net": Decimal(adjust),
+        "net_total": Decimal(payable) + Decimal(adjust),
+        "ledger_ids": [],
+    }
+
+
+def _department_data(
+    team_text: str = "12.5",
+    company_text: str = "5",
+    *,
+    second_has_aggregate: bool = False,
+) -> dict:
+    """一份含成员、零成员团队与未分配人员的部门汇总出参。
+
+    ``second_has_aggregate`` 打开后两位经理都有月结聚合行，用于验证
+    全公司 5% 按人数累计（零成员团队默认不发 5%）。
+    """
+    second_aggregate = Decimal("600.00") if second_has_aggregate else Decimal("0")
+    share_count = 2 if second_has_aggregate else 1
+    aggregate_total = Decimal("1850.00") + second_aggregate
+    payable_total = Decimal("10000.00") + Decimal("2000.00") + aggregate_total
+    return {
+        "settlement_month": "2026-10",
+        "params": {
+            "team_text": team_text,
+            "company_text": company_text,
+            "manager_personal_included": 0,
+            "params_as_of": "2026-10-31",
+        },
+        "teams": [
+            {
+                "manager_id": 1,
+                "manager_name": "张经理",
+                "manager_status": 1,
+                "member_count": 1,
+                "team_base": Decimal("10000.00"),
+                "team_share_amount": Decimal("1250.00"),
+                "company_share_amount": Decimal("600.00"),
+                "has_aggregate": True,
+                "aggregate_amount": Decimal("1850.00"),
+                "team_subtotal": Decimal("1850.00"),
+                "company_base_ratio": Decimal("0.8333"),
+                "adjust_extra": Decimal("0"),
+                "adjust_deduct": Decimal("0"),
+                "adjust_net": Decimal("0"),
+                "net_total": Decimal("1850.00"),
+                "frozen_amount": None,
+                "aggregate_ledger_id": None,
+                "members": [
+                    _member(
+                        2,
+                        "李业务",
+                        performance="10000.00",
+                        payable="10000.00",
+                    )
+                ],
+            },
+            {
+                "manager_id": 3,
+                "manager_name": "王经理",
+                "manager_status": 1,
+                "member_count": 0,
+                "team_base": Decimal("0"),
+                "team_share_amount": Decimal("0"),
+                "company_share_amount": second_aggregate,
+                "has_aggregate": second_has_aggregate,
+                "aggregate_amount": second_aggregate,
+                "team_subtotal": second_aggregate,
+                "company_base_ratio": Decimal("0"),
+                "adjust_extra": Decimal("0"),
+                "adjust_deduct": Decimal("0"),
+                "adjust_net": Decimal("0"),
+                "net_total": second_aggregate,
+                "frozen_amount": None,
+                "aggregate_ledger_id": None,
+                "members": [],
+            },
+        ],
+        "unassigned": [
+            _member(
+                4,
+                "赵未分配",
+                performance="2000.00",
+                payable="2000.00",
+                reason="无经理或上级非营销经理",
+            )
+        ],
+        "company": {
+            "team_base_total": Decimal("10000.00"),
+            "company_base": Decimal("12000.00"),
+            "company_share_pct": company_text,
+            "company_share_amount": Decimal("600.00"),
+            "company_share_count": share_count,
+            "company_share_total": Decimal("600.00") * share_count,
+        },
+        "reconciliation": {
+            "team_member_payable_total": Decimal("10000.00"),
+            "unassigned_payable_total": Decimal("2000.00"),
+            "team_share_total": Decimal("1250.00"),
+            "company_share_total": Decimal("600.00") * share_count,
+            "aggregate_payable_total": aggregate_total,
+            "aggregate_rounding_diff": Decimal("0"),
+            "adjust_net_total": Decimal("0"),
+            "payable_total": payable_total,
+            "team_base_total": Decimal("10000.00"),
+            "unassigned_performance_total": Decimal("2000.00"),
+            "other_team_base": Decimal("0"),
+            "other_team_member_payable_total": Decimal("0"),
+            "other_team_share_total": Decimal("0"),
+            "other_team_company_share_total": Decimal("0"),
+            "other_team_aggregate_total": Decimal("0"),
+            "other_adjust_net_total": Decimal("0"),
+            "company_base": Decimal("12000.00"),
+        },
+        "totals": {},
+    }
+
+
+def _build_department(data: dict) -> bytes:
+    return report_export.build_workbook(
+        code="department-summary",
+        title="部门绩效工资汇总表",
+        filters="统计月份 2026-10",
+        rows=[],
+        summary=None,
+        buckets=None,
+        department=data,
+        template_version="V1.2",
+        operator_name="admin",
+        company_name="衍水集团",
+    )
+
+
+class Test部门汇总签字稿:
+    """签字稿少了团队分成/5% 块/未分配/零成员团队，合计就会小于应发。"""
+
+    def test_五块顺序完整(self):
+        text = _sheet_text(_build_department(_department_data()))
+        titles = [
+            "一、团队分成汇总",
+            "二、全公司分成（每位有聚合行",
+            "三、对账汇总",
+            "四、团队成员明细",
+            "五、未分配人员明细",
+        ]
+        positions = [text.index(title) for title in titles]
+        assert positions == sorted(positions), "签字稿块顺序错乱"
+        assert "财务复核" in text and "领导签字" in text
+
+    def test_零成员团队与未分配行存在(self):
+        text = _sheet_text(_build_department(_department_data()))
+        assert "王经理" in text
+        assert "赵未分配" in text
+        assert "无经理或上级非营销经理" in text
+
+    def test_全公司分成按计入次数累计(self):
+        """两位有聚合行的经理各计一次 5%；零成员团队不产生分成。"""
+        one = _department_data()
+        text = _sheet_text(_build_department(one))
+        assert text.count("全公司分成（每位有聚合行") == 1
+        assert "共 1 次" in text
+        assert "13850" in text
+
+        two = _department_data(second_has_aggregate=True)
+        text = _sheet_text(_build_department(two))
+        assert "共 2 次" in text
+        # 团队成员应付 10000 + 未分配 2000 + 聚合 2450
+        assert "14450" in text
+
+    def test_应发合计取对账口径(self):
+        text = _sheet_text(_build_department(_department_data()))
+        assert "应发合计" in text
+        # Excel 数值单元格读回为 int/float，金额断言只看数值本身
+        assert "13850" in text
+
+    def test_聚合行舍入差单列(self):
+        data = _department_data()
+        data["reconciliation"]["aggregate_rounding_diff"] = Decimal("-0.02")
+        text = _sheet_text(_build_department(data))
+        assert "聚合行舍入差" in text
+        assert "-0.02" in text
+
+    def test_比例标题参数化(self):
+        text = _sheet_text(_build_department(_department_data("20", "6")))
+        assert "团队分成（20%）" in text
+        assert "6%" in text
+        assert "12.5%" not in text
+        assert "5%" not in text.replace("6%", "")

docs/superpowers/plans/2026-10-12-剩余开发计划.md

@@ -45,12 +45,12 @@
 ② 基础数据管理 ✅                  ← 人员引用 ① 岗位角色（外键上游）
 ③ 业务数据管理 ✅                  ← 引用 ② 人员/产品/客户
 │                                    增量（2026-10-15）：二次分配方案（合同刚建立时录入）
-④ 绩效工资核算                     ← 消费 ① 参数快照（按回款到账日）+ ②③ 数据
-│                                    增量（2026-10-15）：月度结算/调整单/负数工资；二次分配分摊（语义待澄清）
-⑦ 复核审批流（新）                 ← 月结/二次分配的审批链（营销后勤→财务→领导）；登录角色体系扩展
-⑤ 报表统计（∥ ⑥ T+ 同步）         ← ⑤ 只读消费 ③④；签字导出模板依赖 ⑦ 与公司信息；⑥ 受 IT 字段映射制约
-⑧ 工作台（新）                     ← 消费 ④⑤⑥⑦ 数据的聚合展示层
-⑨ 备份恢复与运维检查（新）         ← 独立于业务块，可与任一批次并行，建议尽早
+④ 绩效工资核算 ✅                  ← 消费 ① 参数快照（按回款到账日）+ ②③ 数据
+│                                    增量（2026-10-15）：月度结算/调整单/负数工资；二次分配分摊（#1–#3 答复空白，开发侧暂定口径已实现，见 3.3.1）
+⑦ 复核审批流（新）✅               ← 月结/二次分配的审批链（营销后勤→财务→领导）；登录角色体系扩展
+⑤ 报表统计 ✅（∥ ⑥ T+ 同步）       ← ⑤ 只读消费 ③④；签字导出模板依赖 ⑦ 与公司信息；⑥ 受 IT 字段映射制约（待 IT 18 项答复）
+⑧ 工作台（新）✅                   ← 消费 ④⑤⑥⑦ 数据的聚合展示层
+⑨ 备份恢复与运维检查（新）✅       ← 独立于业务块，可与任一批次并行，建议尽早
 ```
 
 ### 1.2 顺序论证（要点）
@@ -68,12 +68,12 @@
 | ① 系统设置与参数体系 | 岗位角色 CRUD + 模板绑定；参数目录/版本化存储/按日期取值/变更留痕/对比/导出；公式纯函数；计算预览。**✅ 已完成**。增量（2026-10-15）：公司信息配置（公司名称、logo 等基本信息）；审批开关配置走参数体系；新增参数一律进参数注册表（用户重申：参数配置尽量灵活） | （原验收已达成）增量：公司信息可配置并用于登录页与导出页眉；审批开关参数化，关闭直接生效、开启走审批 |
 | ② 基础数据管理 | 人员（岗位/工资标准/团队/辐射区域/拓展计划）、产品类型与目录、业务(客户)与归属变更。**✅ 已完成** | 三类数据可维护并留痕；人员岗位引用 ① 岗位表；表含 T+ 预留列；客户分类（个人户/经理个人户/企业户）与归属修改不回溯历史 |
 | ③ 业务数据管理 | 业务明细+指导价、发货（批次/运费/发货日）、回款（批次号/到账日/以物抵顶/手动指定发货批次）、毛利润（不含运费）、客户评级计算器。**✅ 已完成**。增量（2026-10-15）：二次分配方案 —— 只出现在合同刚建立时；合同建立时可手动选择参与二次分配的业务员；分配比例不要求合计 100%；审批开关开启时走审批 | （原验收已达成）增量：合同建立时可录入参与业务员与比例（不强制合计 100%）；开关关闭直接生效；方案修改不影响已结算月份 |
-| ④ 绩效工资核算 | 核算引擎（按回款到账日取参数）、单笔台账+核销防重、经理 12.5%+5%、试用期/年累计 18000、起付 60% 与 R×R、预付款/以物抵顶、月工资（分岗口径）、重算授权留痕。增量（2026-10-15）：月度结算（按确认回款和计算快照生成）；支持负数工资、扣补说明和调整单；结算后锁定 —— 修改源数据不影响已结算结果，只产生提示、历史记录和后续调整；二次分配分摊（语义待澄清，见 3.3） | 任一笔回款可复算到分（零误差）；同笔回款不重复核算；08-13 前后企业户 2‰/3‰ 回溯正确；月工资表可生成。增量：任一月份可出月结快照并锁定；已锁月份源数据变更只产生提示与历史，金额不变；调整单留痕可追溯 |
-| ⑤ 报表统计 | 个人绩效明细、部门汇总、业务统计、回款区间统计、Excel 导出与打印。增量（2026-10-15）：导出模板围绕财务复核和领导签字设计 | 四表口径与台账一致且可下钻单笔；导出精度与页面一致。增量：导出模板含复核/签字栏与公司信息页眉 |
-| ⑥ T+ 数据同步 | 五类数据抓取/清洗/入库、唯一键去重、失败重试、同步日志、周定时+手动触发 | 手动同步幂等（重跑不重复入库）；失败有日志与重试；不覆盖人工字段 |
-| ⑦ 复核审批流（新） | 月结与二次分配的审批链：营销后勤提交 → 财务复核 → 领导审核；审批开关配置（参数化，关闭直接生效）；审批意见与版本；登录角色体系扩展（营销后勤/财务/领导为系统新登录角色，激活闲置的 permissions/role_permissions） | 开关关闭时提交即生效；开启时三级流转全程留痕；每级意见与版本可追溯；重算/改单产生新版本，不覆盖既有审批记录 |
-| ⑧ 工作台（新） | 经营绩效总览、待办事项、风险预警、最近同步/计算/审批动态 | 四区块出数口径与台账/审批/同步日志一致可下钻；预警规则参数化可配置 |
-| ⑨ 备份恢复与运维检查（新） | MySQL 定时备份与恢复演练、运维检查（容器/磁盘/日志/Redis/同步状态）、检查清单与脚本 | 任一备份可按文档恢复成功；检查项脚本化输出健康报告 |
+| ④ 绩效工资核算 | 核算引擎（按回款到账日取参数）、单笔台账+核销防重、经理 12.5%+5%、试用期/年累计 18000、起付 60% 与 R×R、预付款/以物抵顶、月工资（分岗口径）、重算授权留痕。增量（2026-10-15）：月度结算（按确认回款和计算快照生成）；支持负数工资、扣补说明和调整单；结算后锁定 —— 修改源数据不影响已结算结果，只产生提示、历史记录和后续调整；二次分配分摊（#1–#3 答复空白，开发侧暂定口径已实现，见 3.3.1） | 任一笔回款可复算到分（零误差）；同笔回款不重复核算；08-13 前后企业户 2‰/3‰ 回溯正确；月工资表可生成。增量：任一月份可出月结快照并锁定；已锁月份源数据变更只产生提示与历史，金额不变；调整单留痕可追溯 |
+| ⑤ 报表统计 | 个人绩效明细、部门汇总、业务统计、回款区间统计、Excel 导出与打印。增量（2026-10-15）：导出模板围绕财务复核和领导签字设计。**✅ 已完成（2026-10-19）** | 四表口径与台账一致且可下钻单笔；导出精度与页面一致。增量：导出模板含复核/签字栏与公司信息页眉 |
+| ⑥ T+ 数据同步 | 五类数据抓取/清洗/入库、唯一键去重、失败重试、同步日志、周定时+手动触发。**⏳ 阻塞（IT 18 项未答复）** | 手动同步幂等（重跑不重复入库）；失败有日志与重试；不覆盖人工字段 |
+| ⑦ 复核审批流（新） | 月结与二次分配的审批链：营销后勤提交 → 财务复核 → 领导审核；审批开关配置（参数化，关闭直接生效）；审批意见与版本；登录角色体系扩展（营销后勤/财务/领导为系统新登录角色，激活闲置的 permissions/role_permissions）。**✅ 已完成（2026-10-19；权限细目待业务清单）** | 开关关闭时提交即生效；开启时三级流转全程留痕；每级意见与版本可追溯；重算/改单产生新版本，不覆盖既有审批记录 |
+| ⑧ 工作台（新） | 经营绩效总览、待办事项、风险预警、最近同步/计算/审批动态。**✅ 已完成（2026-10-19）** | 四区块出数口径与台账/审批/同步日志一致可下钻；预警规则参数化可配置 |
+| ⑨ 备份恢复与运维检查（新） | MySQL 定时备份与恢复演练、运维检查（容器/磁盘/日志/Redis/同步状态）、检查清单与脚本。**✅ 已完成（2026-10-16）** | 任一备份可按文档恢复成功；检查项脚本化输出健康报告 |
 
 ### 1.4 里程碑（交付批次）
 
@@ -81,9 +81,9 @@
 | --- | --- | --- |
 | **M1** | ① 系统设置与参数体系 | ✅ 已完成（2026-10-13 推送 feat/system-params） |
 | **M2** | ② + ③ | ✅ 已完成（2026-10-13 分支 feat/base-business-data） |
-| **M3** | ① 增量（公司信息/审批开关）+ ③ 增量（二次分配录入）+ ④ 绩效工资核算（含月结/调整单/负数工资）+ ⑦ 复核审批流 | 核心闭环「核算 → 提交 → 财务复核 → 领导审核 → 锁定」；开工前必须澄清：二次分配分摊语义（3.3-1）、经理个人户基数（开放问题 #7）；内部建议顺序 ①增量 → ③增量 → ⑦ 审批框架 → ④ 核算 → 月结与审批串联 |
-| **M4** | ⑤ 报表（含财务复核/领导签字导出模板）+ ⑥ T+ 同步（并行）+ ⑧ 工作台 | ⑥ 受 IT 制约，谁先就绪谁先交；⑧ 消费 ④⑤⑥⑦ 出数，排其后 |
-| **⑨ 备份恢复与运维检查** | 不占独立批次，随任一批次并行（建议 M3 期间并行交付） | 数据持续积累，备份先行；不依赖业务模块进度 |
+| **M3** | ① 增量（公司信息/审批开关）+ ③ 增量（二次分配录入）+ ④ 绩效工资核算（含月结/调整单/负数工资）+ ⑦ 复核审批流。**✅ 已完成（2026-10-19，feat/m3-increments）** | 核心闭环「核算 → 提交 → 财务复核 → 领导审核 → 锁定」已交付；余 3.3-1 答复与审批权限细目。开工前澄清项：二次分配分摊语义（3.3-1）、经理个人户基数（开放问题 #7，已参数化） |
+| **M4** | ⑤ 报表（含财务复核/领导签字导出模板）+ ⑥ T+ 同步（并行）+ ⑧ 工作台。**✅ ⑤⑧ 已完成（2026-10-19）；⑥ 待 IT 18 项答复** | ⑥ 受 IT 制约，谁先就绪谁先交；⑧ 消费 ④⑤⑥⑦ 出数，排其后 |
+| **⑨ 备份恢复与运维检查** | 不占独立批次，随任一批次并行（建议 M3 期间并行交付）。**✅ 已完成（2026-10-16）** | 数据持续积累，备份先行；不依赖业务模块进度 |
 
 ### 1.5 需求依据对照
 
@@ -95,7 +95,7 @@
 | ④ | 需规 2.3.1–2.3.3；对照表 #3/#4/#6/#7/#11/#15/#16/#17/#20–#24；增量见 3.1(一)(四) | 协助 25%；经理 8‰；26 天示例；含运费公式 |
 | ⑤ | 需规 2.4.1–2.4.3；对照表 #22（明细展示 R）；增量见 3.1(四) | — |
 | ⑥ | 需规各「同步 T+」条；对照表 #10/#26/#27 | — |
-| ⑦⑧⑨ 及各块增量 | **2026-10-15 功能清单核查补充（本文第三章，原文照录）**——尚未回填需规与对照表，引用以第三章原文为准 | 二次分配 ≠ 协助 25%：#15 作废的是「协助 25% 分成」条款，二次分配是合同建立时的新机制，关系待业务澄清（3.3-1） |
+| ⑦⑧⑨ 及各块增量 | **2026-10-15 功能清单核查补充（本文第三章，原文照录）**——已随 SRS V2.1（2026-10-19）回填；《问题与解决方案对照表》是否回填由用户决定 | 二次分配 ≠ 协助 25%：#15 作废的是「协助 25% 分成」条款，二次分配是合同建立时的新机制；#1–#3 答复仍空白，开发侧暂定口径已实现（3.3.1） |
 
 ---
 
@@ -610,11 +610,13 @@ git commit -m "test(system): 参数体系模块用例与测试记录"
 | 6 | 导出格式与列集合（是否含批次说明/操作人） | 以需规 2.3.1(四) 字段为准，导出前与业务对一次列 |
 | 7 | 经理个人户业绩如何进入 12.5%+5%（裁决 #7 的具体衔接：该业绩是否计入「本人带领业务员」基数） | ④ 开工前必须确认（影响台账归属字段设计） |
 
+> **2026-10-20 回填**：上述开放问题均已有结论，原表保留为 2026-10-12 设计时快照。#1 → 澄清 #14（首日整 10‰，A 口径）；#2 → #18（按 2‰ 回溯，双种子保留）；#3 → #15（一期仅 FULL，BALANCE 留参数位）；#4 → #16（撤 `probation.salary.basic`，读 `personnel.salary_standard`）；#5 → #17（查看放开登录用户、修改仅管理员）；#6 → 催办清单三、R3；#7 → #12（`manager.personal.included` 参数，默认 0=不计入）。
+
 ---
 
 ## 第三章 2026-10-15 功能清单核查补充（新增需求）
 
-> **来源与地位：** 用户以现有功能清单核查后补充的需求（2026-10-15）。本章 3.1 为**原文照录**（未做任何改写），3.2 起为开发侧归属与推演，**两者严格区分**；引用新需求一律以 3.1 原文为准。本批内容尚未回填需规文档与《问题与解决方案对照表》，是否回填由用户决定。任务级实施计划在对应里程碑开工前按第二章同粒度补齐。
+> **来源与地位：** 用户以现有功能清单核查后补充的需求（2026-10-15）。本章 3.1 为**原文照录**（未做任何改写），3.2 起为开发侧归属与推演，**两者严格区分**；引用新需求一律以 3.1 原文为准。本批内容已随《需求规格说明书》V2.1（2026-10-19）回填；《问题与解决方案对照表》是否回填由用户决定。任务级实施计划在对应里程碑开工前按第二章同粒度补齐。
 
 ### 3.1 需求原文（照录）
 
@@ -656,6 +658,7 @@ git commit -m "test(system): 参数体系模块用例与测试记录"
 #### （六）系统设置
 
 - 公司信息配置：公司名称、logo 等基本信息
+- **2026-10-20 回填**：参数注册表已扩至 22 项（V8/V10 增补审批开关、试用期迁移、预警阈值等）
 
 #### （七）参数配置强调
 
@@ -666,7 +669,7 @@ git commit -m "test(system): 参数体系模块用例与测试记录"
 #### (一) 二次分配方案 → ③ 录入 + ④ 分摊消费 + ⑦ 审批联动 + ④ 调整单衔接
 
 - 「只出现在合同刚建立时」→ 录入入口放业务创建流程（与 `businesses` 创建同事务或紧随其后）；合同建立后是否允许补录/修改未明说，方案修改的留痕方式待设计；
-- 「分配比例不要求合计 100%」→ 余额归属规则需求未说明（3.3-1），存储上只存各参与人比例、不强制校验合计；
+- 「分配比例不要求合计 100%」→ 余额归属规则需求未说明（3.3-1）；存储上只存各参与人比例、不强制校验合计。**2026-10-20 回填**：#1–#3 答复仍空白，开发侧已按暂定口径实现分摊拆行（余额归负责人、行合计守恒，见 3.3.1），答复后如有变更走口径调整；
 - 「修改不影响已结算月份」→ 月结快照天然隔离（快照固化当月结果），方案表需支持编辑 + 修改留痕；
 - 「已结算月份调整只能通过调整单或后续月份扣补」→ 调整单为统一承载机制，见 (四)。
 
@@ -676,7 +679,7 @@ git commit -m "test(system): 参数体系模块用例与测试记录"
 | --- | --- |
 | 经营绩效总览 | ④ 台账/月结（按人/团队/月份汇总） |
 | 待办事项 | ⑦ 待审批单、⑥ 同步失败、回款未指定批次（M2 已有该状态可复用） |
-| 风险预警 | 规则未定义（3.3-5），先落参数化规则容器 |
+| 风险预警 | 规则未定义（3.3-5），先落参数化规则容器。**2026-10-20 回填**：方向已答复参数化（#10），首批三条按默认落 V10 参数（逾期 90 天/审批超时 3 天/同步失败待 ⑥） |
 | 最近同步/计算/审批动态 | ⑥ 同步日志 + ④ 核算/月结记录 + ⑦ 审批记录 |
 
 #### (三) 复核审批 → ⑦ 新模块
@@ -687,9 +690,11 @@ git commit -m "test(system): 参数体系模块用例与测试记录"
 - 「审批意见与版本」→ 审批单带版本号：重算/改单产生新版本，旧版本与意见留痕不覆盖；
 - **登录角色体系扩展**：营销后勤/财务/领导是系统新登录角色（`roles` 表基础设施已有；`permissions`/`role_permissions` 自 M1 起闲置待权限清单，此处激活）——打破一期「营销人员不登录」的前提，但与该原则不矛盾（营销人员仍不登录，后勤/财务/领导是职能角色）。
 
+> **2026-10-20 回填**：审批框架与月结审批对象已交付（2026-10-19）；权限点清单仍待业务方给出（澄清 #6，催办 B-6），当前按 role_code 粗粒度运行，营销后勤暂不可见审批列表。
+
 #### (四) 月结、审批、报表 → ④ 月结子块 + ⑦ 审批 + ⑤ 导出模板
 
-- 月度结算 = 台账之上的月度汇总快照（按月生成、含状态机：草稿/待审/已审定/已锁定）；「确认回款」定义待澄清（3.3-3）；
+- 月度结算 = 台账之上的月度汇总快照（按月生成、含状态机：草稿/待审/已审定/已锁定）；「确认回款」已答复为人工确认、可手动关闭（3.3-3），月结按已确认回款取数；
 - 「支持负数工资」与既有裁决「负绩效按公式正常计算（不归零）」一致，无需改公式口径；
 - 调整单：新单据类型（补发/扣减方向、关联月份、扣补说明、审批状态）；
 - 「结算后锁定」：已锁月份的快照金额不可变；源数据修改只产生提示、历史记录和后续调整 —— 与 #19「归属变更不回溯历史」同一哲学，扩展到全部源数据；
@@ -708,6 +713,8 @@ git commit -m "test(system): 参数体系模块用例与测试记录"
 
 ### 3.3 待澄清问题（不阻塞收录，对应模块开工前须澄清）
 
+> **回填状态（2026-10-20）：** 下表保留 2026-10-15 原始问题与建议默认（历史快照）；业务答复与落实情况见 3.3.1。3.3-1 仍无业务答复，开发侧已按暂定口径实现分摊拆行，答复后如有变更按口径调整。
+
 | # | 事项 | 影响模块 | 建议默认处理 |
 | --- | --- | --- | --- |
 | 1 | 二次分配的**分配标的**（绩效金额？折算毛利润？业绩额？）与**余额归属**（合计 ≠100% 时差额归谁）；与已作废的「协助 25%」（#15）是何关系 | ④ 分摊逻辑与台账归属字段 | 未确认前 ④ 只落录入与存储，不实现分摊 |
@@ -718,27 +725,41 @@ git commit -m "test(system): 参数体系模块用例与测试记录"
 | 6 | 调整单是否走同一审批链；能否跨月调整已锁定月份 | ④⑦ | 建议走同一审批链；「后续月份扣补」为默认路径，跨月调整需授权留痕 —— 待确认 |
 | 7 | 待办事项的范围（除审批/同步失败外还含哪些，回款未指定批次是否算待办） | ⑧ | 按 3.2(二) 默认清单起步，业务确认后扩充 |
 
-### 3.3.1 答复回填（2026-10-19）
+### 3.3.1 答复回填（2026-10-19，2026-10-20 修订）
 
-> 业务方 24 项答复已到（2026-10-16，回执见《待澄清问题清单_2026-10-15(1)(2).xlsx》H 列），2026-10-19 全部落实并实现；本节为 3.3 各项的落定状态，供后续引用。答复与默认不同处以答复为准；仍未答复项见 `docs/待答复催办清单_2026-10-19.md`。
+> **修订说明（2026-10-20）：** 2026-10-19 版本把 24 项统一写成「全部落实并实现」，与 3.3-1 答复空白、开发侧暂定分摊的事实不符。本次按「业务答复 / 落实状态 / 关联项」重写；3.3-1 明确标注为开发侧暂定口径，不视为业务确认。
+>
+> 业务方 24 项答复于 2026-10-16 到达（回执见《待澄清问题清单_2026-10-15(1)(2).xlsx》H 列）。分组：确认按建议默认 13 项（#7/#8/#9/#10/#11/#13/#15/#17/#18/#21/#22/#23/#24，其中 #22 为模糊确认按不计入）；与默认不同、已按答复落实 4 项（#5/#14/#16/#19）；部分答复 3 项（#4/#6/#12）；仍开放 4 项（#1/#2/#3/#20）。未答复项对外跟催见 `docs/待答复催办清单_2026-10-19.md`。
 
-| 3.3# | 事项 | 落定状态 |
-| --- | --- | --- |
-| 1 | 二次分配标的与余额归属（含与 #15 协助 25% 关系） | **答复仍空白；2026-10-19 已按建议默认实现分摊拆行**——分配标的=绩效金额份额（参与人份额×比例、负责人拿余额、行合计守恒、合计>100% 与负责人自分 400 拦截）；协助 25% 维持仅记录不核算；业务答复后如有变更走口径调整 |
-| 2 | 二次分配审批开关默认值 | 已答复：**默认开启**，可手动关闭（已落实 V8 参数种子） |
-| 3 | 「确认回款」定义 | 已答复：需人工确认动作，可手动关闭（系统已按此预留并实现） |
-| 4 | 审批角色权限清单 | 部分答复：先预设角色，名称与职权待客户确认；现按最小授权运行（后勤=提交录入，财务=复核+导出，领导=审核只读），营销后勤暂不可见审批列表 |
-| 5 | 风险预警规则 | 已答复：方向确认参数化；首批三条（逾期 90 天/同步失败/超时 3 天）未点名，按默认起步——⑧ 工作台实现 |
-| 6 | 调整单是否走同一审批链；能否跨月调整已锁月份 | 已答复：同一审批链；跨月调整需管理员授权并留痕（已实现） |
-| 7 | 待办事项范围 | 已答复：待审批 + 同步失败 + 回款未指定批次起步——⑧ 工作台实现 |
+| 3.3# | 事项 | 业务答复（2026-10-16） | 落实状态（2026-10-20） | 关联项 / 证据 |
+| --- | --- | --- | --- | --- |
+| 1 | 二次分配标的与余额归属（含与 #15 协助 25% 的关系） | **未答复**（澄清 #1–#3 空白） | **开发侧暂定口径已实现**（非业务答复）：分配标的=绩效金额份额；参与人=份额×比例，负责人=全额−已分累计（余额归负责人）；行合计守恒；合计>100% 与负责人自分 400 拦截；重算作废全部拆行。协助 25% 维持仅记录不核算。**答复后校准** | 澄清 #1/#2/#3；催办一、A-1 至 A-3；实现 ccfd206 |
+| 2 | 二次分配审批开关默认值 | 默认开启，可手动关闭 | 已落实：V8 参数种子默认开启；关闭时提交即生效（自动通过留痕） | 澄清 #5；`approval.secondary_allocation.enabled` |
+| 3 | 「确认回款」定义 | 需人工确认动作，可手动关闭 | 已落实：月结按已确认回款取数；回款确认状态可手动关闭 | 澄清 #8 |
+| 4 | 审批角色权限清单 | 部分答复：先预设角色，名称与职权待客户确认 | 按最小授权运行：后勤=提交录入，财务=复核+导出，领导=审核只读；营销后勤暂不可见审批列表；permissions/role_permissions 细目待清单 | 澄清 #6；催办一、B-6 |
+| 5 | 风险预警规则 | 方向确认参数化；首批三条未点名 | 已按默认三条实现：`warning.enabled`、`warning.receipt.overdue.days=90`、`warning.approval.timeout.days=3`（V10）；同步失败类待 ⑥ | 澄清 #10 |
+| 6 | 调整单是否走同一审批链；能否跨月调整已锁月份 | 同一审批链；跨月需管理员授权并留痕 | 已落实：调整单走同一审批链；跨月调整需管理员授权并留痕 | 澄清 #9 |
+| 7 | 待办事项范围 | 待审批 + 同步失败 + 回款未指定批次起步 | 已落实：工作台待办三项；⑥ 未建时同步项显式降级为「模块未启用」；待办按角色过滤 | 澄清 #11 |
+
+#### 关联项表（与《待确认事项汇总_2026-10-16》及催办清单的对应）
+
+| 3.3# | 澄清清单编号 | 催办清单位置 | 实现落点 |
+| --- | --- | --- | --- |
+| 1 | #1/#2/#3（#15 为作废条款关系） | 一、A-1 至 A-3 | 二次分摊拆行（ccfd206） |
+| 2 | #5 | 已按答复执行（供核阅） | V8 迁移参数种子 |
+| 3 | #8 | 已按答复执行（供核阅） | 月结取数（settlement_service） |
+| 4 | #6 | 一、B-6 | permissions/role_permissions 待清单；当前按 role_code 粗粒度 |
+| 5 | #10 | 已按答复执行（供核阅） | V10 参数 + dashboard_service |
+| 6 | #9 | 已按答复执行（供核阅） | 调整单审批链与跨月授权 |
+| 7 | #11 | 已按答复执行（供核阅） | dashboard_service 待办三源 |
 
 ### 3.4 对路线图的影响（本次已修订）
 
-1.1 依赖图、1.3 范围与验收、1.4 里程碑、1.5 依据对照均已并入本批内容；第三章待 3.3 各项澄清后，按第二章同粒度补齐 M3 各增量块的任务级实施计划。
+1.1 依赖图、1.3 范围与验收、1.4 里程碑、1.5 依据对照均已并入本批内容。M3 各增量块已于 2026-10-19 交付；3.3-1 仍无业务答复，二次分摊当前按开发侧暂定口径运行，答复后补口径变更计划；3.3-4 的权限细目与 3.3-5 的预警规则仍待业务确认。⑥ T+ 同步在 IT 答复前不排任务级计划。
 
 ## 验证方式
 
 1. 计划内容与 `ai-aux/AI_CONTEXT.md` 需求裁决节逐条核对（口径无冲突、无遗漏作废条款）；
 2. 计划中的命令与项目现有命令体系一致（uv / pnpm / docker compose）；
 3. 首块每个任务有独立可测交付物；测试要点符合「模块整体完成时统一编写」约定（纯函数与 core 级随实现即时测）；
-4. 执行阶段验证：T1 后 `test_schema_sync.py` 通过；T2/T3 后纯单元全绿；T5 后冒烟链路完整；T8 后全量 301+ 用例与 e2e 全绿。
+4. 执行阶段验证：T1 后 `test_schema_sync.py` 通过；T2/T3 后纯单元全绿；T5 后冒烟链路完整；T8 后全量后端用例与 e2e 全绿（以 `backend/tests/README.md` 最新运行记录为准）。

docs/待确认事项汇总_2026-10-16.md

@@ -4,6 +4,7 @@
 > 来源：《待澄清问题清单_2026-10-15.xlsx》（业务方 24 项）、《T+对接数据需求清单》v1.2.1（IT 18 项）、AI_CONTEXT 待办。
 > 答复到齐前，各模块按清单内「建议默认」执行；答复与默认不同时以答复为准。
 > **2026-10-19 更新**：业务答复落实批次与 M3 code-review 修复已落地（见各节标注）；T+ 清单已升 v1.4。
+> **2026-10-20 更新**：开发计划 3.3 已回填（3.3.1），SRS V2.1 已交付；二次分摊已按开发侧暂定口径实现（#1–#3 答复仍空白）。
 
 ## 一、立即待办（需项目内处理）
 
@@ -14,16 +15,16 @@
 
 ## 二、业务方澄清（24 项，**2026-10-16 答复已到**，回执见《待澄清问题清单_2026-10-15(1)(2).xlsx》H 列）
 
-> 答复汇总：**12 项确认按默认执行**（#7/#8/#9/#10/#11/#13/#15/#17/#18/#21/#23/#24）；**4 项答复与默认不同，以答复为准**——#5 二次分配审批开关**默认开启**、#14 **首日整 10‰**、#16 试用期基本工资**按人员档案逐人维护**、#19 备份**加增量补充**；仍未落定见下表「答复状态」列。答复后待回填开发计划 3.3 与需求记录。
+> 答复汇总：**13 项确认按默认执行**（#7/#8/#9/#10/#11/#13/#15/#17/#18/#21/#22/#23/#24，其中 #22 为模糊确认、按不计入）；**4 项答复与默认不同，以答复为准**——#5 二次分配审批开关**默认开启**、#14 **首日整 10‰**、#16 试用期基本工资**按人员档案逐人维护**、#19 备份**加增量补充**；部分答复 3 项（#4/#6/#12），仍开放 4 项（#1–#3/#20）。开发计划 3.3 已于 2026-10-20 回填（3.3.1），未答复项仍按催办清单跟进。
 
 ### 高优先级（阻塞对应模块开工，8 项）
 
 | 清单编号 | 问题 | 建议默认（未答复时执行） | 阻塞模块 | 答复状态 |
 | --- | --- | --- | --- | --- |
-| 1 | 二次分配的分配标的：绩效金额、折算毛利润还是业绩额？ | 未答复前只做方案录入，不实现分摊（已按此交付） | ④ 分摊计算 | ❌ 未答复 |
+| 1 | 二次分配的分配标的：绩效金额、折算毛利润还是业绩额？ | 方案录入已交付；2026-10-19 已按开发侧暂定口径实现分摊拆行（非业务答复，答复后校准） | ④ 分摊计算 | ❌ 未答复 |
 | 2 | 分配比例合计不足 100% 时余额归谁？ | 余额归合同负责营销人员（仅展示不落库） | ④ 分摊计算 | ❌ 未答复 |
 | 3 | 二次分配与已作废的「协助 25% 分成」是什么关系？ | 独立新机制；协助仍仅记录不核算 | ④ | ❌ 未答复 |
-| 4 | 二次分配方案的修改窗口与权限？ | 首次月结前可改（留痕），月结后只读走调整单 | ③/④ | ⚠️ 部分：「暂定可改走调整单」，窗口/权限未定；分摊实现（88 批次）进行中 |
+| 4 | 二次分配方案的修改窗口与权限？ | 首次月结前可改（留痕），月结后只读走调整单 | ③/④ | ⚠️ 部分：「暂定可改走调整单」，窗口/权限未定；分摊已于 2026-10-19 实现（ccfd206） |
 | 6 | 营销后勤/财务/领导的角色权限清单（可见菜单/可操作点）？ | 最小授权：后勤=提交录入，财务=复核+导出，领导=审核只读 | ⑦ / permissions 激活 | ⚠️ 部分：「先预设角色」，名称与职权待客户确认；**#17 参数查看放开已落实（2026-10-19）** |
 | 8 | 回款何时算「确认回款」？ | 需人工确认动作（系统已按此预留） | ④ 月结取数 | ✅ 确认：人工确认，可手动关闭 |
 | 9 | 调整单是否走同一审批链？能否跨月调整已锁月份？ | 同一审批链；后续月份扣补为默认路径 | ④/⑦ | ✅ 确认默认；跨月需管理员授权并留痕 |
@@ -41,7 +42,7 @@
 | 15 | 支付权重「余额模式」是否需要？ | ✅ 一期只用 FULL |
 | 16 | 试用期基本工资（3000 元）是统一参数还是逐人维护？ | ✅ 已落实（2026-10-19，V8）：撤销 `probation.salary.basic` 参数，基本工资读 `personnel.salary_standard` |
 | 20 | 含税口径：毛利/运费/价税合计是否含税？ | ❌ 未答复：「需与客户再次确认具体计算方式」 |
-| 22 | 预收款/保证金/代收款是否计入绩效回款？ | ❌ 未答复：「当前未提到预收款情况」，暂按不计入 |
+| 22 | 预收款/保证金/代收款是否计入绩效回款？ | ⚠️ 模糊答复：「当前未提到预收款情况」，暂按不计入 |
 
 ### 低优先级（已有默认口径，顺手确认，7 项）
 
@@ -83,7 +84,7 @@
 
 | 里程碑 | 内容 | 前置条件 |
 | --- | --- | --- |
-| M3 剩余 | ✅ ④ 核算引擎 + 月结/调整单 + ⑦ 月结审批对象已完成（2026-10-17 M3-core）；**2026-10-19** 业务答复落实批次（首日 10‰/审批开关默认/试用期参数/增量备份/参数查看放开/经理基数开关）与 code-review 21 项修复已落地；**余**：二次分摊实现（#1–#3 答复空白）、业务答复回填批次收口（88 会话） | #20（含税口径）、#1–#3 答复后按同粒度推进 |
+| M3 剩余 | ✅ ④ 核算引擎 + 月结/调整单 + ⑦ 月结审批对象已完成（2026-10-17 M3-core）；**2026-10-19** 业务答复落实批次（首日 10‰/审批开关默认/试用期参数/增量备份/参数查看放开/经理基数开关）与 code-review 21 项修复已落地；二次分摊已按开发侧暂定口径实现（ccfd206），业务答复回填已完成（计划 3.3.1，2026-10-20）；**余**：#1–#3/#20 答复与 #4/#6/#12 补值 | #20（含税口径）、#1–#3 答复后按同粒度推进 |
 | M4 | ⑤ 报表（签字导出模板）+ ⑧ 工作台 | ④ 已交付；报表口径确认后启动 |
 | M4（并行） | ⑥ T+ 数据同步 | IT 答复字段映射后 |
 | code-review 遗留 2 项 | 方案生效版本承载、审批单内容快照 | ④ 分摊语义落定时一并设计 |
@@ -92,6 +93,6 @@
 ## 五、文档待办（低优）
 
 - 需规过时条款未修订（26 天示例、含运费公式、协助 25%）；2026-10-15 补充需求未回填需规——对外引用需规前须先修订或注明以对照表/计划第三章为准
-- 澄清答复已到（2026-10-16）：待回填开发计划 3.3、AI_CONTEXT、《需求规格说明书》升 V2.1；《T+对接数据需求清单》已升 **v1.4**（2026-10-19，本节口径待收编；生成脚本 build_tplus_xlsx.py 已对齐 v1.3、待对齐 v1.4）
+- 澄清答复已到（2026-10-16）：开发计划 3.3 已回填（3.3.1，2026-10-20）、SRS V2.1 已交付（2026-10-19）；《T+对接数据需求清单》已升 **v1.4**（2026-10-19，本节口径待收编；生成脚本 build_tplus_xlsx.py 已对齐 v1.3、待对齐 v1.4）
 - ~~新口径待落实（业务答复与默认不同的 4 项）~~ ✅ **已全部落实（2026-10-19，见第二节标注）**
 - **修复批次引入口径记录（2026-10-19，供业务知会）**：月结的经理聚合与月工资口径按**所属月月末**取参（单笔台账行仍按回款日取参，裁决 #11）——补生成历史月份时结果不随生成日漂移；如业务对月末时点有异议，改 `settlement_service._month_end` 单点即可

docs/待答复催办清单_2026-10-19.md

@@ -1,7 +1,7 @@
 # 待答复事项催办清单（2026-10-19）
 
 > **用途**：对外跟催材料，分三块可直接拆分发送——一、业务方；二、IT/T+ 侧；三、报表口径补充确认（本轮新增）。
-> **基线**：《待澄清问题清单_2026-10-15(1)(2).xlsx》（业务 24 项，2026-10-16 已回 15 项）、《T+对接数据需求清单》v1.3「待IT确认事项」18 项 + v1.4（2026-10-19 勘误）、《问题与解决方案对照表_v2》（28 项已裁决口径）。
+> **基线**：《待澄清问题清单_2026-10-15(1)(2).xlsx》（业务 24 项，2026-10-16 已答复 19 项，其中部分答复 3 项；#1–#3/#20 仍空白）、《T+对接数据需求清单》v1.3「待IT确认事项」18 项 + v1.4（2026-10-19 勘误）、《问题与解决方案对照表_v2》（28 项已裁决口径）。
 > **原则**：已答复项不重复询问；未答复项按清单内「建议默认」执行中，答复后以答复为准并回填文档。
 
 ## 一、业务方（建议回复期限：____）
@@ -10,12 +10,12 @@
 
 | # | 问题（原文口径） | 当前执行默认 | 阻塞 |
 | --- | --- | --- | --- |
-| 1 | 二次分配的分配标的是什么——绩效工资金额、折算毛利润额，还是销售额（业绩额）？ | 未答复前只做方案录入，**不实现分摊计算**（已按此交付） | ④ 分摊计算 |
+| 1 | 二次分配的分配标的是什么——绩效工资金额、折算毛利润额，还是销售额（业绩额）？ | 方案录入已交付；**分摊已按开发侧暂定口径实现**（非业务答复，答复后校准） | ④ 分摊计算 |
 | 2 | 分配比例合计不足 100% 时，剩余部分归属谁？ | 余额归合同负责营销人员（仅展示不落库） | ④ 分摊计算 |
 | 3 | 二次分配与已作废的「协助 25% 分成」是什么关系？ | 独立新机制；协助仍仅记录不核算 | ④ 分摊 |
 | 20 | 含税口径：销售/成本/毛利润/运费各项是否含税？价税合计与折扣折让如何处理？ | 按不含税口径实现中 | ④ 核算口径终验、⑥ 同步建列（是否建税率/税额列） |
 
-> 说明：#1–#3 未答复期间，二次分配方案可正常录入与审批，但**分摊金额不会出现在绩效中**；#20 影响历史算出的绩效是否需要按含税口径重算，越早答复返工面越小。
+> 说明：#1–#3 未答复期间，二次分配方案可正常录入与审批；开发侧已按暂定口径把分摊拆行计入绩效（参与人份额×比例、负责人余额），**标的与余额归属仍待业务答复，答复后可能调整**；#20 影响历史算出的绩效是否需要按含税口径重算，越早答复返工面越小。
 
 ### B. 已有答复但需补具体值 / 勾选（4 项）
 

frontend/README.md

@@ -188,7 +188,7 @@ E2E_ADMIN_PASSWORD=<开发库admin密码> python e2e/system_management.py
 | 页面 | 说明 |
 | --- | --- |
 | `/reports/personal-detail` 个人绩效明细（新组「报表统计」order19） | 人员/月份/客户类型/产品类别/岗位筛选；需规 2.4.1 全部列；行展开懒加载台账快照；页脚**全筛选合计**（注明非当前页）；导出/打印 |
-| `/reports/department-summary` 部门绩效汇总（order20） | 月份必填；全公司分成独立卡片；团队表（基数/团队分成/全公司分成/占比/调整/净合计/月结冻结额），比例列头随在线参数变化；展开成员 → 跨页下钻个人明细（人员与月份带入筛选）；未分配桶（按经理筛选时为空） |
+| `/reports/department-summary` 部门绩效汇总（order20） | 月份必填；全公司分成独立卡片（单经理 5% × 计入人数 = 合计，每位有聚合行的经理各计一次）；团队表（基数/团队分成/全公司分成/占比/调整/净合计/月结冻结额），净合计与月结聚合行同口径，比例列头随在线参数变化；展开成员 → 跨页下钻个人明细（人员与月份带入筛选）；未分配桶（按经理筛选时为空） |
 | `/reports/business-stats` 业务统计（order21） | 关键字/客户/归属人/状态/业务日期；发货额/回款额/绩效（累计口径提示）；展开下钻回款与台账明细 |
 | `/reports/payment-stats` 回款统计（order22） | 月份/回款日期/客户类型/产品类别；8 档区间表（空桶 0，档位由后端下发）；展开桶下钻分页明细；客户类型/产品类别对比表 |
 | `/reports/print/:code` 报表打印（顶层路由，`menu:false`） | 不挂 BasicLayout（打印不带侧栏/顶栏）；页眉公司信息走 public 端点；循环翻页取全量（以接口 total 判定超过 2000 行并提示截断）；部门汇总按「团队分成 → 全公司 5% → 对账 → 成员明细 → 未分配」五块渲染；`@media print` A4 横版 + 每页表头；签字栏；PDF 由浏览器「另存为」覆盖 |
@@ -197,6 +197,7 @@ E2E_ADMIN_PASSWORD=<开发库admin密码> python e2e/system_management.py
 - api 封装 `src/api/reports.ts`（类型 + 查询 + `exportReport` 裸 axios blob；产品类别标签额外覆盖 `mixed`）。
 - 权限：营销后勤「授权范围内」待业务权限清单 #6，当前不开放报表。
 - type-check 与 build 通过（2026-10-19）。e2e `report_flow.py` 已补 12 步：登录 → API 造数 → 四表筛选/导出下载 → 部门下钻筛选生效 → 回款多桶同时展开互不串行 → 打印视图 → 打印截断提示 → 部门汇总打印五块结构；需 `E2E_ADMIN_PASSWORD` 且前后端在跑。
+- **e2e 实跑（2026-10-20）**：`login_flow` 8/8、`system_management` 13/13、`role_home_flow` 4/4、`report_flow` 12/12、`calc_settlement_flow` 8/8 全过。`report_flow` 的回款统计改为展开目标桶后按**本次业务编号**定位明细行，不依赖桶累计值，脚本可重复运行（历史 e2e 数据累计不再造成假失败）。
 
 ## 端到端测试（2026-10-19 追加 · M3 台账与月结）
 
@@ -225,4 +226,7 @@ E2E_ADMIN_PASSWORD=<开发库admin密码> python e2e/system_management.py
 - **ECharts 5.6 按需引入**：`echarts/core` + `BarChart/LineChart/Grid/Tooltip` + `SVGRenderer`（与设计预览一致），只注册用到的模块。
 - **字体红线**：图表 `textStyle.fontFamily` 必须显式设为思源黑体栈——ECharts 在 Windows 默认微软雅黑（字体商用合规要求）。
 - 权限：路由 `allowedRoles: [admin, finance, leader]`，与后端 `require_roles` 双层；营销后勤待权限清单 #6。
+- **普通用户落地页**：守卫不再硬编码回 `/dashboard`，统一走 `src/router/access.ts` 的 `firstAccessiblePath`（工作台 → 业务数据 → 回款），普通用户登录/刷新/直连受限页均落 `/business/businesses`，不会无限重定向；登录回跳也经 `canAccessPath` 校验。菜单与面包屑复用同一份准入判定。
+- 预警徽标用后端 `warnings.total`（全部命中数），`items` 仅为展示上限内的子集。
 - `package.json` 新增依赖 `echarts@5.6.0`（MIT）；type-check 与 build 通过（2026-10-19）。
+- e2e `e2e/role_home_flow.py`：普通用户登录落地、菜单隐藏、刷新保持、直连 `/dashboard` 与报表页回退；需 `E2E_ADMIN_PASSWORD`。

frontend/e2e/common.py

@@ -112,10 +112,12 @@ def expect_toast(page: Page, expected: str, step: str, timeout_ms: int = 5000) -
         fail(step, f"未出现提示「{expected}」，当前提示：{text or '（无）'}")
 
 
-def login(page: Page, username: str, password: str) -> None:
-    """完整登录：打开登录页 → 从 Redis 取验证码答案 → 提交 → 等待进入首页。
+def login(page: Page, username: str, password: str, expected_path: str = "/dashboard") -> None:
+    """完整登录：打开登录页 → 从 Redis 取验证码答案 → 提交 → 等待进入落地页。
 
     验证码 id 经响应监听捕获（前端把它内联在响应体里，不是 URL 参数）。
+    ``expected_path`` 按角色传入：管理员/财务/领导默认工作台，普通用户
+    落业务数据页（工作台权限收紧后守卫按角色解析首页）。
     """
     captcha_ids: list[str] = []
 
@@ -148,8 +150,8 @@ def login(page: Page, username: str, password: str) -> None:
         page.fill('input[autocomplete="off"]', answer)
         page.click("button:has-text('登 录')")
         page.wait_for_timeout(2000)
-        if current_path(page) != "/dashboard":
-            detail = f"未跳转到首页，当前 {page.url}"
+        if current_path(page) != expected_path:
+            detail = f"未跳转到 {expected_path}，当前 {page.url}"
             if message := toast_text(page):
                 detail += f"，提示：{message}"
             fail("登录", detail)

frontend/e2e/report_flow.py

@@ -23,6 +23,7 @@ import os
 import time
 import urllib.error
 import urllib.request
+from decimal import Decimal
 
 from playwright.sync_api import Page, sync_playwright
 
@@ -74,6 +75,15 @@ def step_setup_via_api(page: Page) -> dict:
     if not token:
         fail("API 预置", "localStorage 中未取到 ys_token")
 
+    # 回款统计断言取增量：脚本可重复运行（上次失败/历史 e2e 数据会累计到桶）
+    baseline = _api(
+        "GET", f"/api/v1/reports/payment-stats?settlement_month={MONTH}", token
+    )
+    baseline_bucket = next(b for b in baseline["buckets"] if b["code"] == "61-90")
+    expected_receipt_amount = (
+        Decimal(str(baseline_bucket["receipt_amount"])) + Decimal("10000.00")
+    )
+
     roles = _api("GET", "/api/v1/system/business-roles?pageNum=1&pageSize=50", token)["list"]
     salesperson_role = next((r["id"] for r in roles if r["role_code"] == "salesperson"), None)
     manager_role = next((r["id"] for r in roles if r["role_code"] == "sales_manager"), None)
@@ -159,8 +169,10 @@ def step_setup_via_api(page: Page) -> dict:
     return {
         "month": MONTH,
         "business_no": business["business_no"],
+        "personnel_id": person["id"],
         "personnel_name": person["personnel_name"],
         "manager_name": manager["personnel_name"],
+        "expected_receipt_amount": f"{expected_receipt_amount:.2f}",
     }
 
 
@@ -216,7 +228,7 @@ def step_business_stats(page: Page, ctx: dict) -> None:
     row = page.locator(".el-table__row", has_text=ctx["business_no"]).first
     if row.count() == 0:
         fail("业务统计", f"未找到业务 {ctx['business_no']}")
-    print("✓ 6. 业务统计      关键字筛选命中")
+    print("✓ 7. 业务统计      关键字筛选命中")
 
 
 def step_payment_stats(page: Page, ctx: dict) -> None:
@@ -225,16 +237,136 @@ def step_payment_stats(page: Page, ctx: dict) -> None:
     page.locator(".filters button:has-text('查询')").first.click()
     page.wait_for_timeout(900)
 
-    rows = page.locator(".el-table__row")
+    # 页面同时有区间表与两个维度对比表，必须限定主区间表
+    rows = page.locator('[data-testid="payment-bucket-table"] .el-table__row')
     if rows.count() != 8:
         fail("回款统计", f"区间表应为 8 行，实际 {rows.count()}")
-    bucket = page.locator(".el-table__row", has_text="61-90 天").first
+    bucket = page.locator(
+        '[data-testid="payment-bucket-table"] .el-table__row', has_text="61-90 天"
+    ).first
     if bucket.count() == 0:
         fail("回款统计", "未找到 61-90 天区间行")
     bucket_text = bucket.inner_text()
-    if "10000.00" not in bucket_text:
-        fail("回款统计", f"61-90 天桶回款金额不符：{bucket_text[:120]}")
-    print("✓ 7. 回款统计      8 档区间与目标桶金额正确")
+    expected = ctx["expected_receipt_amount"]
+    if expected not in bucket_text:
+        fail("回款统计", f"61-90 天桶回款金额不符（期望含 {expected}）：{bucket_text[:120]}")
+    print("✓ 8. 回款统计      8 档区间与目标桶金额正确")
+
+
+def step_payment_multi_expand(page: Page, ctx: dict) -> None:
+    """多个区间同时展开时，各自的下钻面板不能串行显示同一批明细。"""
+    page.goto(f"{BASE_URL}/reports/payment-stats", wait_until="networkidle")
+    page.wait_for_timeout(900)
+    page.locator(".filters button:has-text('查询')").first.click()
+    page.wait_for_timeout(900)
+
+    for label in ("61-90 天", "1-30 天"):
+        row = page.locator(
+            '[data-testid="payment-bucket-table"] .el-table__row', has_text=label
+        ).first
+        row.locator(".el-table__expand-icon").click()
+        page.wait_for_timeout(700)
+
+    drills = page.locator(".drill-wrap")
+    if drills.count() != 2:
+        fail("回款多桶展开", f"应展开 2 个下钻面板，实际 {drills.count()}")
+    texts = [drills.nth(index).inner_text() for index in range(2)]
+    # 面板是明细行（每笔 10000），不是桶合计；串行时两个面板文本会完全相同
+    if texts[0] == texts[1] or "10000.00" not in "".join(texts):
+        fail("回款多桶展开", "两个桶面板显示了同一批明细（drill 状态串行）")
+    print("✓ 9. 回款多桶展开  两个下钻面板互不串行")
+
+
+def step_department_drill(page: Page, ctx: dict) -> None:
+    """部门汇总「查看明细」必须把人员与月份带到个人明细页。"""
+    page.goto(f"{BASE_URL}/reports/department-summary", wait_until="networkidle")
+    page.wait_for_timeout(900)
+    page.locator(".filters button:has-text('查询')").first.click()
+    page.wait_for_timeout(900)
+
+    row = page.locator(".el-table__row", has_text=ctx["manager_name"]).first
+    row.locator(".el-table__expand-icon").click()
+    page.wait_for_timeout(500)
+    page.locator("button:has-text('查看明细')").first.click()
+    page.wait_for_timeout(900)
+
+    if f"personnel_id={ctx['personnel_id']}" not in page.url:
+        fail("部门下钻", f"URL 未携带人员筛选：{page.url}")
+    if ctx["personnel_name"] not in page.locator(".filters").inner_text():
+        fail("部门下钻", "个人明细页未应用人员筛选")
+    print("✓ 6. 部门下钻      人员与月份筛选已生效")
+
+
+def step_print_truncated(page: Page, ctx: dict) -> None:
+    """超过打印上限时必须给出截断提示（修复前提示永不出现）。"""
+
+    def handler(route) -> None:
+        rows = [
+            {
+                "ledger_id": index,
+                "business_no": f"E2E-{index}",
+                "ship_date": None,
+                "receipt_date": f"{ctx['month']}-01",
+                "payment_days": 10,
+                "product_category": "municipal_water",
+                "customer_type": "personal",
+                "role_name": "业务员",
+                "rate_text": "1",
+                "weight_r": 1,
+                "performance_amount": "1.00",
+                "payable_amount": "1.00",
+            }
+            for index in range(200)
+        ]
+        route.fulfill(
+            json={
+                "code": 200,
+                "message": "ok",
+                "data": {
+                    "list": rows,
+                    "total": 2500,
+                    "pageNum": 1,
+                    "pageSize": 200,
+                    "summary": {"row_count": 2500},
+                },
+            }
+        )
+
+    page.route("**/api/v1/reports/personal-detail*", handler)
+    try:
+        page.goto(
+            f"{BASE_URL}/reports/print/personal-detail?settlement_month={ctx['month']}",
+            wait_until="networkidle",
+        )
+        page.wait_for_timeout(1200)
+        if page.locator(".truncated").count() == 0:
+            fail("打印截断", "超过 2000 行未显示截断提示")
+        if "2000" not in page.locator(".truncated").inner_text():
+            fail("打印截断", "截断提示未说明上限行数")
+    finally:
+        page.unroute("**/api/v1/reports/personal-detail*", handler)
+    print("✓ 11. 打印截断      超限提示可见")
+
+
+def step_print_department(page: Page, ctx: dict) -> None:
+    """部门汇总打印稿必须含团队分成、5% 块、对账、成员与未分配五块。"""
+    page.goto(
+        f"{BASE_URL}/reports/print/department-summary?settlement_month={ctx['month']}",
+        wait_until="networkidle",
+    )
+    page.wait_for_timeout(1200)
+    for testid in (
+        "print-department-teams",
+        "print-company-block",
+        "print-reconciliation",
+        "print-department-members",
+        "print-unassigned",
+    ):
+        if page.locator(f'[data-testid="{testid}"]').count() == 0:
+            fail("部门打印", f"缺少区块 {testid}")
+    if "财务复核" not in page.locator(".print-page").inner_text():
+        fail("部门打印", "签字栏缺失")
+    print("✓ 12. 部门打印      五块结构与签字栏完整")
 
 
 def step_print_view(page: Page, ctx: dict) -> None:
@@ -253,7 +385,7 @@ def step_print_view(page: Page, ctx: dict) -> None:
     page.wait_for_timeout(400)
     if page.evaluate("window.__printed") is not True:
         fail("打印视图", "window.print 未被调用")
-    print("✓ 8. 打印视图      表格/签字栏/打印调用均正常")
+    print("✓ 10. 打印视图      表格/签字栏/打印调用均正常")
 
 
 def run(page: Page) -> None:
@@ -266,9 +398,13 @@ def run(page: Page) -> None:
     ctx = step_setup_via_api(page)
     step_personal_detail(page, ctx)
     step_department_summary(page, ctx)
+    step_department_drill(page, ctx)
     step_business_stats(page, ctx)
     step_payment_stats(page, ctx)
+    step_payment_multi_expand(page, ctx)
     step_print_view(page, ctx)
+    step_print_truncated(page, ctx)
+    step_print_department(page, ctx)
 
     if console_errors:
         print(f"  （控制台错误 {len(console_errors)} 条：{console_errors[:2]}）")

frontend/e2e/role_home_flow.py

@@ -0,0 +1,141 @@
+"""浏览器端到端验证：普通用户落地页与路由守卫回退。
+
+工作台收紧为 admin/finance/leader 后，登录默认页不能硬编码 `/dashboard`
+（普通用户会被无限重定向）。本脚本验证：
+
+- 普通用户登录后落业务数据页，不出现「无权访问」提示
+- 侧边菜单不出现首页/系统管理/报表统计
+- 刷新后仍保持登录态与落地页
+- 直连 `/dashboard` 或报表页时回退业务数据页并给出权限提示
+
+运行前置：容器、后端 8000、前端 5173 均在运行；设置 ``E2E_ADMIN_PASSWORD``。
+
+运行::
+
+    cd frontend && E2E_ADMIN_PASSWORD=<密码> python e2e/role_home_flow.py
+"""
+
+import json
+import os
+import time
+import urllib.error
+import urllib.request
+import uuid
+
+from playwright.sync_api import Page, sync_playwright
+
+from common import BASE_URL, E2EFailure, current_path, fail, login, toast_text
+
+ADMIN = "admin"
+USERNAME = f"e2e_role_{int(time.time())}_{uuid.uuid4().hex[:4]}"
+PASSWORD = "E2eRole@2026"
+HOME_PATH = "/business/businesses"
+
+
+def _api(method: str, path: str, token: str, body: dict | None = None) -> dict:
+    """带令牌的轻量 API 调用（urllib，避免额外依赖）。"""
+    request = urllib.request.Request(
+        f"{BASE_URL}{path}",
+        method=method,
+        headers={
+            "Authorization": f"Bearer {token}",
+            "Content-Type": "application/json",
+        },
+        data=json.dumps(body).encode() if body is not None else None,
+    )
+    try:
+        with urllib.request.urlopen(request, timeout=20) as response:
+            payload = json.loads(response.read().decode())
+    except urllib.error.HTTPError as exc:
+        detail = exc.read().decode()[:200]
+        raise E2EFailure(f"API {method} {path} HTTP {exc.code}：{detail}") from exc
+    if payload.get("code") != 200:
+        raise E2EFailure(f"API {method} {path} 业务失败：{payload.get('message')}")
+    return payload["data"]
+
+
+def _token(page: Page) -> str:
+    token = page.evaluate("localStorage.getItem('ys_token')")
+    if not token:
+        fail("API", "未取到登录令牌")
+    return token
+
+
+def _create_user(page: Page) -> dict:
+    return _api(
+        "POST",
+        "/api/v1/system/users",
+        _token(page),
+        {
+            "username": USERNAME,
+            "password": PASSWORD,
+            "real_name": "E2E普通用户",
+            "roles": ["user"],
+        },
+    )
+
+
+def run(page: Page) -> None:
+    admin_password = os.environ.get("E2E_ADMIN_PASSWORD", "")
+    if not admin_password:
+        fail("环境变量", "未设置 E2E_ADMIN_PASSWORD（开发库 admin 密码）")
+
+    login(page, ADMIN, admin_password)
+    user = _create_user(page)
+    try:
+        page.evaluate("localStorage.clear()")
+        login(page, USERNAME, PASSWORD, expected_path=HOME_PATH)
+        page.wait_for_timeout(500)
+        print("✓ 1. 普通用户登录 落地业务数据页")
+
+        menu_text = page.locator(".app-menu").inner_text()
+        for forbidden in ("首页", "系统管理", "报表统计"):
+            if forbidden in menu_text:
+                fail("菜单权限", f"普通用户看到了「{forbidden}」菜单")
+        print("✓ 2. 菜单权限     首页/系统管理/报表统计均不可见")
+
+        page.reload(wait_until="networkidle")
+        page.wait_for_timeout(600)
+        if current_path(page) != HOME_PATH:
+            fail("刷新恢复", f"刷新后未保持落地页，当前 {page.url}")
+        print("✓ 3. 刷新恢复     登录态与落地页保持")
+
+        for path in ("/dashboard", "/reports/personal-detail"):
+            page.goto(f"{BASE_URL}{path}", wait_until="networkidle")
+            page.wait_for_timeout(800)
+            if current_path(page) != HOME_PATH:
+                fail("路由守卫", f"直连 {path} 未回退业务数据页，当前 {page.url}")
+            if "无权访问该页面" not in toast_text(page):
+                fail("路由守卫", f"直连 {path} 未给出权限提示")
+        print("✓ 4. 路由守卫     受限页均回退业务数据页并提示")
+    finally:
+        # 无论用例是否失败都清理测试账号（重新以 admin 登录取令牌）
+        page.evaluate("localStorage.clear()")
+        login(page, ADMIN, admin_password)
+        _api("DELETE", f"/api/v1/system/users/{user['id']}", _token(page))
+
+
+def main() -> int:
+    with sync_playwright() as playwright:
+        browser = playwright.chromium.launch(headless=True)
+        page = browser.new_page(viewport={"width": 1440, "height": 900})
+        page.set_default_timeout(15000)
+        try:
+            run(page)
+        except E2EFailure as exc:
+            page.screenshot(path="e2e/failure.png", full_page=True)
+            print(f"\n  ✗ {exc}")
+            print("  已保存截图 e2e/failure.png")
+            return 1
+        except Exception as exc:  # noqa: BLE001 — 测试脚本需打印完整失败原因
+            page.screenshot(path="e2e/failure.png", full_page=True)
+            print(f"\n  ✗ 执行中断：{type(exc).__name__}: {exc}")
+            return 1
+        finally:
+            browser.close()
+    print("\n全部通过。")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

frontend/e2e/system_management.py

@@ -249,18 +249,18 @@ def step_permission_check(page: Page, admin_password: str) -> None:
     page.click("text=退出登录")
     page.wait_for_url("**/login**", timeout=10_000)
 
-    login(page, VIEWER_USERNAME, E2E_PASSWORD)
+    login(page, VIEWER_USERNAME, E2E_PASSWORD, expected_path="/business/businesses")
     if page.locator(".app-menu").get_by_text("系统管理", exact=True).count() != 0:
         fail("菜单权限", "普通用户看到了「系统管理」菜单")
     print("✓ 11. 菜单权限   普通用户看不到「系统管理」分组")
 
     page.goto(f"{BASE_URL}/system/users", wait_until="networkidle")
     page.wait_for_timeout(800)
-    if current_path(page) != "/dashboard":
+    if current_path(page) != "/business/businesses":
         fail("路由守卫", f"普通用户直连管理页未被弹回，当前 {page.url}")
     if "需要管理员权限" not in toast_text(page):
         fail("路由守卫", "直连管理页未给出权限提示")
-    print("✓ 12. 路由守卫   直连 /system/users 被弹回首页")
+    print("✓ 12. 路由守卫   直连 /system/users 被弹回业务数据页")
 
     page.click("text=退出登录")
     page.wait_for_url("**/login**", timeout=10_000)

frontend/src/api/businessRoles.ts

@@ -45,13 +45,26 @@ export interface BusinessRoleUpdateParams {
   sort_order?: number
 }
 
-/** 分页查询岗位角色（含停用） */
+/** 岗位下拉选项（登录即可；基础数据与报表筛选用，含停用） */
+export interface BusinessRoleOption {
+  id: number
+  role_code: string
+  role_name: string
+  status: number
+}
+
+/** 分页查询岗位角色（含停用，仅管理员） */
 export function listBusinessRoles(
   params: BusinessRoleQueryParams,
 ): Promise<PageData<BusinessRoleItem>> {
   return request({ url: '/v1/system/business-roles', method: 'get', params })
 }
 
+/** 只读岗位选项：财务/领导做报表筛选时不走管理员接口 */
+export function listBusinessRoleOptions(): Promise<BusinessRoleOption[]> {
+  return request({ url: '/v1/base/business-roles/options', method: 'get' })
+}
+
 /** 新增岗位 */
 export function createBusinessRole(
   data: BusinessRoleCreateParams,

frontend/src/api/dashboard.ts

@@ -67,7 +67,10 @@ export interface WarningBlock {
   enabled: boolean
   receipt_overdue_days: number
   approval_timeout_days: number
+  /** 当前返回的条目（受展示上限约束） */
   items: WarningItem[]
+  /** 全部规则命中总数；徽标计数用它，不能用 items.length */
+  total: number
   receipt_overdue_total: number
   approval_timeout_total: number
   sync_available: boolean

frontend/src/api/reports.ts

@@ -47,6 +47,9 @@ export interface DepartmentMemberRow {
   personnel_name: string
   role_code: string
   is_probation: number
+  /** regular 普通台账 / manager_personal 经理个人户 / adjustment 仅调整单 */
+  member_kind: string
+  unassigned_reason: string | null
   performance_sum: string | null
   payable_sum: string | null
   adjust_extra: string
@@ -63,7 +66,14 @@ export interface DepartmentTeamRow {
   member_count: number
   team_base: string
   team_share_amount: string
+  /** 单经理 5% 名义值；无月结聚合行时为 0（不发放） */
   company_share_amount: string
+  /** 本月是否有月结聚合行（决定是否发放 5%） */
+  has_aggregate: boolean
+  /** 月结聚合行应发（12.5%+5% 一次舍入）；无聚合行时 0 */
+  aggregate_amount: string
+  /** 团队应发小计 = 聚合行金额 + 调整净额 */
+  team_subtotal: string
   company_base_ratio: string
   adjust_extra: string
   adjust_deduct: string
@@ -74,6 +84,27 @@ export interface DepartmentTeamRow {
   members: DepartmentMemberRow[]
 }
 
+/** 签字稿对账块：应发 = 成员应付 + 未分配应付 + 聚合行金额 + 调整净额 */
+export interface DepartmentReconciliation {
+  team_member_payable_total: string
+  unassigned_payable_total: string
+  team_share_total: string
+  company_share_total: string
+  aggregate_payable_total: string
+  aggregate_rounding_diff: string
+  adjust_net_total: string
+  payable_total: string
+  team_base_total: string
+  unassigned_performance_total: string
+  other_team_base: string
+  other_team_member_payable_total: string
+  other_team_share_total: string
+  other_team_company_share_total: string
+  other_team_aggregate_total: string
+  other_adjust_net_total: string
+  company_base: string
+}
+
 export interface DepartmentSummary {
   settlement_month: string
   params: {
@@ -88,8 +119,14 @@ export interface DepartmentSummary {
     team_base_total: string
     company_base: string
     company_share_pct: string
+    /** 单经理 5% 名义值 */
     company_share_amount: string
+    /** 计入应发的经理数（有聚合行） */
+    company_share_count: number
+    /** 5% 合计 = 单份 × 计入次数 */
+    company_share_total: string
   } | null
+  reconciliation: DepartmentReconciliation | null
   totals: Record<string, string>
 }
 

frontend/src/components/layout/BasicLayout.vue

@@ -4,6 +4,7 @@ import { useRoute, useRouter } from 'vue-router'
 import { ElMessage } from 'element-plus'
 
 import AppMenu from '@/components/layout/AppMenu.vue'
+import { firstAccessiblePath } from '@/router/access'
 import { useUserStore } from '@/store/user'
 
 /** 基础布局：左侧品牌导航（窄屏收为抽屉）+ 顶栏（面包屑/用户/退出）+ 内容区。 */
@@ -13,11 +14,14 @@ const userStore = useUserStore()
 
 const drawerVisible = ref(false)
 
+/** 「首页」按角色解析：普通用户没有工作台权限，固定链到业务数据页 */
+const homePath = computed(() => firstAccessiblePath(router, userStore.roles))
+
 /** 面包屑：首页 + 分组（如有）+ 当前页标题（首页自身不重复） */
 const breadcrumbs = computed<string[]>(() => {
   const items: string[] = []
   if (route.meta.group) items.push(route.meta.group)
-  if (route.meta.title && route.path !== '/dashboard') items.push(route.meta.title)
+  if (route.meta.title && route.path !== homePath.value) items.push(route.meta.title)
   return items
 })
 
@@ -66,7 +70,7 @@ async function handleLogout(): Promise<void> {
             </svg>
           </button>
           <el-breadcrumb separator="/">
-            <el-breadcrumb-item :to="{ path: '/dashboard' }">首页</el-breadcrumb-item>
+            <el-breadcrumb-item :to="{ path: homePath }">首页</el-breadcrumb-item>
             <el-breadcrumb-item v-for="item in breadcrumbs" :key="item">
               {{ item }}
             </el-breadcrumb-item>

frontend/src/composables/useAppMenu.ts

@@ -1,8 +1,8 @@
 import { computed } from 'vue'
 import { useRouter } from 'vue-router'
 
+import { canAccessRoute, isMenuRecord } from '@/router/access'
 import { useUserStore } from '@/store/user'
-import { ROLE_ADMIN } from '@/utils/constants'
 
 /** 菜单节点：从路由表派生，避免菜单标题与页面标题两处维护漂移。 */
 export interface MenuNode {
@@ -30,19 +30,13 @@ export function useAppMenu() {
   const router = useRouter()
   const userStore = useUserStore()
 
-  const isAdmin = computed(() => userStore.roles.includes(ROLE_ADMIN))
   const userRoles = computed(() => userStore.roles)
 
   const menuNodes = computed<MenuNode[]>(() =>
     router
       .getRoutes()
-      .filter((record) => Boolean(record.meta.title) && record.meta.menu !== false)
-      .filter((record) => !record.meta.requiresAdmin || isAdmin.value)
-      // 职能页面（审批等）：任一列入角色命中即可见（admin 须显式列入集合）
-      .filter(
-        (record) =>
-          !record.meta.allowedRoles || record.meta.allowedRoles.some((role) => userRoles.value.includes(role)),
-      )
+      // 准入规则与路由守卫、落地页解析共用，避免三处漂移
+      .filter((record) => isMenuRecord(record) && canAccessRoute(record, userRoles.value))
       .map((record) => ({
         path: record.path,
         title: record.meta.title ?? '',

frontend/src/composables/usePagedList.ts

@@ -13,9 +13,11 @@ import { handledError } from '@/utils/error'
  *   请求的响应允许写入状态，否则旧数据会覆盖新数据、与分页器错位
  * - 请求失败不外抛（提示由拦截器统一完成），loading 仅由最新请求复位
  */
-export function usePagedList<T, F extends object>(
-  fetcher: (params: F & { pageNum: number; pageSize: number }) => Promise<PageData<T>>,
+export function usePagedList<T, F extends object, D extends PageData<T> = PageData<T>>(
+  fetcher: (params: F & { pageNum: number; pageSize: number }) => Promise<D>,
   getFilters: () => F,
+  /** 仅最新请求的响应会触发（fetcher 内的副作用不受序号保护） */
+  onData?: (data: D) => void,
 ) {
   const records = ref([]) as Ref<T[]>
   const total = ref(0)
@@ -39,6 +41,7 @@ export function usePagedList<T, F extends object>(
       if (seq !== requestSeq) return // 已有更新的请求发出，丢弃过期响应
       records.value = data.list
       total.value = data.total
+      onData?.(data)
     } catch (error) {
       handledError(error) // 失败提示由请求拦截器统一完成
     } finally {

frontend/src/router/access.ts

@@ -0,0 +1,62 @@
+import type { RouteRecordNormalized, Router } from 'vue-router'
+
+import { ROLE_ADMIN } from '@/utils/constants'
+
+/**
+ * 路由准入与落地页解析（菜单、守卫、登录回跳共用一份规则）。
+ *
+ * 之前守卫两处兜底硬编码 `/dashboard`，工作台收紧权限后普通用户会被
+ * 反复重定向到同一受限页（无限循环）。集中在这里后，任何角色的落地页
+ * 都必须先通过同一准入判定。
+ */
+
+/** 可作为菜单项/落地页：有标题、未被 menu:false 排除、非免登录页 */
+export function isMenuRecord(record: RouteRecordNormalized): boolean {
+  return Boolean(record.meta.title) && record.meta.menu !== false && !record.meta.public
+}
+
+/** 当前角色是否可访问该路由（与守卫、菜单过滤同源） */
+export function canAccessRoute(record: RouteRecordNormalized, roles: string[]): boolean {
+  if (record.meta.public) return false
+  if (record.meta.requiresAdmin && !roles.includes(ROLE_ADMIN)) return false
+  if (
+    record.meta.allowedRoles &&
+    !record.meta.allowedRoles.some((role) => roles.includes(role))
+  ) {
+    return false
+  }
+  return true
+}
+
+/**
+ * 落地页候选（按优先级）：工作台 → 业务数据 → 回款。
+ * 普通用户没有工作台权限时落到业务数据页，而不是被菜单 order 带到
+ * 基础数据管理页；候选都不可访问时再按菜单顺序兜底。
+ */
+const HOME_CANDIDATES = ['/dashboard', '/business/businesses', '/business/receipts']
+
+function matchedRecord(router: Router, path: string): RouteRecordNormalized | undefined {
+  const matched = router.resolve(path).matched
+  return matched[matched.length - 1]
+}
+
+/** 当前角色可访问的默认首页；无任何可访问页时回登录页（防循环） */
+export function firstAccessiblePath(router: Router, roles: string[]): string {
+  for (const path of HOME_CANDIDATES) {
+    const record = matchedRecord(router, path)
+    if (record && isMenuRecord(record) && canAccessRoute(record, roles)) return path
+  }
+  const fallback = router
+    .getRoutes()
+    .filter(isMenuRecord)
+    .filter((record) => canAccessRoute(record, roles))
+    .sort((a, b) => (a.meta.order ?? 99) - (b.meta.order ?? 99))
+  return fallback[0]?.path ?? '/login'
+}
+
+/** 登录回跳目标是否可信且可访问（防开放重定向与回跳受限页） */
+export function canAccessPath(router: Router, path: string, roles: string[]): boolean {
+  if (!path.startsWith('/') || path.startsWith('//')) return false
+  const record = matchedRecord(router, path)
+  return Boolean(record && isMenuRecord(record) && canAccessRoute(record, roles))
+}

frontend/src/router/index.ts

@@ -1,6 +1,7 @@
 import { createRouter, createWebHistory, type RouteRecordRaw } from 'vue-router'
 import { ElMessage } from 'element-plus'
 
+import { firstAccessiblePath } from '@/router/access'
 import { useUserStore } from '@/store/user'
 import { ROLE_ADMIN } from '@/utils/constants'
 
@@ -265,14 +266,17 @@ router.beforeEach(async (to) => {
   }
 
   if (to.meta.requiresAdmin && !userStore.roles.includes(ROLE_ADMIN)) {
-    ElMessage.error('需要管理员权限')
-    return { path: '/dashboard' }
+    const home = firstAccessiblePath(router, userStore.roles)
+    // 根路径/catch-all 的重定向不提示；用户手输受限 URL 才提示
+    if (!to.redirectedFrom) ElMessage.error('需要管理员权限')
+    return home === to.path ? { path: '/login' } : { path: home }
   }
 
   // 职能页面（审批等）：任一角色命中即可进；菜单隐藏之外的第二道防线
   if (to.meta.allowedRoles && !to.meta.allowedRoles.some((role) => userStore.roles.includes(role))) {
-    ElMessage.error('无权访问该页面')
-    return { path: '/dashboard' }
+    const home = firstAccessiblePath(router, userStore.roles)
+    if (!to.redirectedFrom) ElMessage.error('无权访问该页面')
+    return home === to.path ? { path: '/login' } : { path: home }
   }
 
   return true

frontend/src/views/base/personnel/index.vue

@@ -69,9 +69,9 @@ const managerOptions = ref<SelectOption[]>([])
 
 async function loadRoleOptions(): Promise<void> {
   try {
-    // 岗位规模有限（内置 3 个 + 自定义），一页足够
-    const data = await businessRoleApi.listBusinessRoles({ pageNum: 1, pageSize: 100 })
-    roleOptions.value = data.list.map((r) => ({ id: r.id, name: r.role_name, status: r.status }))
+    // 岗位规模有限（内置 3 个 + 自定义），选项接口一次取全
+    const data = await businessRoleApi.listBusinessRoleOptions()
+    roleOptions.value = data.map((r) => ({ id: r.id, name: r.role_name, status: r.status }))
   } catch (error) {
     handledError(error) // 下拉为空不阻断列表使用
   }

frontend/src/views/dashboard/index.vue

@@ -42,6 +42,8 @@ const ORDER_TYPE_LABELS: Record<number, string> = {
 const month = ref(currentMonth())
 const data = ref<DashboardData | null>(null)
 const loading = ref(false)
+/** 快速切换月份时，只允许最新请求写入数据与 loading */
+let loadSeq = 0
 
 const perfChartRef = ref<HTMLDivElement | null>(null)
 const receiptChartRef = ref<HTMLDivElement | null>(null)
@@ -54,15 +56,18 @@ function currentMonth(): string {
 }
 
 async function load(): Promise<void> {
+  const seq = ++loadSeq
   loading.value = true
   try {
-    data.value = await dashboardApi.getDashboard({ month: month.value })
+    const result = await dashboardApi.getDashboard({ month: month.value })
+    if (seq !== loadSeq) return // 已有更新的请求，丢弃过期响应
+    data.value = result
     await nextTick()
     renderCharts()
   } catch (error) {
-    handledError(error)
+    if (seq === loadSeq) handledError(error)
   } finally {
-    loading.value = false
+    if (seq === loadSeq) loading.value = false
   }
 }
 
@@ -294,7 +299,7 @@ function formatTime(value: string): string {
       <div class="bento-card b-2x2">
         <div class="b-eyebrow">
           风险预警
-          <span class="b-count">{{ data?.warnings.items.length ?? 0 }}</span>
+          <span class="b-count">{{ data?.warnings.total ?? 0 }}</span>
         </div>
         <div v-if="data && !data.warnings.enabled" class="b-empty-mini">
           预警已关闭（参数 warning.enabled=0）
@@ -322,6 +327,12 @@ function formatTime(value: string): string {
           暂无预警（回款逾期 {{ data?.warnings.receipt_overdue_days ?? 90 }} 天 /
           审批超时 {{ data?.warnings.approval_timeout_days ?? 3 }} 天）
         </div>
+        <div
+          v-if="data && data.warnings.total > data.warnings.items.length"
+          class="b-empty-mini"
+        >
+          仅显示前 {{ data.warnings.items.length }} 条，共 {{ data.warnings.total }} 条
+        </div>
         <div class="b-eyebrow b-eyebrow-gap">同步异常预警</div>
         <div class="b-empty-mini">
           {{ data?.warnings.sync_available ? '已接入' : '同步模块未启用（待 T+ 对接）' }}

frontend/src/views/login/index.vue

@@ -4,6 +4,7 @@ import { onMounted, reactive, ref } from 'vue'
 import { useRoute, useRouter } from 'vue-router'
 
 import { getCaptcha } from '@/api/auth'
+import { canAccessPath, firstAccessiblePath } from '@/router/access'
 import { useUserStore } from '@/store/user'
 
 const router = useRouter()
@@ -87,8 +88,15 @@ async function handleSubmit(): Promise<void> {
       captcha_id: captchaId.value,
       captcha_code: form.captcha_code.trim(),
     })
+    // 先拉角色再解析落地页：普通用户不能回跳到受限的工作台
+    await userStore.fetchUserInfo()
     ElMessage.success('登录成功')
-    await router.replace((route.query.redirect as string) || '/dashboard')
+    const redirect = route.query.redirect
+    const target =
+      typeof redirect === 'string' && canAccessPath(router, redirect, userStore.roles)
+        ? redirect
+        : firstAccessiblePath(router, userStore.roles)
+    await router.replace(target)
   } catch {
     // **必须刷新验证码**：后端用 GETDEL 消费验证码，无论何种失败它都已失效。
     // 不刷新的话，用户第二次提交必然报「验证码错误」，且极难自查原因。

frontend/src/views/reports/business-stats/index.vue

@@ -60,14 +60,15 @@ function getFilters(): ReportFilterParams {
   }
 }
 
-async function fetchPage(params: ReportFilterParams) {
-  const data = await reportApi.listBusinessStats(params)
-  summary.value = data.summary
-  return data
+function fetchPage(params: ReportFilterParams) {
+  return reportApi.listBusinessStats(params)
 }
 
+// summary 只能由最新请求写入：fetcher 内的副作用不受 usePagedList 序号保护
 const { records, total, pageNum, pageSize, loading, search, onPageChange, onSizeChange } =
-  usePagedList<BusinessStatsRow, ReportFilterParams>(fetchPage, getFilters)
+  usePagedList(fetchPage, getFilters, (data) => {
+    summary.value = data.summary
+  })
 
 // 进页即加载（报表页默认展示全部，筛选由用户主动收窄）
 onMounted(search)

frontend/src/views/reports/department-summary/index.vue

@@ -1,5 +1,5 @@
 <script setup lang="ts">
-import { reactive, ref } from 'vue'
+import { computed, reactive, ref } from 'vue'
 import { ElMessage } from 'element-plus'
 import { useRouter } from 'vue-router'
 
@@ -14,8 +14,8 @@ import { handledError } from '@/utils/error'
  *
  * 部门口径 = personnel.manager_id 实时上溯的经理团队（无独立部门表）；
  * 团队基数与月结聚合行同源（manager_share 共享模块），因此可与月结
- * 冻结额（frozen_amount）对账。全公司 5% 部分单列——各团队行展示的
- * 5% 金额是同一基数的名义值，不能相加。
+ * 冻结额（frozen_amount）对账。全公司分成每位有聚合行的经理各计一次，
+ * 团队应发小计与月结聚合行同口径；列头比例随在线参数变化。
  */
 
 const router = useRouter()
@@ -35,6 +35,18 @@ void loadManagers()
 const data = ref<DepartmentSummary | null>(null)
 const loading = ref(false)
 const exporting = ref(false)
+let loadSeq = 0
+
+// 比例来自在线参数，列头/卡片不能写死 12.5% / 5%
+const teamShareLabel = computed(
+  () => `团队分成（${data.value?.params?.team_text ?? '—'}%）`,
+)
+const companyShareLabel = computed(
+  () => `全公司分成（${data.value?.params?.company_text ?? '—'}%）`,
+)
+const companyTitle = computed(
+  () => `全公司 ${data.value?.params?.company_text ?? '—'}% 部分（单列）`,
+)
 
 function currentMonth(): string {
   const now = new Date()
@@ -46,16 +58,18 @@ async function load(): Promise<void> {
     ElMessage.warning('请选择统计月份（部门汇总按单月聚合）')
     return
   }
+  const seq = ++loadSeq
   loading.value = true
   try {
-    data.value = await reportApi.getDepartmentSummary({
+    const result = await reportApi.getDepartmentSummary({
       settlement_month: filters.settlement_month,
       manager_id: filters.manager_id,
     })
+    if (seq === loadSeq) data.value = result
   } catch (error) {
-    handledError(error)
+    if (seq === loadSeq) handledError(error)
   } finally {
-    loading.value = false
+    if (seq === loadSeq) loading.value = false
   }
 }
 
@@ -164,11 +178,16 @@ function pct(value: string): string {
     </div>
 
     <div v-loading="loading" class="company-card">
-      <span class="company-title">全公司 5% 部分（单列）</span>
+      <span class="company-title">{{ companyTitle }}</span>
       <span>全公司基数：{{ data?.company?.company_base ?? '—' }}</span>
-      <span>5% 分成：{{ data?.company?.company_share_amount ?? '—' }}</span>
+      <span>
+        单经理 {{ data?.params?.company_text ?? '—' }}% 分成：
+        {{ data?.company?.company_share_amount ?? '—' }} ×
+        {{ data?.company?.company_share_count ?? 0 }} 人 =
+        {{ data?.company?.company_share_total ?? '—' }}
+      </span>
       <span class="company-note">
-        各团队行的 5% 金额基于同一全公司基数，不可跨团队相加
+        每位有月结聚合行的经理各计一次；零成员团队无聚合行，不产生分成
       </span>
     </div>
 
@@ -210,8 +229,8 @@ function pct(value: string): string {
       </el-table-column>
       <el-table-column prop="member_count" label="团队人数" width="90" align="right" />
       <el-table-column prop="team_base" label="团队基数" width="120" align="right" />
-      <el-table-column prop="team_share_amount" label="12.5% 分成" width="120" align="right" />
-      <el-table-column prop="company_share_amount" label="5% 分成" width="110" align="right" />
+      <el-table-column prop="team_share_amount" :label="teamShareLabel" width="150" align="right" />
+      <el-table-column prop="company_share_amount" :label="companyShareLabel" width="150" align="right" />
       <el-table-column label="占全公司比" width="110" align="right">
         <template #default="{ row }">{{ pct(row.company_base_ratio) }}</template>
       </el-table-column>

frontend/src/views/reports/payment-stats/index.vue

@@ -40,66 +40,86 @@ function getFilters() {
   }
 }
 
+interface DrillState {
+  rows: PaymentDetailRow[]
+  total: number
+  pageNum: number
+  pageSize: number
+  loading: boolean
+  seq: number
+}
+
+/** 每个桶独立的下钻状态：同时展开多个区间时互不覆盖 */
+const drills = reactive<Record<string, DrillState>>({})
+
+function drillOf(code: string): DrillState {
+  drills[code] ??= { rows: [], total: 0, pageNum: 1, pageSize: 10, loading: false, seq: 0 }
+  return drills[code]
+}
+
+let loadSeq = 0
+
 async function load(): Promise<void> {
+  const seq = ++loadSeq
   loading.value = true
   try {
-    data.value = await reportApi.getPaymentStats(getFilters())
+    const result = await reportApi.getPaymentStats(getFilters())
+    if (seq !== loadSeq) return
+    data.value = result
+    // 筛选条件变化后旧桶明细不再对应当前口径；同时预建各桶状态，
+    // 避免模板渲染时才创建响应式属性
+    for (const code of Object.keys(drills)) delete drills[code]
+    for (const bucket of result.buckets) drillOf(bucket.code)
   } catch (error) {
-    handledError(error)
+    if (seq === loadSeq) handledError(error)
   } finally {
-    loading.value = false
+    if (seq === loadSeq) loading.value = false
   }
 }
 
-void load()
-
-/** 桶展开下钻：分页明细按 (桶编码) 缓存当前页 */
-const drill = reactive({
-  bucket: '',
-  rows: [] as PaymentDetailRow[],
-  total: 0,
-  pageNum: 1,
-  pageSize: 10,
-  loading: false,
-})
-
-async function loadDrill(bucket: BucketRow, expanded: boolean): Promise<void> {
-  if (!expanded) return
-  drill.bucket = bucket.code
-  drill.pageNum = 1
-  await fetchDrill()
+function loadDrill(bucket: BucketRow, expandedRows: BucketRow[]): void {
+  // Element Plus 2.x 的第二参是当前展开行数组，不是 boolean；
+  // 收起时该桶不在数组里，自然不再请求
+  if (!expandedRows.some((row) => row.code === bucket.code)) return
+  drillOf(bucket.code).pageNum = 1
+  void fetchDrill(bucket.code)
 }
 
-async function fetchDrill(): Promise<void> {
-  if (!drill.bucket) return
-  drill.loading = true
+async function fetchDrill(code: string): Promise<void> {
+  const state = drillOf(code)
+  const seq = ++state.seq
+  state.loading = true
   try {
     const page = await reportApi.listPaymentDetail({
       ...getFilters(),
-      bucket: drill.bucket,
-      pageNum: drill.pageNum,
-      pageSize: drill.pageSize,
+      bucket: code,
+      pageNum: state.pageNum,
+      pageSize: state.pageSize,
     })
-    drill.rows = page.list
-    drill.total = page.total
+    if (seq !== state.seq) return
+    state.rows = page.list
+    state.total = page.total
   } catch (error) {
-    handledError(error)
+    if (seq === state.seq) handledError(error)
   } finally {
-    drill.loading = false
+    if (seq === state.seq) state.loading = false
   }
 }
 
-function onDrillPage(page: number): void {
-  drill.pageNum = page
-  void fetchDrill()
+function onDrillPage(code: string, page: number): void {
+  drillOf(code).pageNum = page
+  void fetchDrill(code)
 }
 
-function onDrillSize(size: number): void {
-  drill.pageSize = size
-  drill.pageNum = 1 // 改每页条数回到第 1 页，否则可能停在空页
-  void fetchDrill()
+function onDrillSize(code: string, size: number): void {
+  const state = drillOf(code)
+  state.pageSize = size
+  state.pageNum = 1 // 改每页条数回到第 1 页，否则可能停在空页
+  void fetchDrill(code)
 }
 
+void load()
+
 async function doExport(): Promise<void> {
   exporting.value = true
   try {
@@ -198,15 +218,22 @@ function today(): string {
 
     <el-table
       v-loading="loading"
+      class="bucket-table"
+      data-testid="payment-bucket-table"
       :data="data?.buckets ?? []"
       row-key="code"
       stripe
       @expand-change="loadDrill"
     >
       <el-table-column type="expand">
-        <template #default>
+        <template #default="{ row }">
           <div class="drill-wrap">
-            <el-table v-loading="drill.loading" :data="drill.rows" size="small" stripe>
+            <el-table
+              v-loading="drillOf(row.code).loading"
+              :data="drillOf(row.code).rows"
+              size="small"
+              stripe
+            >
               <el-table-column prop="business_no" label="合同编号" min-width="150" />
               <el-table-column prop="receipt_batch_no" label="回款批次号" width="160" />
               <el-table-column prop="receipt_date" label="回款日" width="110" />
@@ -226,11 +253,11 @@ function today(): string {
               <el-table-column prop="performance_amount" label="绩效金额" width="120" align="right" />
             </el-table>
             <AppPagination
-              :total="drill.total"
-              :page-num="drill.pageNum"
-              :page-size="drill.pageSize"
-              @page-change="onDrillPage"
-              @size-change="onDrillSize"
+              :total="drillOf(row.code).total"
+              :page-num="drillOf(row.code).pageNum"
+              :page-size="drillOf(row.code).pageSize"
+              @page-change="onDrillPage(row.code, $event)"
+              @size-change="onDrillSize(row.code, $event)"
             />
           </div>
         </template>

frontend/src/views/reports/personal-detail/index.vue

@@ -1,7 +1,7 @@
 <script setup lang="ts">
-import { onMounted, reactive, ref } from 'vue'
+import { onMounted, reactive, ref, watch } from 'vue'
 import { ElMessage } from 'element-plus'
-import { useRouter } from 'vue-router'
+import { useRoute, useRouter } from 'vue-router'
 
 import * as calcApi from '@/api/calc'
 import type { LedgerRow } from '@/api/calc'
@@ -9,7 +9,7 @@ import { CUSTOMER_TYPE_OPTIONS, customerTypeLabel } from '@/api/customers'
 import { listPersonnel } from '@/api/personnel'
 import * as reportApi from '@/api/reports'
 import type { PersonalDetailRow, ReportFilterParams, ReportSummary } from '@/api/reports'
-import { listBusinessRoles } from '@/api/businessRoles'
+import { listBusinessRoleOptions } from '@/api/businessRoles'
 import AppPagination from '@/components/table/AppPagination.vue'
 import { useDropdownOptions } from '@/composables/useDropdownOptions'
 import { usePagedList } from '@/composables/usePagedList'
@@ -23,6 +23,7 @@ import { handledError } from '@/utils/error'
  */
 
 const router = useRouter()
+const route = useRoute()
 
 const filters = reactive({
   settlement_month: '',
@@ -32,14 +33,30 @@ const filters = reactive({
   role_code: '',
 })
 
+/**
+ * 部门汇总「查看明细」下钻：路由 query 必须真正落到筛选条件上。
+ *
+ * 无 query 时也要清空（不能只覆盖）：从菜单再次进入同一页面时组件复用，
+ * 旧的 personnel_id/settlement_month 会残留，用户看到的仍是被下钻者的数据。
+ */
+function applyRouteQuery(): void {
+  const personnelId = Number(route.query.personnel_id)
+  filters.personnel_id =
+    Number.isInteger(personnelId) && personnelId > 0 ? personnelId : undefined
+  const month = route.query.settlement_month
+  filters.settlement_month =
+    typeof month === 'string' && /^\d{4}-\d{2}$/.test(month) ? month : ''
+}
+applyRouteQuery()
+
 const { options: personnelOptions, load: loadPersonnel } = useDropdownOptions(() =>
   listPersonnel({ pageNum: 1, pageSize: 200, status: 1 }).then((data) =>
     data.list.map((item) => ({ label: item.personnel_name, value: item.id })),
   ),
 )
 const { options: roleOptions, load: loadRoles } = useDropdownOptions(() =>
-  listBusinessRoles({ pageNum: 1, pageSize: 200 }).then((data) =>
-    data.list.map((item) => ({ label: item.role_name, value: item.role_code })),
+  listBusinessRoleOptions().then((data) =>
+    data.map((item) => ({ label: item.role_name, value: item.role_code })),
   ),
 )
 void loadPersonnel()
@@ -58,18 +75,28 @@ function getFilters(): ReportFilterParams {
   }
 }
 
-async function fetchPage(params: ReportFilterParams) {
-  const data = await reportApi.listPersonalDetail(params)
-  summary.value = data.summary
-  return data
+function fetchPage(params: ReportFilterParams) {
+  return reportApi.listPersonalDetail(params)
 }
 
+// summary 只能由最新请求写入：fetcher 内的副作用不受 usePagedList 序号保护
 const { records, total, pageNum, pageSize, loading, search, onPageChange, onSizeChange } =
-  usePagedList<PersonalDetailRow, ReportFilterParams>(fetchPage, getFilters)
+  usePagedList(fetchPage, getFilters, (data) => {
+    summary.value = data.summary
+  })
 
-// 进页即加载（与月结/台账页修复后的行为一致；筛选条件默认为「全部」）
+// 进页即加载（筛选条件可能已由部门汇总下钻带入）
 onMounted(search)
 
+// 同一路由再次下钻（例如从部门汇总换成员）时重新应用 query
+watch(
+  () => route.query,
+  () => {
+    applyRouteQuery()
+    search()
+  },
+)
+
 function resetFilters(): void {
   filters.settlement_month = ''
   filters.personnel_id = undefined

frontend/src/views/reports/print/index.vue

@@ -7,7 +7,7 @@ import { customerTypeLabel } from '@/api/customers'
 import * as reportApi from '@/api/reports'
 import type {
   BusinessStatsRow,
-  DepartmentMemberRow,
+  DepartmentSummary,
   PersonalDetailRow,
   ReportFilterParams,
 } from '@/api/reports'
@@ -38,12 +38,30 @@ const title = ref('')
 // 表格行是四类报表的并集，模板按报表编码取用各自字段；用宽松记录类型
 // 避免为四种行形态各写一套模板 props
 const rows = ref<Record<string, any>[]>([])
+const department = ref<DepartmentSummary | null>(null)
 const buckets = ref<Record<string, any>[]>([])
 const groups = ref<{ title: string; rows: Record<string, any>[] }[]>([])
 const loading = ref(true)
 const truncated = ref(false)
 const printedAt = ref(new Date())
 
+/** 部门汇总成员明细展平（团队列冗余，与签字稿其他块一致） */
+const departmentMembers = computed(() =>
+  (department.value?.teams ?? []).flatMap((team) =>
+    team.members.map((member) => ({ manager_name: team.manager_name, ...member })),
+  ),
+)
+
+const MEMBER_KIND_LABELS: Record<string, string> = {
+  regular: '普通绩效',
+  manager_personal: '经理个人户',
+  adjustment: '仅调整单',
+}
+
+function memberKindLabel(kind: string): string {
+  return MEMBER_KIND_LABELS[kind] ?? kind
+}
+
 const filterText = computed(() => {
   const labels: Record<string, string> = {
     settlement_month: '统计月份',
@@ -71,13 +89,19 @@ async function fetchAll<T>(
 ): Promise<{ list: T[]; truncated: boolean }> {
   const collected: T[] = []
   let pageNum = 1
+  let total = 0
   while (collected.length < PRINT_MAX_ROWS) {
     const data = await fetcher({ ...params, pageNum, pageSize: PAGE_SIZE })
+    total = data.total
     collected.push(...data.list)
     if (collected.length >= data.total || data.list.length < PAGE_SIZE) break
     pageNum += 1
   }
-  return { list: collected.slice(0, PRINT_MAX_ROWS), truncated: collected.length > PRINT_MAX_ROWS }
+  // 以接口 total 判定：collected 最多正好等于上限，用长度比较永远为 false
+  return {
+    list: collected.slice(0, PRINT_MAX_ROWS),
+    truncated: total > PRINT_MAX_ROWS,
+  }
 }
 
 function personalRow(row: PersonalDetailRow): Record<string, any> {
@@ -109,18 +133,7 @@ async function load(): Promise<void> {
       truncated.value = data.truncated
     } else if (code.value === 'department-summary') {
       title.value = '部门绩效工资汇总表'
-      const data = await reportApi.getDepartmentSummary(queryParams.value)
-      rows.value = data.teams.flatMap((team) =>
-        team.members.map((member: DepartmentMemberRow) => ({
-          manager_name: team.manager_name,
-          member_name: member.personnel_name,
-          role_code: member.role_code,
-          performance_sum: member.performance_sum,
-          payable_sum: member.payable_sum,
-          adjust_net: member.adjust_net,
-          net_total: member.net_total,
-        })),
-      )
+      department.value = await reportApi.getDepartmentSummary(queryParams.value)
     } else if (code.value === 'business-stats') {
       title.value = '业务统计表'
       const data = await fetchAll(
@@ -169,12 +182,13 @@ onMounted(load)
       <h1 class="title">{{ title }}</h1>
       <div class="meta">
         <span>筛选条件：{{ filterText }}</span>
-        <span>打印时间：{{ printedAt.toLocaleString('zh-CN') }} ｜ 模板版本 V1.0</span>
+        <span>打印时间：{{ printedAt.toLocaleString('zh-CN') }} ｜ 模板版本 V1.2</span>
       </div>
     </header>
 
     <div v-if="truncated" class="truncated">
-      数据量超过打印上限（2000 行），仅打印前 2000 行；请缩小筛选范围后重新打印。
+      数据量超过打印上限（{{ PRINT_MAX_ROWS }} 行），仅打印前 {{ PRINT_MAX_ROWS }}
+      行；请缩小筛选范围后重新打印。
     </div>
 
     <template v-if="code === 'personal-detail'">
@@ -204,23 +218,171 @@ onMounted(load)
       </table>
     </template>
 
-    <template v-else-if="code === 'department-summary'">
-      <table class="sheet-table">
+    <template v-else-if="code === 'department-summary' && department">
+      <h2 class="block-title">一、团队分成汇总</h2>
+      <table class="sheet-table" data-testid="print-department-teams">
         <thead>
           <tr>
-            <th>经理团队</th><th>成员</th><th>岗位</th><th>绩效金额</th>
-            <th>应付绩效</th><th>调整净额</th><th>净合计</th>
+            <th>序号</th><th>经理团队</th><th>状态</th><th>团队人数</th><th>团队基数</th>
+            <th>团队分成（{{ department.params?.team_text ?? '—' }}%）</th>
+            <th>调整净额（仅核对）</th><th>团队应发小计（含聚合分成）</th><th>月结冻结额</th>
           </tr>
         </thead>
         <tbody>
-          <tr v-for="(row, index) in rows" :key="index">
-            <td>{{ row.manager_name }}</td>
-            <td>{{ row.member_name }}</td>
-            <td>{{ row.role_code }}</td>
-            <td class="num">{{ row.performance_sum }}</td>
-            <td class="num">{{ row.payable_sum }}</td>
-            <td class="num">{{ row.adjust_net }}</td>
-            <td class="num">{{ row.net_total }}</td>
+          <tr v-for="(team, index) in department.teams" :key="team.manager_id">
+            <td>{{ index + 1 }}</td>
+            <td>{{ team.manager_name }}</td>
+            <td>{{ team.manager_status === 1 ? '在职' : '离职' }}</td>
+            <td class="num">{{ team.member_count }}</td>
+            <td class="num">{{ team.team_base }}</td>
+            <td class="num">{{ team.team_share_amount }}</td>
+            <td class="num">{{ team.adjust_net }}</td>
+            <td class="num">{{ team.team_subtotal }}</td>
+            <td class="num">{{ team.frozen_amount ?? '—' }}</td>
+          </tr>
+        </tbody>
+      </table>
+
+      <h2 class="block-title">
+        二、全公司分成（每位有聚合行的经理各计一次，共
+        {{ department.company?.company_share_count ?? 0 }} 次）
+      </h2>
+      <table class="sheet-table" data-testid="print-company-block">
+        <thead>
+          <tr>
+            <th>全公司基数</th><th>分成比例</th><th>单经理分成</th>
+            <th>计入次数</th><th>分成合计</th>
+          </tr>
+        </thead>
+        <tbody>
+          <tr>
+            <td class="num">{{ department.company?.company_base ?? '—' }}</td>
+            <td class="num">{{ department.params?.company_text ?? '—' }}%</td>
+            <td class="num">{{ department.company?.company_share_amount ?? '—' }}</td>
+            <td class="num">{{ department.company?.company_share_count ?? 0 }}</td>
+            <td class="num">{{ department.company?.company_share_total ?? '—' }}</td>
+          </tr>
+        </tbody>
+      </table>
+
+      <h2 class="block-title">三、对账汇总</h2>
+      <table class="sheet-table" data-testid="print-reconciliation">
+        <thead>
+          <tr><th>项目</th><th>金额</th><th>说明</th></tr>
+        </thead>
+        <tbody>
+          <tr>
+            <td>团队成员应付合计</td>
+            <td class="num">{{ department.reconciliation?.team_member_payable_total ?? '—' }}</td>
+            <td />
+          </tr>
+          <tr>
+            <td>未分配人员应付合计</td>
+            <td class="num">{{ department.reconciliation?.unassigned_payable_total ?? '—' }}</td>
+            <td />
+          </tr>
+          <tr>
+            <td>团队分成合计（分列）</td>
+            <td class="num">{{ department.reconciliation?.team_share_total ?? '—' }}</td>
+            <td />
+          </tr>
+          <tr>
+            <td>
+              全公司分成合计（{{ department.company?.company_share_count ?? 0 }} 次，分列）
+            </td>
+            <td class="num">{{ department.reconciliation?.company_share_total ?? '—' }}</td>
+            <td>每位有聚合行的经理各计一次</td>
+          </tr>
+          <tr>
+            <td>聚合行舍入差</td>
+            <td class="num">{{ department.reconciliation?.aggregate_rounding_diff ?? '—' }}</td>
+            <td>分列舍入与聚合行一次舍入的差</td>
+          </tr>
+          <tr>
+            <td>调整净额合计（明细层）</td>
+            <td class="num">{{ department.reconciliation?.adjust_net_total ?? '—' }}</td>
+            <td />
+          </tr>
+          <tr>
+            <td>应发合计</td>
+            <td class="num">{{ department.reconciliation?.payable_total ?? '—' }}</td>
+            <td>上列各项之和</td>
+          </tr>
+          <tr>
+            <td>团队基数合计</td>
+            <td class="num">{{ department.reconciliation?.team_base_total ?? '—' }}</td>
+            <td />
+          </tr>
+          <tr>
+            <td>未分配绩效合计</td>
+            <td class="num">{{ department.reconciliation?.unassigned_performance_total ?? '—' }}</td>
+            <td />
+          </tr>
+          <tr>
+            <td>其他未展示基数</td>
+            <td class="num">{{ department.reconciliation?.other_team_base ?? '—' }}</td>
+            <td>筛选时含其他团队与全公司未分配</td>
+          </tr>
+          <tr>
+            <td>全公司基数</td>
+            <td class="num">{{ department.reconciliation?.company_base ?? '—' }}</td>
+            <td />
+          </tr>
+        </tbody>
+      </table>
+
+      <h2 class="block-title">四、团队成员明细</h2>
+      <table class="sheet-table" data-testid="print-department-members">
+        <thead>
+          <tr>
+            <th>经理团队</th><th>成员</th><th>岗位</th><th>来源</th><th>绩效金额</th>
+            <th>应付绩效</th><th>调整补发</th><th>调整扣减</th><th>调整净额</th><th>净合计</th>
+          </tr>
+        </thead>
+        <tbody>
+          <tr
+            v-for="member in departmentMembers"
+            :key="`${member.personnel_id}-${member.member_kind}`"
+          >
+            <td>{{ member.manager_name }}</td>
+            <td>{{ member.personnel_name }}</td>
+            <td>{{ member.role_code }}</td>
+            <td>{{ memberKindLabel(member.member_kind) }}</td>
+            <td class="num">{{ member.performance_sum }}</td>
+            <td class="num">{{ member.payable_sum }}</td>
+            <td class="num">{{ member.adjust_extra }}</td>
+            <td class="num">{{ member.adjust_deduct }}</td>
+            <td class="num">{{ member.adjust_net }}</td>
+            <td class="num">{{ member.net_total }}</td>
+          </tr>
+        </tbody>
+      </table>
+
+      <h2 class="block-title">五、未分配人员明细</h2>
+      <table class="sheet-table" data-testid="print-unassigned">
+        <thead>
+          <tr>
+            <th>人员</th><th>岗位</th><th>未纳入原因</th><th>绩效金额</th><th>应付绩效</th>
+            <th>调整补发</th><th>调整扣减</th><th>调整净额</th><th>净合计</th>
+          </tr>
+        </thead>
+        <tbody>
+          <tr
+            v-for="member in department.unassigned"
+            :key="`${member.personnel_id}-${member.member_kind}`"
+          >
+            <td>{{ member.personnel_name }}</td>
+            <td>{{ member.role_code }}</td>
+            <td>{{ member.unassigned_reason ?? '—' }}</td>
+            <td class="num">{{ member.performance_sum }}</td>
+            <td class="num">{{ member.payable_sum }}</td>
+            <td class="num">{{ member.adjust_extra }}</td>
+            <td class="num">{{ member.adjust_deduct }}</td>
+            <td class="num">{{ member.adjust_net }}</td>
+            <td class="num">{{ member.net_total }}</td>
+          </tr>
+          <tr v-if="department.unassigned.length === 0">
+            <td colspan="9">（无）</td>
           </tr>
         </tbody>
       </table>
@@ -374,6 +536,12 @@ onMounted(load)
   font-variant-numeric: tabular-nums;
 }
 
+.block-title {
+  margin: 18px 0 6px;
+  font-size: 14px;
+  font-weight: 600;
+}
+
 .group-block {
   margin-top: 18px;
 }