delete day3 default params

从奔曜同步本地资源树+publish到deck

plan/2026-05-20_add_two_node.md

@@ -0,0 +1,576 @@
+# Peptide Station 新增三个节点：等待订单完成 + 下料确认 + take-out 同步
+
+> 日期: 2026-05-20  
+> 目标文件: [peptide_station.py](../unilabos/devices/workstation/bioyond_studio/peptide_station/peptide_station.py)  
+> 参考实现:  
+> - [bioyond_cell_workstation.py](/Users/dp/python/yxz/unilabos/devices/workstation/bioyond_studio/bioyond_cell/bioyond_cell_workstation.py)（`wait_for_order_finish`、`get_material_info`）
+> - [bioyond_rpc.py L782-824](../unilabos/devices/workstation/bioyond_studio/bioyond_rpc.py)（`take_out`）
+> - [workstation_architecture.md](../docs/developer_guide/examples/workstation_architecture.md)（HTTP 报送进入 workstation，运行态记录保存在 workstation 内存）
+> 状态: 仅需求草稿，不写代码
+
+---
+
+## 一、需求背景
+
+`BioyondPeptideStation` 当前实验流程在 `start_experiment`（manual_confirm 启动调度器）之后即结束，缺少：
+
+1. **等待奔耀回报实验完成**：调度器跑完后，奔耀通过 LIMS 推送 `POST /report/order_finish` 回调；目前 peptide_station 没有把这条推送封装成 action 节点，下游无法在工作流图上等待结果，也拿不到 `usedMaterials` 等下游所需信息。
+2. **下料引导**：实验完成后操作员需要把样品/产物从仓位里取出，下料前需要看到每个物料对应的 **仓库 / 位置 / 物料名称 / 数量**；下料完成后还需要回写奔耀（调用 `take-out` 接口），让奔耀清空相应库位状态。
+
+本轮新增三个 action 节点，串在 `start_experiment` 之后：
+
+```text
+submit_experiment_dayN
+  -> start_experiment(manual_confirm 上料)
+  -> wait_for_order_finish (等回调 + 生成 unloadTable)
+  -> confirm_unload_materials (manual_confirm 下料确认)
+  -> take_out_materials (调用 take-out 同步奔耀)
+```
+
+---
+
+## 二、关键设计决策
+
+### D1. 本轮只支持单订单，`order_ids` 只做占位
+
+当前不实现多订单等待、乱序回调缓存、并发 wait 隔离。节点输入以 `order_id` 为主，`order_code` 可作为调试兜底。
+
+`order_ids` 可在 handle/返回值中保留为占位字段，但实现只处理第一笔或直接忽略多订单列表。多订单、乱序回调、跨节点重跑复用缓存放到后续迭代。
+
+### D2. `start_experiment` 需要显式透传订单字段
+
+当前 `start_experiment` 只有输入 handles，缺少输出 handles；如果下游 wait 节点要接在 `start_experiment` 之后，必须让 `start_experiment` 透传：
+
+- `order_id`
+- `order_ids`（占位）
+- `order_code`
+- `resultTable`
+
+实现时给 `start_experiment` 增加对应 `ActionOutputHandle`，并在返回值里保留这些字段。`submit_experiment_dayN` 的 `start_experiment` 嵌套字典也应包含 `order_code`，便于工作流编辑器连线。
+
+### D3. `unloadTable` 必须与 `resultTable` 字段一致
+
+`unloadTable` 不新增 `posX/posY/posZ/unit` 列，直接复用现有 `RESULT_TABLE_COLUMNS` 的四列：
+
+```python
+RESULT_TABLE_COLUMNS = [
+    {"name": "设备", "key": "whName"},
+    {"name": "位置", "key": "locationCode"},
+    {"name": "物料名称", "key": "materialName"},
+    {"name": "数量", "key": "quantity"},
+]
+```
+
+`submit_experiment_dayN` 现有上料确认表 `resultTable` 形状如下；`unloadTable` 也必须保持同一 shape，只改 `tableName` 和行数据来源：
+
+```json
+{
+  "data": [
+    {
+      "whName": "自动化堆栈",
+      "locationCode": "A1",
+      "materialName": "96孔板",
+      "quantity": "1"
+    }
+  ],
+  "columns": [
+    {"name": "设备", "key": "whName"},
+    {"name": "位置", "key": "locationCode"},
+    {"name": "物料名称", "key": "materialName"},
+    {"name": "数量", "key": "quantity"}
+  ],
+  "tableName": "resultTable"
+}
+```
+
+`material-info` 官方 schema 中位置坐标字段是 `locations[].x/y/z`，不是 `posX/posY/posZ`。本轮下料表不展示坐标。
+
+### D4. manual_confirm 只做人确认，take-out 放到普通 action
+
+`confirm_unload_materials` 只负责展示下料表、等待操作员确认并透传数据；真正的 `take-out` 调用放到后续普通 action `take_out_materials`。
+
+这样更接近 UniLab manual_confirm 的推荐模式：manual_confirm 是人机确认检查点，副作用由独立普通 action 执行。
+
+### D5. 本轮不做 unload context 缓存
+
+虽然 workstation architecture 文档支持在 workstation 内存保存 HTTP 报送记录，但本轮暂不实现 `unload_context_cache` 或 order report 缓存。
+
+因此本轮限制如下：
+
+- `wait_for_order_finish` 只等待本次进入节点之后到达的 `/report/order_finish`。
+- 如果 push 早于 wait 节点到达，本轮不自动补救。
+- 如果用户在 `confirm_unload_materials` approve 时忘记勾选，节点失败；当前架构不支持原地重新弹出同一个 manual_confirm，也不在本轮实现失败节点重跑。
+- 后续若要支持重跑复用，应在 `BioyondPeptideStation` 实例上新增 station runtime 的 `unload_context_cache`，按 `orderCode` 缓存 `unloadTable/material_ids/order_id` 等上下文。
+
+---
+
+## 三、节点 1：`wait_for_order_finish`（等推送 + 生成 unloadTable）
+
+### 行为
+
+1. 解析单订单目标：
+   - 首选 `order_id`。
+   - 如果没有 `order_code`，通过 `self.hardware_interface.order_report(order_id)` 取返回数据中的 `code` 作为 `orderCode`。
+   - `order_ids` 仅占位，本轮不实现多订单循环。
+2. 设置 `self.last_order_code = order_code`、`self.last_order_report = None`，并 `self.order_finish_event.clear()`。
+3. 阻塞在 `self.order_finish_event.wait(timeout=timeout_seconds)` 等 LIMS 推送。
+4. peptide_station override `process_order_finish_report(report_request, used_materials)`：
+   - 先调用 `super().process_order_finish_report(...)` 保留父类行为（状态发布、物料同步等）。
+   - 当 `report_request.data.orderCode == self.last_order_code` 时，把 `report_request.data` 存入 `self.last_order_report`，并 `set()` event。
+   - 非当前订单推送只记录日志，本轮不缓存。
+5. 解除阻塞后解析 `status`：
+   - `"30"` -> `success`
+   - `"-11"` -> `abnormal_stop`
+   - `"-12"` -> `manual_stop`
+   - 其它 -> `unknown_<status>`
+   - 超时 -> `timeout`
+6. 对 `report.usedMaterials[].materialId` 调用 `self.hardware_interface.material_info(material_id)`，带本地函数级 `material_info_cache` 避免重复请求。
+7. 组装 `unloadTable`、`material_ids`、`preintake_ids`、`unload_summary` 并作为输出 handles 暴露。
+
+### 入参（goal_default）
+
+| 参数 | 类型 | 说明 |
+|------|------|------|
+| `order_id` | `str` | 来自 `start_experiment` 透传输出，必填优先 |
+| `order_code` | `str` | 调试兜底；若已知 orderCode 可跳过 `order_report` 反查 |
+| `order_ids` | `List[str]` | 占位字段；本轮不实现多订单 |
+| `timeout_seconds` | `int` | 默认 `36000`（10h） |
+| `poll_mode` | `bool` | 默认 `False`；如需要可沿用 bioyond_cell 的轮询等待风格 |
+
+### 输出 handles
+
+| key | data_type | 说明 |
+|-----|-----------|------|
+| `order_finish_status` | `str` | `success` / `abnormal_stop` / `manual_stop` / `timeout` / `unknown_*` |
+| `order_finish_report` | `json` | 完整 `report_request.data` |
+| `used_materials` | `json` | JSON 化后的 `usedMaterials` 列表 |
+| `material_ids` | `json` | 从 `used_materials` 抽出的 `materialId` 列表，可为空 |
+| `preintake_ids` | `json` | 本轮默认 `[]`，保留扩展点 |
+| `unloadTable` | `table` | 下料表，字段与 `resultTable` 一致 |
+| `unload_summary` | `json` | `{ "order_code": ..., "total_items": N, "missing_material_info": [...] }` |
+| `order_id` | `str` | 透传给后续节点 |
+| `order_code` | `str` | 透传给后续节点 |
+| `order_ids` | `json` | 占位透传 |
+
+### `unloadTable` 组装规则
+
+返回结构：
+
+```json
+{
+  "data": [
+    {
+      "whName": "自动化堆栈",
+      "locationCode": "A1",
+      "materialName": "多肽产物",
+      "quantity": "10 mg"
+    }
+  ],
+  "columns": [
+    {"name": "设备", "key": "whName"},
+    {"name": "位置", "key": "locationCode"},
+    {"name": "物料名称", "key": "materialName"},
+    {"name": "数量", "key": "quantity"}
+  ],
+  "tableName": "unloadTable"
+}
+```
+
+每行字段：
+
+| key | 数据来源 |
+|-----|----------|
+| `whName` | `material_info.locations` 中匹配 `usedMaterials.locationId` 的 location 的 `whName`；匹配不到取第一条 location 的 `whName`；失败为空串 |
+| `locationCode` | 匹配 location 的 `code`；匹配不到取第一条 location 的 `code`；再兜底 `usedMaterials.locationId` |
+| `materialName` | `material_info.name`；失败为空串 |
+| `quantity` | `usedMaterials.usedQuantity`，若 `material_info.unit` 存在则拼成字符串（如 `"10 mg"`） |
+
+`material-info` 失败时不抛异常，对应行尽量保留 `locationCode` / `quantity`，`whName` 和 `materialName` 用空串，并把 `materialId` 放入 `unload_summary.missing_material_info`。
+
+### 接口依赖
+
+| 接口 | 调用方式 | 用途 |
+|------|----------|------|
+| `process_order_finish_report` 钩子 | 基类 HTTP 服务已注册 | 接 LIMS push |
+| `POST /api/lims/order/order-report` | `self.hardware_interface.order_report(order_id)` | 从 `order_id` 反查 `orderCode` |
+| `POST /api/lims/storage/material-info` | `self.hardware_interface.material_info(material_id)` | 查 `whName/locationCode/materialName/unit` |
+
+#### `order_report(order_id)` API 形式
+
+请求：
+
+```json
+{
+  "apiKey": "B10B5995",
+  "requestTime": "2026-05-20T10:50:00.123Z",
+  "data": "<orderId UUID>"
+}
+```
+
+响应中本节点只依赖 `data.code`：
+
+```json
+{
+  "code": 1,
+  "message": null,
+  "timestamp": 1779255000000,
+  "data": {
+    "id": "<orderId UUID>",
+    "name": "实验260520-103000",
+    "code": "EXP260520-103000",
+    "status": 30,
+    "statusName": "完成"
+  }
+}
+```
+
+`BioyondV1RPC.order_report(order_id)` 已经返回响应中的 `data`，所以实现中应读取 `raw.get("code")`，不是 `raw["data"]["code"]`。
+
+#### `material_info(material_id)` API 形式
+
+请求：
+
+```json
+{
+  "apiKey": "B10B5995",
+  "requestTime": "2026-05-20T10:50:00.123Z",
+  "data": "<materialId UUID>"
+}
+```
+
+响应中本节点依赖：
+
+```json
+{
+  "id": "<materialId UUID>",
+  "name": "多肽产物",
+  "unit": "mg",
+  "locations": [
+    {
+      "id": "<locationId UUID>",
+      "whid": "<warehouse UUID>",
+      "whName": "自动化堆栈",
+      "code": "A1",
+      "x": 1,
+      "y": 1,
+      "z": 1,
+      "quantity": 10
+    }
+  ]
+}
+```
+
+`BioyondV1RPC.material_info(material_id)` 已经返回响应中的 `data`。字段映射：
+
+| unloadTable key | 来源 |
+|-----------------|------|
+| `whName` | 匹配 `locationId` 的 `locations[].whName` |
+| `locationCode` | 匹配 `locationId` 的 `locations[].code` |
+| `materialName` | `name` |
+| `quantity` | `usedMaterials[].usedQuantity` + `unit` |
+
+### 实现要点
+
+- `BioyondPeptideStation.__init__` 末尾追加 `self.order_finish_event = threading.Event()`、`self.last_order_code = None`、`self.last_order_report = None`。
+- 新增 `process_order_finish_report` override，先 `super()`，再做单订单匹配。
+- `used_materials` 参数是 `MaterialUsage` dataclass 列表；输出前必须转成 JSON dict。
+- `unloadTable` 复用 `RESULT_TABLE_COLUMNS`，不新增 `UNLOAD_TABLE_COLUMNS`。
+
+---
+
+## 四、节点 2：`confirm_unload_materials`（人工下料确认）
+
+### 行为
+
+1. 接收节点 1 输出的 `order_id` / `order_code` / `material_ids` / `preintake_ids` / `unloadTable`。
+2. 进入 `NodeType.MANUAL_CONFIRM` 阻塞，操作员根据 `unloadTable` 物理下料。
+3. 操作员勾选 `materials_unloaded=True` 并 approve 后，节点函数体继续。
+4. 校验 `materials_unloaded == True`：
+   - 为 True：返回确认结果，并透传 `order_id/material_ids/preintake_ids/unloadTable` 给节点 3。
+   - 为 False：抛 `RuntimeError("下料未确认，拒绝继续 take-out")`。
+
+### 入参（goal_default）
+
+| 参数 | 类型 | 说明 |
+|------|------|------|
+| `order_id` | `str` | 来自节点 1，必填 |
+| `order_code` | `str` | 来自节点 1，日志/排错用 |
+| `material_ids` | `List[str]` | 来自节点 1，可为空 |
+| `preintake_ids` | `List[str]` | 来自节点 1，默认 `[]` |
+| `unloadTable` | `table` | 来自节点 1，供人工确认展示 |
+| `materials_unloaded` | `bool` | manual_confirm 勾选字段，默认 `False` |
+| `timeout_seconds` | `int` | 默认 `3600` |
+| `assignee_user_ids` | `List[str]` | `placeholder_keys={"assignee_user_ids": "unilabos_manual_confirm"}` |
+
+### 输出 handles
+
+| key | data_type | 说明 |
+|-----|-----------|------|
+| `unload_confirmed` | `bool` | 是否已人工确认下料 |
+| `order_id` | `str` | 透传 |
+| `order_code` | `str` | 透传 |
+| `material_ids` | `json` | 透传 |
+| `preintake_ids` | `json` | 透传 |
+| `unloadTable` | `table` | 透传 |
+
+### 实现要点
+
+- 装饰器使用 `node_type=NodeType.MANUAL_CONFIRM`。
+- `always_free=True`、`placeholder_keys`、`feedback_interval=300` 与现有 `start_experiment` 保持一致。
+- 本节点不调用 `take_out`，只做确认与透传。
+- 忘记勾选后不会自动重新显示下料指引；本轮不实现缓存或失败节点原地重跑。
+
+---
+
+## 五、节点 3：`take_out_materials`（调用 take-out 同步奔耀）
+
+### 行为
+
+1. 接收节点 2 透传的 `order_id` / `material_ids` / `preintake_ids`。
+2. 校验 `order_id` 非空。
+3. 调用 `self.hardware_interface.take_out(order_id, preintake_ids=preintake_ids, material_ids=material_ids)`。
+4. 返回 `take_out_result`、`unloaded_count`、`success`。
+
+### 入参
+
+| 参数 | 类型 | 说明 |
+|------|------|------|
+| `order_id` | `str` | 必填 |
+| `material_ids` | `List[str]` | 可为空；为空时由奔耀按 `orderId` 处理的能力以后现场确认 |
+| `preintake_ids` | `List[str]` | 可为空，默认 `[]` |
+| `order_code` | `str` | 日志/排错用 |
+
+### 输出 handles
+
+| key | data_type | 说明 |
+|-----|-----------|------|
+| `take_out_result` | `json` | `take-out` 原始响应 `{code, message, timestamp, data}` |
+| `unloaded_count` | `int` | `len(material_ids)` |
+| `success` | `bool` | `take_out_result.code == 1` 且 `data` 不为 False |
+
+### 接口依赖
+
+| 接口 | 调用方式 | 用途 |
+|------|----------|------|
+| `POST /api/lims/order/take-out` | `self.hardware_interface.take_out(order_id, preintake_ids, material_ids)` | 通知奔耀同步取出 |
+
+请求体 schema（helper script 已核对）：
+
+```json
+{
+  "apiKey": "B10B5995",
+  "requestTime": "2026-05-20T10:50:00.123Z",
+  "data": {
+    "orderId": "<UUID>",
+    "preintakeIds": [],
+    "materialIds": ["<UUID-1>", "<UUID-2>"]
+  }
+}
+```
+
+响应 schema：
+
+```json
+{
+  "code": 1,
+  "message": null,
+  "timestamp": 1779255000000,
+  "data": true
+}
+```
+
+`BioyondV1RPC.take_out(...)` 返回完整响应包，因此 `take_out_materials` 应保留原始包到 `take_out_result`。
+
+### 实现要点
+
+- 本轮不修改 `sample_waste_removal`，保持 backward compatibility。
+- 新节点只调用现有完整能力的 `take_out(...)`。
+- `preintake_ids` / `material_ids` 都按可选列表处理，默认 `[]`。
+- `take-out` 返回 `code != 1` 时返回 `success=False` 并记录 warning；是否抛异常留作开放问题。
+
+---
+
+## 六、端到端工作流连线
+
+```mermaid
+flowchart LR
+    submit["submit_experiment_dayN"] --> start["start_experiment<br/>manual_confirm: 上料"]
+    start -->|order_id, order_code| wait["wait_for_order_finish<br/>等 order_finish + 生成 unloadTable"]
+    wait -->|order_id, material_ids,<br/>preintake_ids, unloadTable| confirm["confirm_unload_materials<br/>manual_confirm: 操作员下料确认"]
+    confirm -->|order_id, material_ids,<br/>preintake_ids| takeout["take_out_materials<br/>调用 take-out 同步奔耀"]
+    bioyond["奔耀 LIMS"] -.HTTP POST /report/order_finish.-> wait
+    wait -.material-info.-> bioyond
+    takeout -.take-out.-> bioyond
+```
+
+---
+
+## 七、影响面与兼容性
+
+- **`peptide_station.py`**
+  - 修改 `start_experiment`：增加 `order_id/order_code/order_ids/resultTable` 输出 handles，并在返回值透传。
+  - 增加 `wait_for_order_finish`、`confirm_unload_materials`、`take_out_materials` 三个 action。
+  - 增加 `process_order_finish_report` override。
+  - 增加 `_build_unload_table(...)` 等私有辅助方法。
+- **`bioyond_rpc.py` 不动**
+  - `take_out` 已有完整 schema 能力。
+  - `sample_waste_removal` 本轮不改，保持兼容。
+- **基类 `station.py` 不动**
+  - override 中保留 `super().process_order_finish_report(...)` 调用。
+- **HTTP 服务不动**
+  - `WorkstationHTTPService` 已支持 `/report/order_finish`。
+- **本轮不做缓存**
+  - 不新增 `unload_context_cache`。
+  - 不支持 push 早于 wait 的自动补救。
+  - 不支持失败 manual_confirm 原地重跑。
+- **测试**：补在现有路径 `unilabos/devices/workstation/bioyond_studio/peptide_station/tests/test_peptide_station_contracts.py`
+  1. `start_experiment` 输出 handles/返回值透传 `order_id/order_code/order_ids/resultTable`。
+  2. `process_order_finish_report` orderCode 匹配 / 不匹配时 event 是否触发。
+  3. `wait_for_order_finish` 单订单成功、超时、状态映射、`used_materials` JSON 化。
+  4. `_build_unload_table` 列顺序严格等于 `RESULT_TABLE_COLUMNS`，且无 `posX/posY/posZ/unit` 列。
+  5. `material-info` 失败时不抛异常，`missing_material_info` 正确记录。
+  6. `confirm_unload_materials` 未勾选时报错，勾选后透传下游字段且不调用 `take_out`。
+  7. `take_out_materials` 调用 `hardware_interface.take_out(order_id, preintake_ids, material_ids)`，不调用 `sample_waste_removal`。
+
+---
+
+## 八、待人类确认的开放问题
+
+1. **过滤产物 vs 全量**：`usedMaterials` 同时包含试剂、耗材、样品（`typeMode` 区分），下料表是否需要默认排除试剂/耗材？当前默认全量列出。
+2. **take-out 失败是否阻塞工作流**：本计划暂定返回 `success=False` 并 warning，不抛异常；如果希望奔耀仓位状态必须一致，可改为抛 `RuntimeError`。
+3. **后续缓存/重跑能力**：如果要支持 push 早到、忘勾选后重跑复用 `unloadTable`，后续应在 `BioyondPeptideStation` station runtime 上实现 `unload_context_cache`，但本轮不做。
+4. **多订单**：本轮只保留 `order_ids` 占位，不实现多订单等待、乱序回调或并发 wait。
+
+---
+
+## 附录 A：API schema 核对摘要
+
+使用 `temp_benyao/scripts/api_helper.py --root temp_benyao/peptide` 核对：
+
+### A.1 `POST /api/lims/storage/material-info`
+
+请求：
+
+```json
+{
+  "apiKey": "B10B5995",
+  "requestTime": "2026-05-20T10:50:00.123Z",
+  "data": "<materialId UUID>"
+}
+```
+
+响应关键字段：
+
+```json
+{
+  "id": "<materialId UUID>",
+  "typeName": "样品",
+  "code": "MAT-001",
+  "barCode": "BC-001",
+  "name": "多肽产物",
+  "quantity": 10,
+  "lockQuantity": 0,
+  "unit": "mg",
+  "status": 1,
+  "isUse": true,
+  "locations": [
+    {
+      "id": "<locationId UUID>",
+      "whid": "<warehouse UUID>",
+      "whName": "自动化堆栈",
+      "code": "A1",
+      "x": 1,
+      "y": 1,
+      "z": 1,
+      "quantity": 10
+    }
+  ],
+  "detail": []
+}
+```
+
+注意：schema 没有 `posX/posY/posZ`，本轮也不展示坐标。
+
+### A.2 `POST /api/lims/order/take-out`
+
+请求：
+
+```json
+{
+  "apiKey": "B10B5995",
+  "requestTime": "2026-05-20T10:50:00.123Z",
+  "data": {
+    "orderId": "<orderId UUID>",
+    "preintakeIds": [],
+    "materialIds": ["<materialId UUID>"]
+  }
+}
+```
+
+响应：
+
+```json
+{
+  "code": 1,
+  "message": null,
+  "timestamp": 1779255000000,
+  "data": true
+}
+```
+
+源码中已有 `BioyondV1RPC.take_out(order_id, preintake_ids=None, material_ids=None)`，本轮复用它。
+
+### A.3 `/report/order_finish`
+
+`/report/order_finish` 不在 Peptide JSON OpenAPI specs 中；schema 依据：
+
+- `unilabos/devices/workstation/workstation_http_service.py`
+- `temp_benyao/peptide/docs/reference/api_manual.md`
+
+关键字段：
+
+```json
+{
+  "token": "token-from-lims",
+  "request_time": "2026-05-20 10:50:00.123",
+  "data": {
+    "orderCode": "EXP260520-103000",
+    "orderName": "实验260520-103000",
+    "startTime": "2026-05-20 09:00:00",
+    "endTime": "2026-05-20 10:50:00",
+    "status": "30",
+    "usedMaterials": [
+      {
+        "materialId": "<materialId UUID>",
+        "locationId": "<locationId UUID>",
+        "typeMode": "1",
+        "usedQuantity": 10
+      }
+    ]
+  }
+}
+```
+
+`WorkstationHTTPService` 会把 `usedMaterials[]` 转成 `MaterialUsage` dataclass 列表传给 `process_order_finish_report(report_request, used_materials)`；peptide 输出 `used_materials` handle 前需要转回 JSON dict：
+
+```json
+[
+  {
+    "materialId": "<materialId UUID>",
+    "locationId": "<locationId UUID>",
+    "typeMode": "1",
+    "usedQuantity": 10
+  }
+]
+```
+
+---
+
+## 附录 B：本轮不实现的内容
+
+- 不做 station runtime 的 `unload_context_cache`。
+- 不做多订单。
+- 不做 push 早到后的补救。
+- 不做 failed manual_confirm 原地重跑。
+- 不改前端。
+- 不改 `sample_waste_removal`。

plan/2026-05-21_01_peptide_reset_four_checkbox_plan.md

@@ -0,0 +1,461 @@
+# Peptide Four-Checkbox Reset Plan
+
+Date: 2026-05-21 16:30
+Status: Proposal only / not executed
+
+## Scope
+
+This plan replaces `2026-05-21_1556_peptide_reset_sirna_reference_plan.md` for Peptide reset work.
+
+User direction captured here:
+
+- `take_out` is unnecessary for Peptide reset.
+- Do not add a material-cache refresh checkbox.
+- Change reset to four checkbox-controlled operations:
+  - 调度器复位
+  - 订单状态复位
+  - 库位复位
+  - 仪器复位
+- The first three checkboxes default to checked.
+- The fourth checkbox, 仪器复位 / `reset_devices`, defaults to unchecked.
+- Replace the current public `reset` action with:
+  - `reset_auto`: normal ILab action node. This is the renamed/replaced version of the current reset implementation.
+  - `reset_manual`: manual-confirm action node with a physical cleanup confirmation message.
+
+## Evidence Summary
+
+Current Peptide source:
+
+- Reset action code is currently in `unilabos/devices/workstation/bioyond_studio/peptide_station/peptide_station.py`.
+- Current Peptide reset selects `scheduler_reset`, `reset_order_status`, and `reset_location`, and passes ids to order/location resets.
+- `BioyondV1RPC.reset_devices()` already calls `/api/lims/device/reset-devices` with only `apiKey` and `requestTime`.
+- `BioyondV1RPC.scheduler_reset()` already calls `/api/lims/scheduler/reset` with only `apiKey` and `requestTime`.
+- `BioyondV1RPC.reset_order_status(order_id)` and `reset_location(location_id)` currently send `data`, but live probes showed that omitted `data` succeeds.
+
+Live Peptide no-data reset probes using `temp_benyao/peptide/peptide_station_config.example.json`:
+
+- `POST /api/lims/order/reset-order-status` with request keys `["apiKey", "requestTime"]` returned HTTP 200 and `code=1`.
+- `POST /api/lims/scheduler/reset` with request keys `["apiKey", "requestTime"]` returned HTTP 200 and `code=1`.
+- `POST /api/lims/storage/reset-location` with request keys `["apiKey", "requestTime"]` returned HTTP 200 and `code=1`.
+- `reset-devices` was not live-probed in this session, but the current RPC wrapper already sends no `data`.
+
+Raw findings:
+
+- `temp_benyao/peptide/_findings/2026-05-21_1613_reset_order_status_no_data_live.md`
+- `temp_benyao/peptide/_findings/2026-05-21_1615_remaining_resets_no_data_live.md`
+
+## Proposed Public Actions
+
+### `reset_auto`
+
+Normal action node. This is the auto/no-manual-confirm path. It replaces the current public `reset` action; do not leave a second public `reset` action unless a later compatibility request explicitly asks for an alias.
+
+Checkbox schema rule:
+
+- Use plain `bool` annotations in the action signature.
+- Do not use `Annotated[bool, Field(...)]` for these checkbox params in this implementation plan.
+- The current AST registry schema path does not unwrap `Annotated[...]`; plain `bool` is required so generated JSON Schema marks the fields as boolean and the renderer can show checkboxes.
+- Put human-facing labels/descriptions in the method docstring or action description. If field-level `Field(description=...)` metadata is required later, add registry `Annotated` support and a schema test as a separate change.
+
+Decorator shape:
+
+```python
+@action(
+    always_free=True,
+    goal_default={
+        "reset_scheduler": True,
+        "reset_order_status": True,
+        "reset_location": True,
+        "reset_devices": False,
+    },
+    description="自动复位调度器/订单状态/库位，可选仪器复位",
+)
+def reset_auto(
+    self,
+    reset_scheduler: bool = True,
+    reset_order_status: bool = True,
+    reset_location: bool = True,
+    reset_devices: bool = False,
+    **kwargs: Any,
+) -> Dict[str, Any]:
+    """自动复位调度器/订单状态/库位，可选仪器复位。
+
+    Args:
+        reset_scheduler[调度器复位]: 调用 /api/lims/scheduler/reset，默认勾选。
+        reset_order_status[订单状态复位]: 调用 /api/lims/order/reset-order-status，默认勾选。
+        reset_location[库位复位]: 调用 /api/lims/storage/reset-location，默认勾选。
+        reset_devices[仪器复位]: 调用 /api/lims/device/reset-devices，默认不勾选。
+    """
+    ...
+```
+
+Implementation notes:
+
+- Use real plain-`bool` parameters, not hidden `**kwargs` and not `Annotated`, so the action renderer can expose four checkboxes.
+- Rename/replace the existing `reset` action as `reset_auto`; the implementation should not keep the old id-shaped `reset` action as another public path by default.
+- Keep the three routine reset defaults checked.
+- Keep `reset_devices` unchecked because it can be broader and more disruptive.
+- Do not require or resolve order ids or location ids.
+- Do not call `take_out`.
+- Do not call `refresh_material_cache`.
+
+### `reset_manual`
+
+Manual-confirm node. It should show the operator a physical cleanup warning, then execute the same reset helper as `reset_auto` after the operator confirms.
+
+Actual manual-confirm decorator pattern in this repo:
+
+- Use `@action(node_type=NodeType.MANUAL_CONFIRM)`.
+- Set `always_free=True`.
+- Add `placeholder_keys={"assignee_user_ids": "unilabos_manual_confirm"}`.
+- Include `timeout_seconds: int` and `assignee_user_ids: list[str]`.
+- Add `goal_default` for `timeout_seconds` and `assignee_user_ids`.
+- Manual-confirm actions are normally side-effect-light, but existing Peptide `start_experiment` is already a `MANUAL_CONFIRM` action that performs scheduler start after the operator gate, so a reset-after-confirm pattern is compatible with current Peptide style.
+
+Proposed confirmation text:
+
+```text
+请确认G3、CEM、Tecan、撕膜机、封膜机、打标机、旋转堆栈上下料位、3个转台等位置的物料已清理完毕；
+请开门检查冰箱、IDOT、酶标仪、离心机、LCMS内部没有遗留物料。
+```
+
+Decorator/function shape:
+
+```python
+RESET_MANUAL_CONFIRM_MESSAGE = (
+    "请确认G3、CEM、Tecan、撕膜机、封膜机、打标机、旋转堆栈上下料位、3个转台等位置的物料已清理完毕；\n"
+    "请开门检查冰箱、IDOT、酶标仪、离心机、LCMS内部没有遗留物料。"
+)
+
+@action(
+    always_free=True,
+    node_type=NodeType.MANUAL_CONFIRM,
+    placeholder_keys={"assignee_user_ids": "unilabos_manual_confirm"},
+    goal_default={
+        "reset_scheduler": True,
+        "reset_order_status": True,
+        "reset_location": True,
+        "reset_devices": False,
+        "physical_cleanup_confirmed": False,
+        "timeout_seconds": 3600,
+        "assignee_user_ids": [],
+    },
+    feedback_interval=300,
+    description=RESET_MANUAL_CONFIRM_MESSAGE,
+)
+def reset_manual(
+    self,
+    reset_scheduler: bool = True,
+    reset_order_status: bool = True,
+    reset_location: bool = True,
+    reset_devices: bool = False,
+    physical_cleanup_confirmed: bool = False,
+    timeout_seconds: int = 3600,
+    assignee_user_ids: Optional[List[str]] = None,
+    **kwargs: Any,
+) -> Dict[str, Any]:
+    """人工确认物理清理后执行复位。
+
+    Args:
+        reset_scheduler[调度器复位]: 调用 /api/lims/scheduler/reset，默认勾选。
+        reset_order_status[订单状态复位]: 调用 /api/lims/order/reset-order-status，默认勾选。
+        reset_location[库位复位]: 调用 /api/lims/storage/reset-location，默认勾选。
+        reset_devices[仪器复位]: 调用 /api/lims/device/reset-devices，默认不勾选。
+        physical_cleanup_confirmed[物理清理确认]: 确认清理提示中的物料检查已经完成，默认不勾选。
+    """
+    ...
+```
+
+Execution rule:
+
+- If `physical_cleanup_confirmed` is false, return a blocked result and do not call any reset API.
+- If it is true, call the same internal helper as `reset_auto`.
+- Return `confirmation_message` in the result payload so call logs preserve the exact operator instruction text.
+
+Renderer caveat:
+
+- `description` should carry the warning in generated action metadata.
+- `physical_cleanup_confirmed` must remain a plain `bool` so it renders as a checkbox.
+- The cleanup warning should be carried by the action `description` and the docstring param description. Do not rely on `Field(description=...)` unless registry `Annotated` support has been implemented and tested.
+- If the current frontend does not show action descriptions or docstring field descriptions reliably, add a read-only string parameter such as `confirmation_message: str = RESET_MANUAL_CONFIRM_MESSAGE` with `goal_default`, or use a handle-based display only after renderer behavior is verified.
+
+## Shared Internal Helper
+
+Both public actions should delegate to one helper, for example:
+
+```python
+def _execute_reset_operations(
+    self,
+    *,
+    reset_scheduler: bool,
+    reset_order_status: bool,
+    reset_location: bool,
+    reset_devices: bool,
+) -> Dict[str, Any]:
+    ...
+```
+
+Call order:
+
+1. `scheduler_reset`
+2. `reset_order_status`
+3. `reset_location`
+4. `reset_devices`
+
+Result shape:
+
+```python
+{
+    "selected_operations": [
+        {"key": "reset_scheduler", "label": "调度器复位", "selected": True},
+        {"key": "reset_order_status", "label": "订单状态复位", "selected": True},
+        {"key": "reset_location", "label": "库位复位", "selected": True},
+        {"key": "reset_devices", "label": "仪器复位", "selected": False},
+    ],
+    "executed_calls": [
+        {"operation": "scheduler_reset", "endpoint": "/api/lims/scheduler/reset", "result": {"code": 1}},
+    ],
+    "skipped_operations": [
+        {"operation": "reset_devices", "reason": "checkbox_disabled"},
+    ],
+    "warnings": [],
+}
+```
+
+Failure handling:
+
+- Execute selected operations sequentially and record each result.
+- If an operation returns non-`1` code, add a warning and continue unless the caller later requests fail-fast.
+- If an RPC method raises, catch it, record an error entry, and continue to the next selected operation unless fail-fast is introduced.
+
+## RPC Wrapper Adjustment
+
+Adjust the two id-shaped wrappers to no-data calls:
+
+- `BioyondV1RPC.reset_order_status()` should no longer require `order_id`.
+- `BioyondV1RPC.reset_location()` should no longer require `location_id`.
+
+Current no-data wrappers already exist:
+
+- `scheduler_reset()`
+- `reset_devices()`
+
+Suggested RPC signatures:
+
+```python
+def scheduler_reset(self) -> int: ...
+def reset_order_status(self) -> int: ...
+def reset_location(self) -> int: ...
+def reset_devices(self) -> int: ...
+```
+
+Compatibility option:
+
+```python
+def reset_order_status(self, order_id: Optional[str] = None) -> int:
+    del order_id
+    ...
+
+def reset_location(self, location_id: Optional[str] = None) -> int:
+    del location_id
+    ...
+```
+
+This keeps older code from crashing while making the actual wire request no-data.
+
+## Adjusted Runtime API Schemas
+
+These are the schemas Peptide reset code should target at runtime after the live no-data probes. They intentionally omit `data`, even though OpenAPI models nullable `data` for these endpoints.
+
+All four requests use:
+
+```json
+{
+  "apiKey": "string",
+  "requestTime": "date-time"
+}
+```
+
+No `data` field should be sent by default.
+
+All four responses use:
+
+```json
+{
+  "code": 1,
+  "message": "",
+  "timestamp": 0
+}
+```
+
+### 调度器复位
+
+Endpoint:
+
+```text
+POST /api/lims/scheduler/reset
+```
+
+Adjusted request:
+
+```json
+{
+  "apiKey": "B10B5995",
+  "requestTime": "2026-05-21T08:15:16.494Z"
+}
+```
+
+Live response:
+
+```json
+{
+  "code": 1,
+  "message": "",
+  "timestamp": 1779351316072
+}
+```
+
+Notes:
+
+- OpenAPI says `data` is nullable int32.
+- Live Peptide accepted omitted `data`.
+
+### 订单状态复位
+
+Endpoint:
+
+```text
+POST /api/lims/order/reset-order-status
+```
+
+Adjusted request:
+
+```json
+{
+  "apiKey": "B10B5995",
+  "requestTime": "2026-05-21T08:13:34.750Z"
+}
+```
+
+Live response:
+
+```json
+{
+  "code": 1,
+  "message": "",
+  "timestamp": 1779351214422
+}
+```
+
+Notes:
+
+- OpenAPI says `data` is nullable string.
+- Live Peptide accepted omitted `data`.
+- Do not model this as order-id scoped unless Bioyond confirms backend behavior.
+
+### 库位复位
+
+Endpoint:
+
+```text
+POST /api/lims/storage/reset-location
+```
+
+Adjusted request:
+
+```json
+{
+  "apiKey": "B10B5995",
+  "requestTime": "2026-05-21T08:15:18.924Z"
+}
+```
+
+Live response:
+
+```json
+{
+  "code": 1,
+  "message": "",
+  "timestamp": 1779351318565
+}
+```
+
+Notes:
+
+- OpenAPI says `data` is nullable string.
+- Live Peptide accepted omitted `data`.
+- Do not model this as location-id scoped unless Bioyond confirms backend behavior.
+
+### 仪器复位
+
+Endpoint:
+
+```text
+POST /api/lims/device/reset-devices
+```
+
+Adjusted request:
+
+```json
+{
+  "apiKey": "B10B5995",
+  "requestTime": "date-time"
+}
+```
+
+Expected response shape:
+
+```json
+{
+  "code": 1,
+  "message": "",
+  "timestamp": 0
+}
+```
+
+Notes:
+
+- OpenAPI says `data` is nullable string.
+- Current `BioyondV1RPC.reset_devices()` already sends no `data`.
+- This endpoint was not live-probed in the no-data reset session.
+- Keep checkbox default unchecked.
+
+## Tests To Add Before Implementation
+
+1. `reset_auto` is not `NodeType.MANUAL_CONFIRM`.
+2. `reset_manual` has `node_type=NodeType.MANUAL_CONFIRM`.
+3. `reset_manual` metadata includes:
+   - `always_free=True`
+   - `placeholder_keys={"assignee_user_ids": "unilabos_manual_confirm"}`
+   - `timeout_seconds=3600`
+   - `assignee_user_ids=[]`
+   - `physical_cleanup_confirmed=False`
+4. Both reset actions expose four real boolean params:
+   - `reset_scheduler`
+   - `reset_order_status`
+   - `reset_location`
+   - `reset_devices`
+5. The generated registry schema marks those reset params as JSON Schema `type: boolean`, not `object` or `string`, so the frontend can render checkboxes.
+6. `reset_auto` replaces the current public `reset` action. Unless a later compatibility request adds an alias, no old id-shaped public `reset` action remains.
+7. Goal defaults are:
+   - first three reset checkboxes `True`
+   - `reset_devices=False`
+8. `reset_manual(..., physical_cleanup_confirmed=False)` does not call any RPC reset method.
+9. `reset_auto()` with defaults calls:
+   - `scheduler_reset()`
+   - `reset_order_status()`
+   - `reset_location()`
+   - not `reset_devices()`
+10. `reset_auto(reset_devices=True)` also calls `reset_devices()`.
+11. `reset_order_status()` and `reset_location()` RPC wrappers send no `data` key.
+12. No reset path calls `take_out`.
+13. No reset path calls `refresh_material_cache`.
+
+## Non-Goals
+
+- Do not implement `take_out` in reset.
+- Do not refresh `material_cache` from reset.
+- Do not resolve order ids or location ids for reset.
+- Do not add Project/cache/browser cleanup routes.
+- Do not make `reset_devices` default-on.
+- Do not execute this plan during planning.

unilabos/devices/workstation/bioyond_studio/bioyond_rpc.py

@@ -415,21 +415,25 @@ class BioyondV1RPC(BaseRequest):
             return {}
         return response.get("data", {})
 
-    def reset_location(self, location_id: str) -> int:
+    def reset_location(self, location_id: Optional[str] = None) -> int:
         """复位库位
 
+        现场实测 ``POST /api/lims/storage/reset-location`` 不传 ``data`` 即可成功
+        （见 ``temp_benyao/peptide/_findings/2026-05-21_1615_remaining_resets_no_data_live.md``），
+        因此默认无 ``data`` 字段；保留 ``location_id`` 仅为兼容旧调用，传入会被忽略。
+
         参数:
-            location_id: 库位ID
+            location_id: 兼容入参，已被忽略；新逻辑不再以 location 为粒度复位。
 
         返回值:
             int: 成功返回1，失败返回0
         """
+        del location_id
         response = self.post(
             url=f'{self.host}/api/lims/storage/reset-location',
             params={
                 "apiKey": self.api_key,
                 "requestTime": self.get_current_time_iso8601(),
-                "data": location_id,
             })
         if not response or response['code'] != 1:
             return 0
@@ -779,6 +783,49 @@ class BioyondV1RPC(BaseRequest):
 
         return response.get("data", {})
 
+    def take_out(
+        self,
+        order_id: str,
+        preintake_ids: list[str] | None = None,
+        material_ids: list[str] | None = None,
+    ) -> dict:
+        """取出订单关联通量/物料
+
+        参数:
+            order_id: 订单ID
+            preintake_ids: 通量ID列表，可为空
+            material_ids: 物料ID列表，可为空
+
+        返回值:
+            dict: 服务端响应包，失败返回空字典
+        """
+        if not order_id:
+            self._logger.error("取出订单关联通量/物料错误: 缺少订单ID")
+            return {}
+
+        params = {
+            "orderId": order_id,
+            "preintakeIds": list(preintake_ids or []),
+            "materialIds": list(material_ids or []),
+        }
+
+        response = self.post(
+            url=f'{self.host}/api/lims/order/take-out',
+            params={
+                "apiKey": self.api_key,
+                "requestTime": self.get_current_time_iso8601(),
+                "data": params,
+            })
+
+        if not response:
+            return {}
+
+        if response['code'] != 1:
+            self._logger.error(f"取出订单关联通量/物料错误: {response.get('message', '')}")
+            return response
+
+        return response
+
     def cancel_order(self, json_str: str) -> bool:
         """取消指定任务
 
@@ -886,21 +933,25 @@ class BioyondV1RPC(BaseRequest):
             return {}
         return response.get("data", {})
 
-    def reset_order_status(self, order_id: str) -> int:
+    def reset_order_status(self, order_id: Optional[str] = None) -> int:
         """复位订单状态
 
+        现场实测 ``POST /api/lims/order/reset-order-status`` 不传 ``data`` 即可成功
+        （见 ``temp_benyao/peptide/_findings/2026-05-21_1613_reset_order_status_no_data_live.md``），
+        因此默认无 ``data`` 字段；保留 ``order_id`` 仅为兼容旧调用，传入会被忽略。
+
         参数:
-            order_id: 订单ID
+            order_id: 兼容入参，已被忽略；新逻辑不再以单订单为粒度复位。
 
         返回值:
             int: 成功返回1，失败返回0
         """
+        del order_id
         response = self.post(
             url=f'{self.host}/api/lims/order/reset-order-status',
             params={
                 "apiKey": self.api_key,
                 "requestTime": self.get_current_time_iso8601(),
-                "data": order_id,
             })
         if not response or response['code'] != 1:
             return 0

unilabos/devices/workstation/bioyond_studio/debug_call_log.py

@@ -8,7 +8,7 @@ returned. Outside of an active session the wrapped method delegates to the
 original (unwrapped) implementation, leaving non-debug behavior intact.
 
 The session writes a Markdown file under ``out_dir`` mirroring the format of
-``temp_benyao/peptide/_logs/2026-04-30_160316_day3_samplefile_only_raw_calls.md``
+``bioyond_debug_records/2026-04-30_160316_day3_samplefile_only_raw_calls.md``
 minus the "Raw Payload Argument" section.
 
 This module has no dependency on ``BioyondV1RPC`` itself; the only contract is

unilabos/devices/workstation/bioyond_studio/peptide_station/__init__.py

@@ -0,0 +1,3 @@
+from .peptide_station import BioyondPeptideStation, fetch_workflow_list, load_peptide_config
+
+__all__ = ["BioyondPeptideStation", "fetch_workflow_list", "load_peptide_config"]

unilabos/devices/workstation/bioyond_studio/sirna_station/__init__.py

@@ -1,3 +0,0 @@
-from .sirna_station import BioyondSirnaStation, fetch_workflow_list, load_sirna_config
-
-__all__ = ["BioyondSirnaStation", "fetch_workflow_list", "load_sirna_config"]

unilabos/devices/workstation/bioyond_studio/station.py

@@ -56,13 +56,17 @@ class ConnectionMonitor:
     def _monitor_loop(self):
         while self._running:
             try:
-                # 使用 lightweight API 检查连接
-                # query_matial_type_list 是比较快的查询
-                start_time = time.time()
-                result = self.workstation.hardware_interface.material_type_list()
+                # 使用轻量级调度状态接口检查连接，避免启动时打印完整物料类型列表。
+                result = self.workstation.hardware_interface.scheduler_status()
 
                 status = "online" if result else "offline"
-                msg = "Connection established" if status == "online" else "Failed to get material type list"
+                if status == "online":
+                    msg = (
+                        f"Scheduler status={result.get('status')}, "
+                        f"hasTask={result.get('hasTask')}"
+                    )
+                else:
+                    msg = "Failed to get scheduler status"
 
                 if status != self._last_status:
                     logger.info(f"Bioyond连接状态变更: {self._last_status} -> {status}")
@@ -730,7 +734,7 @@ class BioyondWorkstation(WorkstationBase):
         """解析 ``debug_log_dir`` 为绝对路径。"""
         configured = (getattr(self, "bioyond_config", {}) or {}).get("debug_log_dir")
         default_dir = getattr(self, "_DEBUG_LOG_DEFAULT_DIR", None)
-        candidate = configured or default_dir or "temp_benyao/_logs/bioyond_debug"
+        candidate = configured or default_dir or "bioyond_debug_records"
         path = Path(candidate)
         if not path.is_absolute():
             repo_root = Path(__file__).resolve().parents[4]

unilabos/resources/bioyond/__init__.py

@@ -1 +1,9 @@
-from . import sirna_materials  # noqa: F401  ensure @resource classes are importable for PLR deserialize
+try:
+    from . import peptide_materials  # noqa: F401  ensure @resource classes are importable for PLR deserialize
+except Exception:  # pragma: no cover - 允许轻量环境导入非资源辅助函数
+    peptide_materials = None  # type: ignore[assignment]
+
+try:
+    from . import sirna_materials  # noqa: F401  ensure @resource classes are importable for PLR deserialize
+except Exception:  # pragma: no cover - 允许轻量环境导入非资源辅助函数
+    sirna_materials = None  # type: ignore[assignment]

unilabos/resources/bioyond/decks.py

@@ -258,13 +258,13 @@ class BIOYOND_PeptideStation_Deck(Deck):
         ],
         "xy_col_row",
     )
-    WAREHOUSE_BIOYOND_KEY_AXIS = dict.fromkeys(WAREHOUSE_BIOYOND_AXIS, "row_col")
+    WAREHOUSE_BIOYOND_KEY_AXIS = dict.fromkeys(WAREHOUSE_BIOYOND_AXIS, "col_row")
 
     def __init__(
         self,
         name: str = "PeptideStation_Deck",
-        size_x: float = 3500.0,
-        size_y: float = 1800.0,
+        size_x: float = 2700.0,
+        size_y: float = 2000.0,
         size_z: float = 1500.0,
         category: str = "deck",
         setup: bool = False
@@ -276,6 +276,8 @@ class BIOYOND_PeptideStation_Deck(Deck):
     @classmethod
     def deserialize(cls, data: dict, allow_marshal: bool = False):
         if data.get("children") and data.get("setup") is True:
+            data = data.copy()
+            data["setup"] = False
             # 已有序列化子资源，跳过 setup 避免重复创建
             result = super(BIOYOND_PeptideStation_Deck, cls).deserialize(data, allow_marshal=allow_marshal)
         else:
@@ -293,81 +295,140 @@ class BIOYOND_PeptideStation_Deck(Deck):
             if key_axis and not hasattr(child, "bioyond_key_axis"):
                 child.bioyond_key_axis = key_axis
 
+    def _frontend_y_flipped_coordinate(self, display_x: float, display_y: float, child) -> Coordinate:
+        """把期望显示坐标转换为兼容前端 y 轴翻转的存储坐标。"""
+        return Coordinate(display_x, self.get_size_y() - display_y - child.get_size_y(), 0.0)
+
     def setup(self) -> None:
         # 多肽工作站仓库配置
         # 基于 2026-05-09 live API probe 发现的实际仓库拓扑 (12个仓库)
-        # 数据来源: temp_benyao/peptide/_logs/warehouse_discovery_raw_live_2026-05-09.json
+        # 数据来源: Bioyond 现场仓库发现结果。
         self.warehouses = {
-            # 主自动化堆栈 - live API: code 10-17 -> x=17, y=10，显示为 10 行×17 列
+            # 主自动化堆栈 - live API: code 10-17 -> x=17, y=10，显示为 17 行×10 列
             "自动化堆栈": bioyond_warehouse_numeric_stack(
-                "自动化堆栈", rows=10, columns=17, bioyond_axis="xy_col_row", bioyond_key_axis="row_col"
+                "自动化堆栈",
+                rows=17,
+                columns=10,
+                bioyond_axis="xy_col_row",
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
 
             # 低温存储
             "低温冰箱仓库": bioyond_warehouse_live_grid(
-                "低温冰箱仓库", rows=2, columns=3, slot_keys=["1", "2", "3", "4", "5", "6"]
+                "低温冰箱仓库",
+                rows=3,
+                columns=2,
+                slot_keys=["1", "2", "3", "4", "5", "6"],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
 
             # 移液站库位
             "Tecan移液站库": bioyond_warehouse_live_grid(
-                "Tecan移液站库", rows=1, columns=18, slot_keys=[str(index) for index in range(1, 19)]
+                "Tecan移液站库",
+                rows=18,
+                columns=1,
+                slot_keys=[str(index) for index in range(1, 19)],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
             "G3移液站库": bioyond_warehouse_live_grid(
                 "G3移液站库",
-                rows=1,
-                columns=18,
+                rows=18,
+                columns=1,
                 slot_keys=["1", "2", "3", "4", "垃圾桶", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15", "16", "17", "18"],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
             "IDOT移液站库": bioyond_warehouse_live_grid(
                 "IDOT移液站库",
-                rows=1,
-                columns=12,
+                rows=12,
+                columns=1,
                 slot_keys=[f"0009-{index:04d}" for index in range(1, 13)],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
 
             # 缓冲库位
             "G3缓冲库": bioyond_warehouse_live_grid(
-                "G3缓冲库", rows=1, columns=5, slot_keys=[str(index) for index in range(1, 6)]
+                "G3缓冲库",
+                rows=5,
+                columns=1,
+                slot_keys=[str(index) for index in range(1, 6)],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
             "盖板缓冲库": bioyond_warehouse_live_grid(
-                "盖板缓冲库", rows=1, columns=7, slot_keys=[str(index) for index in range(1, 8)]
+                "盖板缓冲库",
+                rows=7,
+                columns=1,
+                slot_keys=[str(index) for index in range(1, 8)],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
             "配平板缓冲库": bioyond_warehouse_live_grid(
-                "配平板缓冲库", rows=1, columns=3, slot_keys=[str(index) for index in range(1, 4)]
+                "配平板缓冲库",
+                rows=3,
+                columns=1,
+                slot_keys=[str(index) for index in range(1, 4)],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
             "IDOT缓冲库": bioyond_warehouse_live_grid(
-                "IDOT缓冲库", rows=1, columns=2, slot_keys=["1", "1"]
+                "IDOT缓冲库",
+                rows=2,
+                columns=1,
+                slot_keys=["1", "1"],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
             "固相合成板底座缓冲位": bioyond_warehouse_live_grid(
                 "固相合成板底座缓冲位",
-                rows=1,
-                columns=4,
+                rows=4,
+                columns=1,
                 slot_keys=[f"0015-{index:04d}" for index in range(1, 5)],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
 
             # 设备库位
             "离心机库位": bioyond_warehouse_live_grid(
-                "离心机库位", rows=1, columns=4, slot_keys=[f"0017-{index:04d}" for index in range(1, 5)]
+                "离心机库位",
+                rows=4,
+                columns=1,
+                slot_keys=[f"0017-{index:04d}" for index in range(1, 5)],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
             "热封膜机位": bioyond_warehouse_live_grid(
-                "热封膜机位", rows=1, columns=2, slot_keys=[f"0016-{index:04d}" for index in range(1, 3)]
+                "热封膜机位",
+                rows=2,
+                columns=1,
+                slot_keys=[f"0016-{index:04d}" for index in range(1, 3)],
+                bioyond_key_axis="col_row",
+                frontend_y_flip=True,
             ),
         }
 
-        # 仓库位置布局 (需根据实际硬件布局调整)
+        # 仓库显示布局：紧凑排列；存储 y 坐标按前端兼容翻转预先反向。
+        display_layout = {
+            "自动化堆栈": (0.0, 0.0),
+            "Tecan移液站库": (1520.0, 0.0),
+            "G3移液站库": (1710.0, 0.0),
+            "IDOT移液站库": (1900.0, 0.0),
+            "G3缓冲库": (2090.0, 0.0),
+            "盖板缓冲库": (2090.0, 580.0),
+            "低温冰箱仓库": (2280.0, 0.0),
+            "配平板缓冲库": (2280.0, 370.0),
+            "IDOT缓冲库": (2470.0, 370.0),
+            "固相合成板底座缓冲位": (2280.0, 740.0),
+            "离心机库位": (2470.0, 740.0),
+            "热封膜机位": (2280.0, 1210.0),
+        }
         self.warehouse_locations = {
-            "自动化堆栈": Coordinate(0.0, 0.0, 0.0),
-            "Tecan移液站库": Coordinate(0.0, 1150.0, 0.0),
-            "G3移液站库": Coordinate(0.0, 1300.0, 0.0),
-            "IDOT移液站库": Coordinate(0.0, 1450.0, 0.0),
-            "G3缓冲库": Coordinate(0.0, 1600.0, 0.0),
-            "盖板缓冲库": Coordinate(850.0, 1600.0, 0.0),
-            "低温冰箱仓库": Coordinate(2700.0, 0.0, 0.0),
-            "配平板缓冲库": Coordinate(2700.0, 300.0, 0.0),
-            "IDOT缓冲库": Coordinate(2700.0, 450.0, 0.0),
-            "固相合成板底座缓冲位": Coordinate(2700.0, 600.0, 0.0),
-            "离心机库位": Coordinate(2700.0, 750.0, 0.0),
-            "热封膜机位": Coordinate(2700.0, 900.0, 0.0),
+            name: self._frontend_y_flipped_coordinate(x, y, self.warehouses[name])
+            for name, (x, y) in display_layout.items()
         }
 
         for warehouse_name, warehouse in self.warehouses.items():

unilabos/resources/bioyond/peptide_materials.py

@@ -0,0 +1,247 @@
+"""Peptide Station Material Resource Definitions."""
+
+from __future__ import annotations
+
+from collections import OrderedDict
+
+try:
+    from pylabrobot.resources import Container, Plate, TipRack
+except Exception:  # pragma: no cover - 允许无 pylabrobot 的轻量动作测试导入
+    class _FallbackResource:
+        def __init__(self, *args, **kwargs):
+            self.args = args
+            self.kwargs = kwargs
+
+    class Container(_FallbackResource):  # type: ignore[no-redef]
+        pass
+
+    class Plate(_FallbackResource):  # type: ignore[no-redef]
+        pass
+
+    class TipRack(_FallbackResource):  # type: ignore[no-redef]
+        pass
+
+try:
+    from unilabos.registry.decorators import resource
+except Exception:  # pragma: no cover - 允许无完整 registry 依赖时导入常量
+    def resource(*args, **kwargs):
+        def decorator(cls):
+            return cls
+
+        return decorator
+
+
+def _ensure_itemized_ordering(kwargs: dict) -> None:
+    if kwargs.get("ordering") is None and kwargs.get("ordered_items") is None:
+        kwargs["ordering"] = OrderedDict()
+
+
+class _PeptideTipRack(TipRack):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("size_x", 127.76)
+        kwargs.setdefault("size_y", 85.48)
+        kwargs.setdefault("size_z", 64.0)
+        kwargs.setdefault("with_tips", True)
+        _ensure_itemized_ordering(kwargs)
+        super().__init__(*args, **kwargs)
+
+
+class _PeptidePlate(Plate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("size_x", 127.76)
+        kwargs.setdefault("size_y", 85.48)
+        kwargs.setdefault("size_z", 14.35)
+        kwargs.setdefault("plate_type", "skirted")
+        _ensure_itemized_ordering(kwargs)
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_1000ul_tip_rack",
+    category=["labware", "tip_rack"],
+    description="1000uL tip rack for Bioyond peptide station",
+)
+class BioyondPeptide_1000ul_TipRack(_PeptideTipRack):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_1000ul_tip_rack")
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_200ul_tip_rack",
+    category=["labware", "tip_rack"],
+    description="200uL tip rack for Bioyond peptide station",
+)
+class BioyondPeptide_200ul_TipRack(_PeptideTipRack):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_200ul_tip_rack")
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_50ul_tip_rack",
+    category=["labware", "tip_rack"],
+    description="50uL tip rack for Bioyond peptide station",
+)
+class BioyondPeptide_50ul_TipRack(_PeptideTipRack):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_50ul_tip_rack")
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_96_well_deep_well_plate",
+    category=["labware", "plate"],
+    description="96 well deep well plate for Bioyond peptide station",
+)
+class BioyondPeptide_96WellDeepWellPlate(_PeptidePlate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_96_well_deep_well_plate")
+        kwargs.setdefault("size_z", 44.0)
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_96_well_synthesis_plate",
+    category=["labware", "plate"],
+    description="96 well solid-phase synthesis plate for Bioyond peptide station",
+)
+class BioyondPeptide_96WellSynthesisPlate(_PeptidePlate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_96_well_synthesis_plate")
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_96_well_synthesis_plate_base",
+    category=["labware", "adapter"],
+    description="96 well solid-phase synthesis plate base for Bioyond peptide station",
+)
+class BioyondPeptide_96WellSynthesisPlateBase(_PeptidePlate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_96_well_synthesis_plate_base")
+        kwargs.setdefault("size_z", 20.0)
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_96_well_balance_plate",
+    category=["labware", "plate"],
+    description="96 well balance plate for Bioyond peptide station",
+)
+class BioyondPeptide_96WellBalancePlate(_PeptidePlate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_96_well_balance_plate")
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_384_well_plate",
+    category=["labware", "plate"],
+    description="384 well plate for Bioyond peptide station",
+)
+class BioyondPeptide_384WellPlate(_PeptidePlate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_384_well_plate")
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_384_lcms_plate",
+    category=["labware", "plate"],
+    description="384 well LCMS plate for Bioyond peptide station",
+)
+class BioyondPeptide_384LCMSPlate(_PeptidePlate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_384_lcms_plate")
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_384_balance_plate",
+    category=["labware", "plate"],
+    description="384 well balance plate for Bioyond peptide station",
+)
+class BioyondPeptide_384BalancePlate(_PeptidePlate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_384_balance_plate")
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_cover_plate",
+    category=["labware", "cover"],
+    description="Cover plate for Bioyond peptide station",
+)
+class BioyondPeptide_CoverPlate(_PeptidePlate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_cover_plate")
+        kwargs.setdefault("size_z", 8.0)
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_sealing_base",
+    category=["labware", "adapter"],
+    description="Sealing base for Bioyond peptide station",
+)
+class BioyondPeptide_SealingBase(_PeptidePlate):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("model", "bioyond_peptide_sealing_base")
+        kwargs.setdefault("size_z", 20.0)
+        super().__init__(*args, **kwargs)
+
+
+@resource(
+    id="bioyond_peptide_reagent_trough",
+    category=["labware", "trough"],
+    description="Reagent trough for Bioyond peptide station",
+)
+class BioyondPeptide_ReagentTrough(Container):
+    def __init__(self, *args, **kwargs):
+        kwargs.setdefault("size_x", 127.76)
+        kwargs.setdefault("size_y", 85.48)
+        kwargs.setdefault("size_z", 44.0)
+        kwargs.setdefault("max_volume", 300000.0)
+        kwargs.setdefault("model", "bioyond_peptide_reagent_trough")
+        super().__init__(*args, **kwargs)
+
+
+DEFAULT_PEPTIDE_MATERIAL_TYPE_MAPPINGS = {
+    "bioyond_peptide_1000ul_tip_rack": ["1000μL枪头盒", "3a1890bb-736e-cfdd-3213-eb314e8a60f9"],
+    "bioyond_peptide_200ul_tip_rack": ["200μL枪头盒", "3a1890bb-36d1-964a-18bd-0bf0f2877a7b"],
+    "bioyond_peptide_50ul_tip_rack": ["50μL枪头盒", "3a1890bc-5fae-361c-cc09-e6f2f6dcd71d"],
+    "bioyond_peptide_96_well_deep_well_plate": ["96孔深孔板", "3a1890bc-1fa8-fe39-9faa-12279ed4569b"],
+    "bioyond_peptide_96_well_synthesis_plate": ["96孔固相合成板", "3a1871cb-99f3-f01d-23e2-08dbbd0045b5"],
+    "bioyond_peptide_96_well_synthesis_plate_base": ["96孔固相合成板底座", "3a1b997e-241b-64f0-80d1-47bca08799d1"],
+    "bioyond_peptide_96_well_balance_plate": ["96孔配平板", "3a187661-2378-1e20-fa5c-a27d49fdc15d"],
+    "bioyond_peptide_384_well_plate": ["384孔酶标板", "3a1890bf-2148-ed20-92bd-d85869947d9a"],
+    "bioyond_peptide_384_lcms_plate": ["384孔LCMS板", "3a1e6a8b-cb61-74da-a089-8e6f197f80f0"],
+    "bioyond_peptide_384_balance_plate": ["384孔配平板", "3a18be7e-47cc-888c-fc68-055753286826"],
+    "bioyond_peptide_cover_plate": ["防挥发盖板", "3a19d5a6-b0e2-b486-e5eb-bcabc632f4de"],
+    "bioyond_peptide_sealing_base": ["封膜底座", "3a1d1d7b-e33b-6975-165d-c56cba5ed345"],
+    "bioyond_peptide_reagent_trough": ["12道试剂槽", "3a18b431-ac58-ca2e-9680-2a4f5880ea45"],
+}
+
+
+MATERIAL_TYPE_CODE_TO_CLASS = {
+    "0001": BioyondPeptide_96WellSynthesisPlate,
+    "0002": BioyondPeptide_96WellBalancePlate,
+    "0008": BioyondPeptide_200ul_TipRack,
+    "0009": BioyondPeptide_1000ul_TipRack,
+    "0011": BioyondPeptide_96WellDeepWellPlate,
+    "0012": BioyondPeptide_50ul_TipRack,
+    "0016": BioyondPeptide_384WellPlate,
+    "0018": BioyondPeptide_384WellPlate,
+    "0024": BioyondPeptide_ReagentTrough,
+    "0026": BioyondPeptide_384BalancePlate,
+    "0035": BioyondPeptide_CoverPlate,
+    "0039": BioyondPeptide_96WellSynthesisPlateBase,
+    "0041": BioyondPeptide_SealingBase,
+    "0049": BioyondPeptide_384LCMSPlate,
+}
+
+
+def get_material_class_by_type_code(type_code: str):
+    """Return a peptide material class by Bioyond material type code."""
+    return MATERIAL_TYPE_CODE_TO_CLASS.get(type_code)

unilabos/resources/bioyond/sirna_materials.py

@@ -1,126 +0,0 @@
-"""Sirna Station Material Resource Definitions
-
-Defines PyLabRobot resource classes for Bioyond Sirna station materials.
-Each class is decorated with @resource for AST-based registry discovery.
-"""
-
-from collections import OrderedDict
-
-from pylabrobot.resources import Plate, TipRack, Container
-
-from unilabos.registry.decorators import resource
-
-
-@resource(
-    id="bioyond_sirna_g3_200ul_tip_rack",
-    category=["labware", "tip_rack"],
-    description="G3-200ul枪头盒 for Sirna station",
-)
-class BioyondSirna_G3_200ul_TipRack(TipRack):
-    """G3-200ul tip rack for Sirna liquid handling."""
-
-    def __init__(self, *args, **kwargs):
-        kwargs.setdefault("size_x", 127.76)
-        kwargs.setdefault("size_y", 85.48)
-        kwargs.setdefault("size_z", 64.0)
-        kwargs.setdefault("model", "bioyond_sirna_g3_200ul_tip_rack")
-        kwargs.setdefault("with_tips", True)
-        if kwargs.get("ordering") is None and kwargs.get("ordered_items") is None:
-            kwargs["ordering"] = OrderedDict()
-        super().__init__(*args, **kwargs)
-
-
-@resource(
-    id="bioyond_sirna_g3_50ul_tip_rack",
-    category=["labware", "tip_rack"],
-    description="G3-50ul枪头盒 for Sirna station",
-)
-class BioyondSirna_G3_50ul_TipRack(TipRack):
-    """G3-50ul tip rack for Sirna liquid handling."""
-
-    def __init__(self, *args, **kwargs):
-        kwargs.setdefault("size_x", 127.76)
-        kwargs.setdefault("size_y", 85.48)
-        kwargs.setdefault("size_z", 64.0)
-        kwargs.setdefault("model", "bioyond_sirna_g3_50ul_tip_rack")
-        kwargs.setdefault("with_tips", True)
-        if kwargs.get("ordering") is None and kwargs.get("ordered_items") is None:
-            kwargs["ordering"] = OrderedDict()
-        super().__init__(*args, **kwargs)
-
-
-@resource(
-    id="bioyond_sirna_384_well_plate",
-    category=["labware", "plate"],
-    description="384孔板 for Sirna assays",
-)
-class BioyondSirna_384WellPlate(Plate):
-    """384-well plate for Sirna reporter gene detection."""
-
-    def __init__(self, *args, **kwargs):
-        kwargs.setdefault("size_x", 127.76)
-        kwargs.setdefault("size_y", 85.48)
-        kwargs.setdefault("size_z", 14.35)
-        kwargs.setdefault("model", "bioyond_sirna_384_well_plate")
-        kwargs.setdefault("plate_type", "skirted")
-        if kwargs.get("ordering") is None and kwargs.get("ordered_items") is None:
-            kwargs["ordering"] = OrderedDict()
-        super().__init__(*args, **kwargs)
-
-
-@resource(
-    id="bioyond_sirna_cell_culture_plate",
-    category=["labware", "plate"],
-    description="细胞培养板 for Sirna cell culture",
-)
-class BioyondSirna_CellCulturePlate(Plate):
-    """Cell culture plate for Sirna experiments."""
-
-    def __init__(self, *args, **kwargs):
-        kwargs.setdefault("size_x", 127.76)
-        kwargs.setdefault("size_y", 85.48)
-        kwargs.setdefault("size_z", 14.35)
-        kwargs.setdefault("model", "bioyond_sirna_cell_culture_plate")
-        kwargs.setdefault("plate_type", "skirted")
-        if kwargs.get("ordering") is None and kwargs.get("ordered_items") is None:
-            kwargs["ordering"] = OrderedDict()
-        super().__init__(*args, **kwargs)
-
-
-@resource(
-    id="bioyond_sirna_reagent_trough",
-    category=["labware", "trough"],
-    description="试剂槽 for Sirna reagents",
-)
-class BioyondSirna_ReagentTrough(Container):
-    """Reagent trough for Sirna station reagents (RiboGreen, etc.)."""
-
-    def __init__(self, *args, **kwargs):
-        kwargs.setdefault("size_x", 127.76)
-        kwargs.setdefault("size_y", 85.48)
-        kwargs.setdefault("size_z", 44.0)
-        kwargs.setdefault("max_volume", 300000.0)
-        kwargs.setdefault("model", "bioyond_sirna_reagent_trough")
-        super().__init__(*args, **kwargs)
-
-
-# Material type code mapping for dynamic instantiation
-MATERIAL_TYPE_CODE_TO_CLASS = {
-    "0016": BioyondSirna_G3_200ul_TipRack,
-    "0017": BioyondSirna_G3_50ul_TipRack,
-    "0015": BioyondSirna_384WellPlate,
-    "0001": BioyondSirna_CellCulturePlate,
-    "0006": BioyondSirna_ReagentTrough,
-}
-
-
-def get_material_class_by_type_code(type_code: str):
-    """Get resource class by Bioyond material type code.
-
-    Args:
-        type_code: Bioyond materialTypeCode (e.g., "0016", "0017")
-
-    Returns:
-        Resource class or None if not found
-    """
-    return MATERIAL_TYPE_CODE_TO_CLASS.get(type_code)

unilabos/resources/bioyond/warehouses.py

@@ -25,6 +25,7 @@ def bioyond_warehouse_numeric_stack(
     columns: int = 17,
     bioyond_axis: str = "xy_row_col",
     bioyond_key_axis: str = "row_col",
+    frontend_y_flip: bool = False,
 ) -> WareHouse:
     """创建 Bioyond 数字库位堆栈，库位名使用服务端返回的 行-列 格式。
 
@@ -46,17 +47,22 @@ def bioyond_warehouse_numeric_stack(
     item_dx = 147.0
     item_dy = 106.0
     item_dz = 130.0
-    locations = [
-        Coordinate(dx + col * item_dx, dy + row * item_dy, dz)
-        for row in range(num_items_y)
-        for col in range(num_items_x)
-    ]
+    resource_size_x = 127.0
+    resource_size_y = 86.0
+    resource_size_z = 25.0
+    size_y = dy + item_dy * num_items_y
+    locations = []
+    for row in range(num_items_y):
+        display_y = dy + row * item_dy
+        y = size_y - display_y - resource_size_y if frontend_y_flip else display_y
+        for col in range(num_items_x):
+            locations.append(Coordinate(dx + col * item_dx, y, dz))
     holders = create_homogeneous_resources(
         klass=ResourceHolder,
         locations=locations,
-        resource_size_x=127.0,
-        resource_size_y=86.0,
-        resource_size_z=25.0,
+        resource_size_x=resource_size_x,
+        resource_size_y=resource_size_y,
+        resource_size_z=resource_size_z,
         name_prefix=name,
     )
     if bioyond_key_axis == "row_col":
@@ -76,7 +82,7 @@ def bioyond_warehouse_numeric_stack(
     warehouse = BioyondWareHouse(
         name=name,
         size_x=dx + item_dx * num_items_x,
-        size_y=dy + item_dy * num_items_y,
+        size_y=size_y,
         size_z=dz + item_dz * num_items_z,
         num_items_x=num_items_x,
         num_items_y=num_items_y,
@@ -97,6 +103,7 @@ def bioyond_warehouse_live_grid(
     slot_keys: list[str] | None = None,
     bioyond_axis: str = "xy_col_row",
     bioyond_key_axis: str = "row_col",
+    frontend_y_flip: bool = False,
 ) -> WareHouse:
     """创建 Bioyond 实测库位网格，按服务端 code 保存位点标签。
 
@@ -112,17 +119,22 @@ def bioyond_warehouse_live_grid(
     item_dx = 147.0
     item_dy = 106.0
     item_dz = 130.0
-    locations = [
-        Coordinate(dx + col * item_dx, dy + row * item_dy, dz)
-        for row in range(num_items_y)
-        for col in range(num_items_x)
-    ]
+    resource_size_x = 127.0
+    resource_size_y = 86.0
+    resource_size_z = 25.0
+    size_y = dy + item_dy * num_items_y
+    locations = []
+    for row in range(num_items_y):
+        display_y = dy + row * item_dy
+        y = size_y - display_y - resource_size_y if frontend_y_flip else display_y
+        for col in range(num_items_x):
+            locations.append(Coordinate(dx + col * item_dx, y, dz))
     holders = create_homogeneous_resources(
         klass=ResourceHolder,
         locations=locations,
-        resource_size_x=127.0,
-        resource_size_y=86.0,
-        resource_size_z=25.0,
+        resource_size_x=resource_size_x,
+        resource_size_y=resource_size_y,
+        resource_size_z=resource_size_z,
         name_prefix=name,
     )
     keys = slot_keys or [str(index + 1) for index in range(num_items_x * num_items_y)]
@@ -139,7 +151,7 @@ def bioyond_warehouse_live_grid(
     return BioyondWareHouse(
         name=name,
         size_x=dx + item_dx * num_items_x,
-        size_y=dy + item_dy * num_items_y,
+        size_y=size_y,
         size_z=dz + item_dz * num_items_z,
         num_items_x=num_items_x,
         num_items_y=num_items_y,

unilabos/resources/plr_additional_res_reg.py

@@ -18,3 +18,7 @@ def register():
     from unilabos.devices.liquid_handling.rviz_backend import UniLiquidHandlerRvizBackend
     from unilabos.devices.liquid_handling.laiyu.backend.laiyu_v_backend import UniLiquidHandlerLaiyuBackend
 
+    # noinspection PyUnresolvedReferences
+    from unilabos.resources.bioyond.decks import BIOYOND_SirnaStation_Deck
+    # noinspection PyUnresolvedReferences
+    from unilabos.resources.bioyond.decks import BIOYOND_PeptideStation_Deck