add:skill&agent

This commit is contained in:
ZiWei
2026-03-06 16:54:31 +08:00
parent 9c6f7c7505
commit 47a29a0c2f
14 changed files with 4214 additions and 0 deletions

View File

@@ -0,0 +1,24 @@
---
name: add-device
description: Guide for adding new devices to Uni-Lab-OS (接入新设备). Walks through device category selection (thing model), communication protocol, command protocol collection, driver creation, registry YAML, and graph file setup. Use when the user wants to add/integrate a new device, create a device driver, write a device class, configure device registry, or mentions 接入设备/添加设备/设备驱动/物模型.
---
# 添加新设备到 Uni-Lab-OS
**第一步:** 使用 Read 工具读取 `docs/ai_guides/add_device.md`,获取完整的设备接入指南并严格遵循。
该指南包含:
- 8 步完整流程(设备类别、通信协议、指令收集、接口对齐、驱动创建、注册表、图文件、验证)
- 所有物模型代码模板(注射泵、电磁阀、蠕动泵、温控、电机等)
- 通信协议代码片段Serial、Modbus、TCP、HTTP、OPC UA
- 现有设备接口快照用于第四步对齐包含参数名、status_types、方法签名
- 常见错误检查清单
**Cursor 工具映射:**
| 指南中的操作 | Cursor 中使用的工具 |
|---|---|
| 向用户确认设备类别、协议等信息 | 使用 AskQuestion 工具 |
| 搜索已有设备注册表 | 使用 Grep 在 `unilabos/registry/devices/` 中搜索 |
| 读取用户提供的协议文档/SDK 代码 | 使用 Read 工具 |
| 第四步对齐:查找同类设备接口 | 优先使用 Grep 搜索仓库中的最新注册表;指南中的「现有设备接口快照」作为兜底参考 |

View File

@@ -0,0 +1,323 @@
---
name: add-protocol
description: Guide for adding new experiment protocols to Uni-Lab-OS (添加新实验操作协议). Walks through ROS Action definition, Pydantic model creation, protocol generator implementation, and registration. Use when the user wants to add a new protocol, create a compile function, implement an experiment operation, or mentions 协议/protocol/编译/compile/实验操作.
---
# 添加新实验操作协议Protocol
Protocol 是对实验有意义的完整动作(如泵转移、过滤、溶解),需要多设备协同。`compile/` 中的生成函数根据设备连接图将抽象操作"编译"为设备指令序列。
添加一个 Protocol 需修改 **6 个文件**,按以下流程执行。
---
## 第一步:确认协议信息
向用户确认:
| 信息 | 示例 |
|------|------|
| 协议英文名 | `MyNewProtocol` |
| 操作描述 | 将固体样品研磨至目标粒径 |
| Goal 参数(必需 + 可选) | `vessel: dict`, `time: float = 300.0` |
| Result 字段 | `success: bool`, `message: str` |
| 需要哪些设备协同 | 研磨器、搅拌器 |
---
## 第二步:创建 ROS Action 定义
路径:`unilabos_msgs/action/<ActionName>.action`
三段式结构Goal / Result / Feedback`---` 分隔:
```
# Goal
Resource vessel
float64 time
string mode
---
# Result
bool success
string return_info
---
# Feedback
string status
string current_device
builtin_interfaces/Duration time_spent
builtin_interfaces/Duration time_remaining
```
**类型映射:**
| Python 类型 | ROS 类型 | 说明 |
|------------|----------|------|
| `dict` | `Resource` | 容器/设备引用,自定义消息类型 |
| `float` | `float64` | |
| `int` | `int32` | |
| `str` | `string` | |
| `bool` | `bool` | |
> `Resource` 是 `unilabos_msgs/msg/Resource.msg` 中定义的自定义消息类型。
---
## 第三步:注册 Action 到 CMakeLists
`unilabos_msgs/CMakeLists.txt``set(action_files ...)` 块中添加:
```cmake
"action/MyNewAction.action"
```
> 调试时需编译:`cd unilabos_msgs && colcon build && source ./install/local_setup.sh && cd ..`
> PR 合并后 CI/CD 自动发布,`mamba update ros-humble-unilabos-msgs` 即可。
---
## 第四步:创建 Pydantic 模型
`unilabos/messages/__init__.py` 中添加(位于 `# Start Protocols``# End Protocols` 之间):
```python
class MyNewProtocol(BaseModel):
# === 必需参数 ===
vessel: dict = Field(..., description="目标容器")
# === 可选参数 ===
time: float = Field(300.0, description="操作时间 (秒)")
mode: str = Field("default", description="操作模式")
def model_post_init(self, __context):
"""参数验证和修正"""
if self.time <= 0:
self.time = 300.0
```
**规则:**
- 参数名必须与 `.action` 文件中 Goal 字段完全一致
- `dict` 类型对应 `.action` 中的 `Resource`
- 将类名加入文件末尾的 `__all__` 列表
---
## 第五步:实现协议生成函数
路径:`unilabos/compile/<protocol_name>_protocol.py`
```python
import networkx as nx
from typing import List, Dict, Any
def generate_my_new_protocol(
G: nx.DiGraph,
vessel: dict,
time: float = 300.0,
mode: str = "default",
**kwargs,
) -> List[Dict[str, Any]]:
"""将 MyNewProtocol 编译为设备动作序列。
Args:
G: 设备连接图NetworkX节点为设备/容器,边为物理连接
vessel: 目标容器 {"id": "reactor_1"}
time: 操作时间(秒)
mode: 操作模式
Returns:
动作列表,每个元素为:
- dict: 单步动作
- list[dict]: 并行动作
"""
from unilabos.compile.utils.vessel_parser import get_vessel
vessel_id, vessel_data = get_vessel(vessel)
actions = []
# 查找相关设备(通过图的连接关系)
# 生成动作序列
actions.append({
"device_id": "target_device_id",
"action_name": "some_action",
"action_kwargs": {"param": "value"}
})
# 等待
actions.append({
"action_name": "wait",
"action_kwargs": {"time": time}
})
return actions
```
### 动作字典格式
```python
# 单步动作(发给子设备)
{"device_id": "pump_1", "action_name": "set_position", "action_kwargs": {"position": 10.0}}
# 发给工作站自身
{"device_id": "self", "action_name": "my_action", "action_kwargs": {...}}
# 等待
{"action_name": "wait", "action_kwargs": {"time": 5.0}}
# 并行动作(列表嵌套)
[
{"device_id": "pump_1", "action_name": "set_position", "action_kwargs": {"position": 10.0}},
{"device_id": "stirrer_1", "action_name": "start_stir", "action_kwargs": {"stir_speed": 300}}
]
```
### 关于 `vessel` 参数类型
现有协议的 `vessel` 参数类型不统一:
- 新协议趋势:使用 `dict`(如 `{"id": "reactor_1"}`
- 旧协议:使用 `str`(如 `"reactor_1"`
- 兼容写法:`Union[str, dict]`
**建议新协议统一使用 `dict` 类型**,通过 `get_vessel()` 兼容两种输入。
### 公共工具函数(`unilabos/compile/utils/`
| 函数 | 用途 |
|------|------|
| `get_vessel(vessel)` | 解析容器参数为 `(vessel_id, vessel_data)`,兼容 dict 和 str |
| `find_solvent_vessel(G, solvent)` | 根据溶剂名查找容器(精确→命名规则→模糊→液体类型) |
| `find_reagent_vessel(G, reagent)` | 根据试剂名查找容器(支持固体和液体) |
| `find_connected_stirrer(G, vessel)` | 查找与容器相连的搅拌器 |
| `find_solid_dispenser(G)` | 查找固体加样器 |
### 协议内专属查找函数
许多协议在自己的文件内定义了专属的 `find_*` 函数(不在 `utils/` 中)。编写新协议时,优先复用 `utils/` 中的公共函数;如需特殊查找逻辑,在协议文件内部定义即可:
```python
def find_my_special_device(G: nx.DiGraph, vessel: str) -> str:
"""查找与容器相关的特殊设备"""
for node in G.nodes():
if 'my_device_type' in G.nodes[node].get('class', '').lower():
return node
raise ValueError("未找到特殊设备")
```
### 复用已有协议
复杂协议通常组合已有协议:
```python
from unilabos.compile.pump_protocol import generate_pump_protocol_with_rinsing
actions.extend(generate_pump_protocol_with_rinsing(
G, from_vessel=solvent_vessel, to_vessel=vessel, volume=volume
))
```
### 图查询模式
```python
# 查找与容器相连的特定类型设备
for neighbor in G.neighbors(vessel_id):
node_data = G.nodes[neighbor]
if "heater" in node_data.get("class", ""):
heater_id = neighbor
break
# 查找最短路径(泵转移)
path = nx.shortest_path(G, source=from_vessel_id, target=to_vessel_id)
```
---
## 第六步:注册协议生成函数
`unilabos/compile/__init__.py` 中:
1. 顶部添加导入:
```python
from .my_new_protocol import generate_my_new_protocol
```
2.`action_protocol_generators` 字典中添加映射:
```python
action_protocol_generators = {
# ... 已有协议
MyNewProtocol: generate_my_new_protocol,
}
```
---
## 第七步:配置图文件
在工作站的图文件中,将协议名加入 `protocol_type`
```json
{
"id": "my_station",
"class": "workstation",
"config": {
"protocol_type": ["PumpTransferProtocol", "MyNewProtocol"]
}
}
```
---
## 第八步:验证
```bash
# 1. 模块可导入
python -c "from unilabos.messages import MyNewProtocol; print(MyNewProtocol.model_fields)"
# 2. 生成函数可导入
python -c "from unilabos.compile import action_protocol_generators; print(list(action_protocol_generators.keys()))"
# 3. 启动测试(可选)
unilab -g <graph>.json --complete_registry
```
---
## 工作流清单
```
协议接入进度:
- [ ] 1. 确认协议名、参数、涉及设备
- [ ] 2. 创建 .action 文件 (unilabos_msgs/action/<Name>.action)
- [ ] 3. 注册到 CMakeLists.txt
- [ ] 4. 创建 Pydantic 模型 (unilabos/messages/__init__.py) + 更新 __all__
- [ ] 5. 实现生成函数 (unilabos/compile/<name>_protocol.py)
- [ ] 6. 注册到 compile/__init__.py
- [ ] 7. 配置图文件 protocol_type
- [ ] 8. 验证
```
---
## 高级模式
实现复杂协议时,详见 [reference.md](reference.md)协议运行时数据流、mock graph 测试模式、单位解析工具(`unit_parser.py`)、复杂协议组合模式(以 dissolve 为例)。
---
## 现有协议速查
| 协议 | Pydantic 类 | 生成函数 | 核心参数 |
|------|-------------|---------|---------|
| 泵转移 | `PumpTransferProtocol` | `generate_pump_protocol_with_rinsing` | `from_vessel, to_vessel, volume` |
| 简单转移 | `TransferProtocol` | `generate_pump_protocol` | `from_vessel, to_vessel, volume` |
| 加样 | `AddProtocol` | `generate_add_protocol` | `vessel, reagent, volume` |
| 过滤 | `FilterProtocol` | `generate_filter_protocol` | `vessel, filtrate_vessel` |
| 溶解 | `DissolveProtocol` | `generate_dissolve_protocol` | `vessel, solvent, volume` |
| 加热/冷却 | `HeatChillProtocol` | `generate_heat_chill_protocol` | `vessel, temp, time` |
| 搅拌 | `StirProtocol` | `generate_stir_protocol` | `vessel, time` |
| 分离 | `SeparateProtocol` | `generate_separate_protocol` | `from_vessel, separation_vessel, solvent` |
| 蒸发 | `EvaporateProtocol` | `generate_evaporate_protocol` | `vessel, pressure, temp, time` |
| 清洗 | `CleanProtocol` | `generate_clean_protocol` | `vessel, solvent, volume` |
| 离心 | `CentrifugeProtocol` | `generate_centrifuge_protocol` | `vessel, speed, time` |
| 抽气充气 | `EvacuateAndRefillProtocol` | `generate_evacuateandrefill_protocol` | `vessel, gas` |

View File

@@ -0,0 +1,207 @@
# 协议高级参考
本文件是 SKILL.md 的补充包含协议运行时数据流、测试模式、单位解析工具和复杂协议组合模式。Agent 在需要实现这些功能时按需阅读。
---
## 1. 协议运行时数据流
从图文件到协议执行的完整链路:
```
实验图 JSON
↓ graphio.read_node_link_json()
physical_setup_graph (NetworkX DiGraph)
↓ ROS2WorkstationNode._setup_protocol_names(protocol_type)
为每个 protocol_name 创建 ActionServer
↓ 收到 Action Goal
_create_protocol_execute_callback()
↓ convert_from_ros_msg_with_mapping(goal, mapping)
protocol_kwargs (Python dict)
↓ 向 Host 查询 Resource 类型参数的当前状态
protocol_kwargs 更新vessel 带上 children、data 等)
↓ protocol_steps_generator(G=physical_setup_graph, **protocol_kwargs)
List[Dict] 动作序列
↓ 逐步 execute_single_action / 并行 create_task
子设备 ActionClient 执行
```
### `_setup_protocol_names` 核心逻辑
```python
def _setup_protocol_names(self, protocol_type):
if isinstance(protocol_type, str):
self.protocol_names = [p.strip() for p in protocol_type.split(",")]
else:
self.protocol_names = protocol_type
self.protocol_action_mappings = {}
for protocol_name in self.protocol_names:
protocol_type = globals()[protocol_name] # 从 messages 模块取 Pydantic 类
self.protocol_action_mappings[protocol_name] = get_action_type(protocol_type)
```
### `_create_protocol_execute_callback` 关键步骤
1. `convert_from_ros_msg_with_mapping(goal, action_value_mapping["goal"])` — ROS Goal → Python dict
2.`Resource` 类型字段,通过 `resource_get` Service 查询 Host 的最新资源状态
3. `protocol_steps_generator(G=physical_setup_graph, **protocol_kwargs)` — 调用编译函数
4. 遍历 steps`dict` 串行执行,`list` 并行执行
5. `execute_single_action` 通过 `_action_clients[device_id]` 向子设备发送 Action Goal
6. 执行完毕后通过 `resource_update` Service 更新资源状态
---
## 2. 测试模式
### 2.1 协议文件内测试函数
许多协议文件末尾有 `test_*` 函数,主要测试参数解析工具:
```python
def test_dissolve_protocol():
"""测试溶解协议的各种参数解析"""
volumes = ["10 mL", "?", 10.0, "1 L", "500 μL"]
for vol in volumes:
result = parse_volume_input(vol)
print(f"体积解析: {vol}{result}mL")
masses = ["2.9 g", "?", 2.5, "500 mg"]
for mass in masses:
result = parse_mass_input(mass)
print(f"质量解析: {mass}{result}g")
```
### 2.2 使用 mock graph 测试协议生成器
推荐的端到端测试模式:
```python
import pytest
import networkx as nx
from unilabos.compile.stir_protocol import generate_stir_protocol
@pytest.fixture
def topology_graph():
"""创建测试拓扑图"""
G = nx.DiGraph()
G.add_node("flask_1", **{"class": "flask", "type": "container"})
G.add_node("stirrer_1", **{"class": "virtual_stirrer", "type": "device"})
G.add_edge("stirrer_1", "flask_1")
return G
def test_generate_stir_protocol(topology_graph):
"""测试搅拌协议生成"""
actions = generate_stir_protocol(
G=topology_graph,
vessel="flask_1",
time="5 min",
stir_speed=300.0
)
assert len(actions) >= 1
assert actions[0]["device_id"] == "stirrer_1"
```
**要点:**
-`nx.DiGraph()` 构建最小拓扑
- `add_node(id, **attrs)` 设置 `class``type``data`
- `add_edge(src, dst)` 建立物理连接
- 协议内的 `find_*` 函数依赖这些节点和边
---
## 3. 单位解析工具
路径:`unilabos/compile/utils/unit_parser.py`
| 函数 | 输入 | 返回 | 默认值 |
|------|------|------|--------|
| `parse_volume_input(input, default_unit)` | `"100 mL"`, `"2.5 L"`, `"500 μL"`, `10.0`, `"?"` | mL (float) | 50.0 |
| `parse_mass_input(input)` | `"19.3 g"`, `"500 mg"`, `2.5`, `"?"` | g (float) | 1.0 |
| `parse_time_input(input)` | `"30 min"`, `"1 h"`, `"300"`, `60.0`, `"?"` | 秒 (float) | 60.0 |
支持的单位:
- **体积**: mL, L, μL/uL, milliliter, liter, microliter
- **质量**: g, mg, kg, gram, milligram, kilogram
- **时间**: s/sec/second, min/minute, h/hr/hour, d/day
特殊值 `"?"``"unknown"``"tbd"` 返回默认值。
---
## 4. 复杂协议组合模式
`dissolve_protocol` 为例,展示如何组合多个子操作:
### 整体流程
```
1. 解析参数 (parse_volume_input, parse_mass_input, parse_time_input)
2. 设备发现 (find_connected_heatchill, find_connected_stirrer, find_solid_dispenser)
3. 判断溶解类型 (液体 vs 固体)
4. 组合动作序列:
a. heat_chill_start / start_stir (启动加热/搅拌)
b. wait (等待温度稳定)
c. pump_protocol_with_rinsing (液体转移, 通过 extend 拼接)
或 add_solid (固体加样)
d. heat_chill / stir / wait (溶解等待)
e. heat_chill_stop (停止加热)
```
### 关键代码模式
**设备发现 → 条件组合:**
```python
heatchill_id = find_connected_heatchill(G, vessel_id)
stirrer_id = find_connected_stirrer(G, vessel_id)
solid_dispenser_id = find_solid_dispenser(G)
actions = []
# 启动阶段
if heatchill_id and temp > 25.0:
actions.append({
"device_id": heatchill_id,
"action_name": "heat_chill_start",
"action_kwargs": {"vessel": {"id": vessel_id}, "temp": temp}
})
actions.append({"action_name": "wait", "action_kwargs": {"time": 30}})
elif stirrer_id:
actions.append({
"device_id": stirrer_id,
"action_name": "start_stir",
"action_kwargs": {"vessel": {"id": vessel_id}, "stir_speed": stir_speed}
})
# 转移阶段(复用已有协议)
pump_actions = generate_pump_protocol_with_rinsing(
G=G, from_vessel=solvent_vessel, to_vessel=vessel_id, volume=volume
)
actions.extend(pump_actions)
# 等待阶段
if heatchill_id:
actions.append({
"device_id": heatchill_id,
"action_name": "heat_chill",
"action_kwargs": {"vessel": {"id": vessel_id}, "temp": temp, "time": time}
})
else:
actions.append({"action_name": "wait", "action_kwargs": {"time": time}})
```
---
## 5. 关键路径
| 内容 | 路径 |
|------|------|
| 协议执行回调 | `unilabos/ros/nodes/presets/workstation.py` |
| ROS 消息映射 | `unilabos/ros/msgs/message_converter.py` |
| 物理拓扑图 | `unilabos/resources/graphio.py` (`physical_setup_graph`) |
| 单位解析 | `unilabos/compile/utils/unit_parser.py` |
| 容器解析 | `unilabos/compile/utils/vessel_parser.py` |
| 溶解协议(组合示例) | `unilabos/compile/dissolve_protocol.py` |

View File

@@ -0,0 +1,371 @@
---
name: add-resource
description: Guide for adding new resources (materials, bottles, carriers, decks, warehouses) to Uni-Lab-OS (添加新物料/资源). Covers Bottle, Carrier, Deck, WareHouse definitions and registry YAML. Use when the user wants to add resources, define materials, create a deck layout, add bottles/carriers/plates, or mentions 物料/资源/resource/bottle/carrier/deck/plate/warehouse.
---
# 添加新物料资源
Uni-Lab-OS 的资源体系基于 PyLabRobot通过扩展实现 Bottle、Carrier、WareHouse、Deck 等实验室物料管理。
---
## 第一步:确认资源类型
向用户确认需要添加的资源类型:
| 类型 | 基类 | 用途 | 示例 |
|------|------|------|------|
| **Bottle** | `Well` (PyLabRobot) | 单个容器(瓶、小瓶、烧杯、反应器) | 试剂瓶、粉末瓶 |
| **BottleCarrier** | `ItemizedCarrier` | 多槽位载架(放多个 Bottle | 6 位试剂架、枪头盒 |
| **WareHouse** | `ItemizedCarrier` | 堆栈/仓库(放多个 Carrier | 4x4 堆栈 |
| **Deck** | `Deck` (PyLabRobot) | 工作站台面(放多个 WareHouse | 反应站 Deck |
**层级关系:** `Deck``WareHouse``BottleCarrier``Bottle`
还需确认:
- 资源所属的项目/场景(如 bioyond、battery、通用
- 尺寸参数(直径、高度、最大容积等)
- 布局参数(行列数、间距等)
---
## 第二步:创建资源定义
### 文件位置
```
unilabos/resources/
├── <project>/ # 按项目分组
│ ├── bottles.py # Bottle 工厂函数
│ ├── bottle_carriers.py # Carrier 工厂函数
│ ├── warehouses.py # WareHouse 工厂函数
│ └── decks.py # Deck 类定义
├── itemized_carrier.py # Bottle, BottleCarrier, ItemizedCarrier 基类
├── warehouse.py # WareHouse 基类
└── container.py # 通用容器
```
### 2A. 添加 Bottle工厂函数
```python
from unilabos.resources.itemized_carrier import Bottle
def My_Reagent_Bottle(
name: str,
diameter: float = 70.0, # 瓶体直径 (mm)
height: float = 120.0, # 瓶体高度 (mm)
max_volume: float = 500000.0, # 最大容积 (μL)
barcode: str = None,
) -> Bottle:
"""创建试剂瓶"""
return Bottle(
name=name,
diameter=diameter,
height=height,
max_volume=max_volume,
barcode=barcode,
model="My_Reagent_Bottle", # 唯一标识,用于注册表和物料映射
)
```
**Bottle 参数:**
- `name`: 实例名称(运行时唯一)
- `diameter`: 瓶体直径 (mm)
- `height`: 瓶体高度 (mm)
- `max_volume`: 最大容积 (**μL**注意单位500mL = 500000)
- `barcode`: 条形码(可选)
- `model`: 模型标识,与注册表 key 一致
### 2B. 添加 BottleCarrier工厂函数
```python
from pylabrobot.resources import ResourceHolder
from pylabrobot.resources.carrier import create_ordered_items_2d
from unilabos.resources.itemized_carrier import BottleCarrier
def My_6SlotCarrier(name: str) -> BottleCarrier:
"""创建 3x2 六槽位载架"""
sites = create_ordered_items_2d(
klass=ResourceHolder,
num_items_x=3, # 列数
num_items_y=2, # 行数
dx=10.0, # X 起始偏移
dy=10.0, # Y 起始偏移
dz=5.0, # Z 偏移
item_dx=42.0, # X 间距
item_dy=35.0, # Y 间距
size_x=20.0, # 槽位宽
size_y=20.0, # 槽位深
size_z=50.0, # 槽位高
)
carrier = BottleCarrier(
name=name,
size_x=146.0, # 载架总宽
size_y=80.0, # 载架总深
size_z=55.0, # 载架总高
sites=sites,
model="My_6SlotCarrier",
)
carrier.num_items_x = 3
carrier.num_items_y = 2
carrier.num_items_z = 1
# 预装 Bottle可选
ordering = ["A01", "A02", "A03", "B01", "B02", "B03"]
for i in range(6):
carrier[i] = My_Reagent_Bottle(f"{ordering[i]}")
return carrier
```
### 2C. 添加 WareHouse工厂函数
```python
from unilabos.resources.warehouse import warehouse_factory
def my_warehouse_4x4(name: str) -> "WareHouse":
"""创建 4行x4列 堆栈仓库"""
return warehouse_factory(
name=name,
num_items_x=4, # 列数
num_items_y=4, # 行数
num_items_z=1, # 层数(通常为 1
dx=137.0, # X 起始偏移
dy=96.0, # Y 起始偏移
dz=120.0, # Z 起始偏移
item_dx=137.0, # X 间距
item_dy=125.0, # Y 间距
item_dz=10.0, # Z 间距(多层时用)
resource_size_x=127.0, # 槽位宽
resource_size_y=85.0, # 槽位深
resource_size_z=100.0, # 槽位高
model="my_warehouse_4x4",
)
```
**`warehouse_factory` 参数说明:**
| 参数 | 说明 |
|------|------|
| `num_items_x/y/z` | 列数/行数/层数 |
| `dx, dy, dz` | 第一个槽位的起始坐标偏移 |
| `item_dx, item_dy, item_dz` | 相邻槽位间距 |
| `resource_size_x/y/z` | 单个槽位的物理尺寸 |
| `col_offset` | 列命名偏移(如设 4 则从 A05 开始) |
| `row_offset` | 行命名偏移(如设 5 则从 F 行开始) |
| `layout` | 排序方式:`"col-major"`(列优先,默认)/ `"row-major"`(行优先) |
| `removed_positions` | 要移除的位置索引列表 |
自动生成 `ResourceHolder` 槽位,命名规则为 `A01, B01, C01, D01, A02, ...`(列优先)或 `A01, A02, A03, A04, B01, ...`(行优先)。
### 2D. 添加 Deck类定义
```python
from pylabrobot.resources import Deck, Coordinate
class MyStation_Deck(Deck):
def __init__(
self,
name: str = "MyStation_Deck",
size_x: float = 2700.0,
size_y: float = 1080.0,
size_z: float = 1500.0,
category: str = "deck",
setup: bool = False,
) -> None:
super().__init__(name=name, size_x=size_x, size_y=size_y, size_z=size_z)
if setup:
self.setup()
def setup(self) -> None:
self.warehouses = {
"仓库A": my_warehouse_4x4("仓库A"),
"仓库B": my_warehouse_4x4("仓库B"),
}
self.warehouse_locations = {
"仓库A": Coordinate(-200.0, 400.0, 0.0),
"仓库B": Coordinate(2350.0, 400.0, 0.0),
}
for wh_name, wh in self.warehouses.items():
self.assign_child_resource(wh, location=self.warehouse_locations[wh_name])
```
**Deck 要点:**
- 继承 `pylabrobot.resources.Deck`
- `setup()` 创建 WareHouse 并通过 `assign_child_resource` 放置到指定坐标
- `setup` 参数控制是否在构造时自动调用 `setup()`(图文件中通过 `config.setup: true` 触发)
---
## 第三步:创建注册表 YAML
路径:`unilabos/registry/resources/<project>/<type>.yaml`
### Bottle 注册
```yaml
My_Reagent_Bottle:
category:
- bottles
class:
module: unilabos.resources.my_project.bottles:My_Reagent_Bottle
type: pylabrobot
description: 我的试剂瓶
handles: []
icon: ''
init_param_schema: {}
version: 1.0.0
```
### Carrier 注册
```yaml
My_6SlotCarrier:
category:
- bottle_carriers
class:
module: unilabos.resources.my_project.bottle_carriers:My_6SlotCarrier
type: pylabrobot
handles: []
icon: ''
init_param_schema: {}
version: 1.0.0
```
### Deck 注册
```yaml
MyStation_Deck:
category:
- deck
class:
module: unilabos.resources.my_project.decks:MyStation_Deck
type: pylabrobot
description: 我的工作站 Deck
handles: []
icon: ''
init_param_schema: {}
registry_type: resource
version: 1.0.0
```
**注册表规则:**
- `class.module` 格式为 `python.module.path:ClassName_or_FunctionName`
- `class.type` 固定为 `pylabrobot`
- Key`My_Reagent_Bottle`)必须与工厂函数名 / 类名一致
- `category` 按类型标注(`bottles`, `bottle_carriers`, `deck` 等)
---
## 第四步:在图文件中引用
### Deck 在工作站中的引用
工作站节点通过 `deck` 字段引用Deck 作为子节点:
```json
{
"id": "my_station",
"children": ["my_deck"],
"deck": {
"data": {
"_resource_child_name": "my_deck",
"_resource_type": "unilabos.resources.my_project.decks:MyStation_Deck"
}
}
},
{
"id": "my_deck",
"parent": "my_station",
"type": "deck",
"class": "MyStation_Deck",
"config": {"type": "MyStation_Deck", "setup": true}
}
```
### 物料类型映射(外部系统对接时)
如果工作站需要与外部系统同步物料,在 config 中配置 `material_type_mappings`
```json
"material_type_mappings": {
"My_Reagent_Bottle": ["试剂瓶", "external-type-uuid"],
"My_6SlotCarrier": ["六槽载架", "external-type-uuid"]
}
```
---
## 第五步:注册 PLR 扩展(如需要)
如果添加了新的 Deck 类,需要在 `unilabos/resources/plr_additional_res_reg.py` 中导入,使 `find_subclass` 能发现它:
```python
def register():
from unilabos.resources.my_project.decks import MyStation_Deck
```
---
## 第六步:验证
```bash
# 1. 资源可导入
python -c "from unilabos.resources.my_project.bottles import My_Reagent_Bottle; print(My_Reagent_Bottle('test'))"
# 2. Deck 可创建
python -c "
from unilabos.resources.my_project.decks import MyStation_Deck
d = MyStation_Deck('test', setup=True)
print(d.children)
"
# 3. 启动测试
unilab -g <graph>.json --complete_registry
```
---
## 工作流清单
```
资源接入进度:
- [ ] 1. 确定资源类型Bottle / Carrier / WareHouse / Deck
- [ ] 2. 创建资源定义(工厂函数/类)
- [ ] 3. 创建注册表 YAML (unilabos/registry/resources/<project>/<type>.yaml)
- [ ] 4. 在图文件中引用(如需要)
- [ ] 5. 注册 PLR 扩展Deck 类需要)
- [ ] 6. 验证
```
---
## 高级模式
实现复杂资源系统时,详见 [reference.md](reference.md):类继承体系完整图、序列化/反序列化流程、Bioyond 物料双向同步、非瓶类资源ElectrodeSheet / Magazine、仓库工厂 layout 模式。
---
## 现有资源参考
| 项目 | Bottles | Carriers | WareHouses | Decks |
|------|---------|----------|------------|-------|
| bioyond | `bioyond/bottles.py` | `bioyond/bottle_carriers.py` | `bioyond/warehouses.py`, `YB_warehouses.py` | `bioyond/decks.py` |
| battery | — | `battery/bottle_carriers.py` | — | — |
| 通用 | — | — | `warehouse.py` | — |
### 关键路径
| 内容 | 路径 |
|------|------|
| Bottle/Carrier 基类 | `unilabos/resources/itemized_carrier.py` |
| WareHouse 基类 + 工厂 | `unilabos/resources/warehouse.py` |
| PLR 注册 | `unilabos/resources/plr_additional_res_reg.py` |
| 资源注册表 | `unilabos/registry/resources/` |
| 图文件加载 | `unilabos/resources/graphio.py` |
| 资源跟踪器 | `unilabos/resources/resource_tracker.py` |

View File

@@ -0,0 +1,292 @@
# 资源高级参考
本文件是 SKILL.md 的补充,包含类继承体系、序列化/反序列化、Bioyond 物料同步、非瓶类资源和仓库工厂模式。Agent 在需要实现这些功能时按需阅读。
---
## 1. 类继承体系
```
PyLabRobot
├── Resource (PLR 基类)
│ ├── Well
│ │ └── Bottle (unilabos) → 瓶/小瓶/烧杯/反应器
│ ├── Deck
│ │ └── 自定义 Deck 类 (unilabos) → 工作站台面
│ ├── ResourceHolder → 槽位占位符
│ └── Container
│ └── Battery (unilabos) → 组装好的电池
├── ItemizedCarrier (unilabos, 继承 Resource)
│ ├── BottleCarrier (unilabos) → 瓶载架
│ └── WareHouse (unilabos) → 堆栈仓库
├── ItemizedResource (PLR)
│ └── MagazineHolder (unilabos) → 子弹夹载架
└── ResourceStack (PLR)
└── Magazine (unilabos) → 子弹夹洞位
```
### Bottle 类细节
```python
class Bottle(Well):
def __init__(self, name, diameter, height, max_volume,
size_x=0.0, size_y=0.0, size_z=0.0,
barcode=None, category="container", model=None, **kwargs):
super().__init__(
name=name,
size_x=diameter, # PLR 用 diameter 作为 size_x/size_y
size_y=diameter,
size_z=height, # PLR 用 height 作为 size_z
max_volume=max_volume,
category=category,
model=model,
bottom_type="flat",
cross_section_type="circle"
)
```
注意 `size_x = size_y = diameter``size_z = height`
### ItemizedCarrier 核心方法
| 方法 | 说明 |
|------|------|
| `__getitem__(identifier)` | 通过索引或 Excel 标识(如 `"A01"`)访问槽位 |
| `__setitem__(identifier, resource)` | 向槽位放入资源 |
| `get_child_identifier(child)` | 获取子资源的标识符 |
| `capacity` | 总槽位数 |
| `sites` | 所有槽位字典 |
---
## 2. 序列化与反序列化
### PLR ↔ UniLab 转换
| 函数 | 位置 | 方向 |
|------|------|------|
| `ResourceTreeSet.from_plr_resources(resources)` | `resource_tracker.py` | PLR → UniLab |
| `ResourceTreeSet.to_plr_resources()` | `resource_tracker.py` | UniLab → PLR |
### `from_plr_resources` 流程
```
PLR Resource
↓ build_uuid_mapping (递归生成 UUID)
↓ resource.serialize() → dict
↓ resource.serialize_all_state() → states
↓ resource_plr_inner (递归构建 ResourceDictInstance)
ResourceTreeSet
```
关键:每个 PLR 资源通过 `unilabos_uuid` 属性携带 UUID`unilabos_extra` 携带扩展数据(如 `class` 名)。
### `to_plr_resources` 流程
```
ResourceTreeSet
↓ collect_node_data (收集 UUID、状态、扩展数据)
↓ node_to_plr_dict (转为 PLR 字典格式)
↓ find_subclass(type_name, PLRResource) (查找 PLR 子类)
↓ sub_cls.deserialize(plr_dict) (反序列化)
↓ loop_set_uuid, loop_set_extra (递归设置 UUID 和扩展)
PLR Resource
```
### Bottle 序列化
```python
class Bottle(Well):
def serialize(self) -> dict:
data = super().serialize()
return {**data, "diameter": self.diameter, "height": self.height}
@classmethod
def deserialize(cls, data: dict, allow_marshal=False):
barcode_data = data.pop("barcode", None)
instance = super().deserialize(data, allow_marshal=allow_marshal)
if barcode_data and isinstance(barcode_data, str):
instance.barcode = barcode_data
return instance
```
---
## 3. Bioyond 物料同步
### 双向转换函数
| 函数 | 位置 | 方向 |
|------|------|------|
| `resource_bioyond_to_plr(materials, type_mapping, deck)` | `graphio.py` | Bioyond → PLR |
| `resource_plr_to_bioyond(resources, type_mapping, warehouse_mapping)` | `graphio.py` | PLR → Bioyond |
### `resource_bioyond_to_plr` 流程
```
Bioyond 物料列表
↓ reverse_type_mapping: {typeName → (model, UUID)}
↓ 对每个物料:
typeName → 查映射 → model (如 "BIOYOND_PolymerStation_Reactor")
initialize_resource({"name": unique_name, "class": model})
↓ 设置 unilabos_extra (material_bioyond_id, material_bioyond_name 等)
↓ 处理 detail (子物料/坐标)
↓ 按 locationName 放入 deck.warehouses 对应槽位
PLR 资源列表
```
### `resource_plr_to_bioyond` 流程
```
PLR 资源列表
↓ 遍历每个资源:
载架(capacity > 1): 生成 details 子物料 + 坐标
单瓶: 直接映射
↓ type_mapping 查找 typeId
↓ warehouse_mapping 查找位置 UUID
↓ 组装 Bioyond 格式 (name, typeName, typeId, quantity, Parameters, locations)
Bioyond 物料列表
```
### BioyondResourceSynchronizer
工作站通过 `ResourceSynchronizer` 自动同步物料:
```python
class BioyondResourceSynchronizer(ResourceSynchronizer):
def sync_from_external(self) -> bool:
all_data = []
all_data.extend(api_client.stock_material('{"typeMode": 0}')) # 耗材
all_data.extend(api_client.stock_material('{"typeMode": 1}')) # 样品
all_data.extend(api_client.stock_material('{"typeMode": 2}')) # 试剂
unilab_resources = resource_bioyond_to_plr(
all_data,
type_mapping=self.workstation.bioyond_config["material_type_mappings"],
deck=self.workstation.deck
)
# 更新 deck 上的资源
```
---
## 4. 非瓶类资源
### ElectrodeSheet极片
路径:`unilabos/resources/battery/electrode_sheet.py`
```python
class ElectrodeSheet(ResourcePLR):
"""片状材料(极片、隔膜、弹片、垫片等)"""
_unilabos_state = {
"diameter": 0.0,
"thickness": 0.0,
"mass": 0.0,
"material_type": "",
"color": "",
"info": "",
}
```
工厂函数:`PositiveCan`, `PositiveElectrode`, `NegativeCan`, `NegativeElectrode`, `SpringWasher`, `FlatWasher`, `AluminumFoil`
### Battery电池
```python
class Battery(Container):
"""组装好的电池"""
_unilabos_state = {
"color": "",
"electrolyte_name": "",
"open_circuit_voltage": 0.0,
}
```
### Magazine / MagazineHolder子弹夹
```python
class Magazine(ResourceStack):
"""子弹夹洞位,可堆叠 ElectrodeSheet"""
# direction, max_sheets
class MagazineHolder(ItemizedResource):
"""多洞位子弹夹"""
# hole_diameter, hole_depth, max_sheets_per_hole
```
工厂函数 `magazine_factory()``create_homogeneous_resources` 生成洞位,可选预填 `ElectrodeSheet``Battery`
---
## 5. 仓库工厂模式参考
### 实际 warehouse 工厂函数示例
```python
# 行优先 4x4 仓库
def bioyond_warehouse_1x4x4(name: str) -> WareHouse:
return warehouse_factory(
name=name,
num_items_x=4, num_items_y=4, num_items_z=1,
dx=10.0, dy=10.0, dz=10.0,
item_dx=147.0, item_dy=106.0, item_dz=130.0,
layout="row-major", # A01,A02,A03,A04, B01,...
)
# 右侧 4x4 仓库(列名偏移)
def bioyond_warehouse_1x4x4_right(name: str) -> WareHouse:
return warehouse_factory(
name=name,
num_items_x=4, num_items_y=4, num_items_z=1,
dx=10.0, dy=10.0, dz=10.0,
item_dx=147.0, item_dy=106.0, item_dz=130.0,
col_offset=4, # A05,A06,A07,A08
layout="row-major",
)
# 竖向仓库(站内试剂存放)
def bioyond_warehouse_reagent_storage(name: str) -> WareHouse:
return warehouse_factory(
name=name,
num_items_x=1, num_items_y=2, num_items_z=1,
dx=10.0, dy=10.0, dz=10.0,
item_dx=147.0, item_dy=106.0, item_dz=130.0,
layout="vertical-col-major",
)
# 行偏移F 行开始)
def bioyond_warehouse_5x3x1(name: str, row_offset: int = 0) -> WareHouse:
return warehouse_factory(
name=name,
num_items_x=3, num_items_y=5, num_items_z=1,
dx=10.0, dy=10.0, dz=10.0,
item_dx=159.0, item_dy=183.0, item_dz=130.0,
row_offset=row_offset, # 0→A行起5→F行起
layout="row-major",
)
```
### layout 类型说明
| layout | 命名顺序 | 适用场景 |
|--------|---------|---------|
| `col-major` (默认) | A01,B01,C01,D01, A02,B02,... | 列优先,标准堆栈 |
| `row-major` | A01,A02,A03,A04, B01,B02,... | 行优先Bioyond 前端展示 |
| `vertical-col-major` | 竖向排列,标签从底部开始 | 竖向仓库(试剂存放、测密度) |
---
## 6. 关键路径
| 内容 | 路径 |
|------|------|
| Bottle/Carrier 基类 | `unilabos/resources/itemized_carrier.py` |
| WareHouse 类 + 工厂 | `unilabos/resources/warehouse.py` |
| ResourceTreeSet 转换 | `unilabos/resources/resource_tracker.py` |
| Bioyond 物料转换 | `unilabos/resources/graphio.py` |
| Bioyond 仓库定义 | `unilabos/resources/bioyond/warehouses.py` |
| 电池资源 | `unilabos/resources/battery/` |
| PLR 注册 | `unilabos/resources/plr_additional_res_reg.py` |

View File

@@ -0,0 +1,500 @@
---
name: add-workstation
description: Guide for adding new workstations to Uni-Lab-OS (接入新工作站). Walks through workstation type selection, sub-device composition, external system integration, driver creation, registry YAML, deck setup, and graph file configuration. Use when the user wants to add/integrate a new workstation, create a workstation driver, configure a station with sub-devices, set up deck and materials, or mentions 工作站/工站/station/workstation.
---
# Uni-Lab-OS 工作站接入指南
工作站workstation是组合多个子设备的大型设备拥有独立的物料管理系统PLR Deck和工作流引擎。本指南覆盖从需求分析到验证的全流程。
> **前置知识**:工作站接入基于 `docs/ai_guides/add_device.md` 的通用设备接入框架,但有显著差异。阅读本指南前无需先读通用指南。
## 第一步:确定工作站类型
向用户确认以下信息:
**Q1: 工作站的业务场景?**
| 类型 | 基类 | 适用场景 | 示例 |
|------|------|----------|------|
| **Protocol 工作站** | `ProtocolNode` | 标准化学操作协议(过滤、转移、加热等) | FilterProtocolStation |
| **外部系统工作站** | `WorkstationBase` | 与外部 LIMS/MES 系统对接,有专属 API | BioyondStation |
| **硬件控制工作站** | `WorkstationBase` | 直接控制 PLC/硬件,无外部系统 | CoinCellAssembly |
**Q2: 工作站英文名称?**(如 `my_reaction_station`
**Q3: 与外部系统的交互方式?**
| 方式 | 适用场景 | 需要的配置 |
|------|----------|-----------|
| 无外部系统 | Protocol 工作站、纯硬件控制 | 无 |
| HTTP API | LIMS/MES 系统(如 Bioyond | `api_host`, `api_key` |
| Modbus TCP | PLC 控制 | `address`, `port` |
| OPC UA | 工业设备 | `url` |
**Q4: 子设备组成?**
- 列出所有子设备(如反应器、泵、阀、传感器等)
- 哪些是已有设备类型?哪些需要新增?
- 子设备之间的硬件代理关系(如泵通过串口设备通信)
**Q5: 物料管理需求?**
- 是否需要 Deck物料面板
- 物料类型plate、tip_rack、bottle 等)
- 是否需要与外部物料系统同步?
---
## 第二步:理解工作站架构
工作站与普通设备的核心差异:
| 维度 | 普通设备 | 工作站 |
|------|---------|--------|
| 基类 | 无(纯 Python 类) | `WorkstationBase``ProtocolNode` |
| ROS 节点 | `BaseROS2DeviceNode` | `ROS2WorkstationNode` |
| 状态管理 | `self.data` 字典 | 通常不用 `self.data`,用 `@property` 直接访问 |
| 子设备 | 无 | `children` 列表,通过 `self._children` 访问 |
| 物料 | 无 | `self.deck`PLR Deck |
| 图文件角色 | `parent: null``parent: "<station>"` | `parent: null`,含 `children``deck` |
### 继承体系
`WorkstationBase` (ABC) → `ProtocolNode` (通用协议) / `BioyondWorkstation` (→ ReactionStation, DispensingStation) / `CoinCellAssemblyWorkstation` (硬件控制)
### ROS 层
`ROS2WorkstationNode` 额外负责:初始化 children 子设备节点、为子设备创建 ActionClient、配置硬件代理、为 protocol_type 创建协议 ActionServer。
---
## 第三步:创建驱动文件
文件路径:`unilabos/devices/workstation/<station_name>/<station_name>.py`
### 模板 A基于外部系统的工作站
适用于与 LIMS/MES 等外部系统对接的场景。
```python
import logging
from typing import Dict, Any, Optional, List
from pylabrobot.resources import Deck
from unilabos.devices.workstation.workstation_base import WorkstationBase
try:
from unilabos.ros.nodes.presets.workstation import ROS2WorkstationNode
except ImportError:
ROS2WorkstationNode = None
class MyWorkstation(WorkstationBase):
"""工作站描述"""
_ros_node: "ROS2WorkstationNode"
def __init__(
self,
config: dict = None,
deck: Optional[Deck] = None,
protocol_type: list = None,
**kwargs,
):
super().__init__(deck=deck, **kwargs)
self.config = config or {}
self.logger = logging.getLogger(f"MyWorkstation")
# 外部系统连接配置
self.api_host = self.config.get("api_host", "")
self.api_key = self.config.get("api_key", "")
# 工作站业务状态(不同于 self.data 模式)
self._status = "Idle"
def post_init(self, ros_node: "ROS2WorkstationNode") -> None:
super().post_init(ros_node)
# 在这里启动后台服务、连接监控等
# ============ 子设备访问 ============
def _get_child_device(self, device_id: str):
"""通过 ID 获取子设备节点"""
return self._children.get(device_id)
# ============ 动作方法 ============
async def scheduler_start(self, **kwargs) -> Dict[str, Any]:
"""启动调度器"""
return {"success": True}
async def create_order(self, json_str: str, **kwargs) -> Dict[str, Any]:
"""创建工单"""
return {"success": True}
# ============ 属性 ============
@property
def workflow_sequence(self) -> str:
return "[]"
@property
def material_info(self) -> str:
return "{}"
```
### 模板 B基于硬件控制的工作站
适用于直接与 PLC/硬件通信的场景。
```python
import logging
from typing import Dict, Any, Optional
from pylabrobot.resources import Deck
from unilabos.devices.workstation.workstation_base import WorkstationBase
try:
from unilabos.ros.nodes.presets.workstation import ROS2WorkstationNode
except ImportError:
ROS2WorkstationNode = None
class MyHardwareWorkstation(WorkstationBase):
"""硬件控制工作站"""
_ros_node: "ROS2WorkstationNode"
def __init__(
self,
config: dict = None,
deck: Optional[Deck] = None,
address: str = "192.168.1.100",
port: str = "502",
debug_mode: bool = False,
*args,
**kwargs,
):
super().__init__(deck=deck, *args, **kwargs)
self.config = config or {}
self.address = address
self.port = int(port)
self.debug_mode = debug_mode
self.logger = logging.getLogger("MyHardwareWorkstation")
# 初始化通信客户端
if not debug_mode:
from unilabos.device_comms.modbus_plc.client import ModbusTcpClient
self.client = ModbusTcpClient(host=self.address, port=self.port)
else:
self.client = None
def post_init(self, ros_node: "ROS2WorkstationNode") -> None:
super().post_init(ros_node)
# ============ 硬件读写 ============
def _read_register(self, name: str):
"""读取 Modbus 寄存器"""
if self.debug_mode:
return 0
# 实际读取逻辑
pass
# ============ 动作方法 ============
async def start_process(self, **kwargs) -> Dict[str, Any]:
"""启动加工流程"""
return {"success": True}
async def stop_process(self, **kwargs) -> Dict[str, Any]:
"""停止加工流程"""
return {"success": True}
# ============ 属性(从硬件实时读取)============
@property
def sys_status(self) -> str:
return str(self._read_register("SYS_STATUS"))
```
### 模板 CProtocol 工作站
适用于标准化学操作协议的场景,直接使用 `ProtocolNode`
```python
from typing import List, Optional
from pylabrobot.resources import Resource as PLRResource
from unilabos.devices.workstation.workstation_base import ProtocolNode
class MyProtocolStation(ProtocolNode):
"""Protocol 工作站 — 使用标准化学操作协议"""
def __init__(
self,
protocol_type: List[str],
deck: Optional[PLRResource] = None,
*args,
**kwargs,
):
super().__init__(protocol_type=protocol_type, deck=deck, *args, **kwargs)
```
> Protocol 工作站通常不需要自定义驱动类,直接使用 `ProtocolNode` 并在注册表和图文件中配置 `protocol_type` 即可。
---
## 第四步:创建子设备驱动(如需要)
工作站的子设备本身是独立设备。按 `docs/ai_guides/add_device.md` 的标准流程创建。
子设备的关键约束:
- 在图文件中 `parent` 指向工作站 ID
- 图文件中在工作站的 `children` 数组里列出
- 如需硬件代理,在子设备的 `config.hardware_interface.name` 指向通信设备 ID
---
## 第五步:创建注册表 YAML
路径:`unilabos/registry/devices/<station_name>.yaml`
### 最小配置
```yaml
my_workstation:
category:
- workstation
class:
module: unilabos.devices.workstation.my_station.my_station:MyWorkstation
type: python
```
启动时 `--complete_registry` 自动补全 `status_types``action_value_mappings`
### 完整配置参考
```yaml
my_workstation:
description: "我的工作站"
version: "1.0.0"
category:
- workstation
- my_category
class:
module: unilabos.devices.workstation.my_station.my_station:MyWorkstation
type: python
status_types:
workflow_sequence: String
material_info: String
action_value_mappings:
scheduler_start:
type: UniLabJsonCommandAsync
goal: {}
result:
success: success
create_order:
type: UniLabJsonCommandAsync
goal:
json_str: json_str
result:
success: success
init_param_schema:
config:
type: object
deck:
type: object
protocol_type:
type: array
```
### 子设备注册表
子设备有独立的注册表文件,需要在 `category` 中包含工作站标识:
```yaml
my_reactor:
category:
- reactor
- my_workstation
class:
module: unilabos.devices.workstation.my_station.my_reactor:MyReactor
type: python
```
---
## 第六步:配置 Deck 资源(如需要)
如果工作站有物料管理需求,需要定义 Deck 类。
### 使用已有 Deck 类
查看 `unilabos/resources/` 目录下是否有适用的 Deck 类。
### 创建自定义 Deck
`unilabos/resources/<category>/decks.py` 中定义:
```python
from pylabrobot.resources import Deck
from pylabrobot.resources.coordinate import Coordinate
def MyStation_Deck(name: str = "MyStation_Deck") -> Deck:
deck = Deck(name=name, size_x=2700.0, size_y=1080.0, size_z=1500.0)
# 在 deck 上定义子资源位置carrier、plate 等)
return deck
```
`unilabos/resources/<category>/` 下注册或通过注册表引用。
---
## 第七步:配置图文件
图文件路径:`unilabos/test/experiments/<station_name>.json`
### 完整结构
```json
{
"nodes": [
{
"id": "my_station",
"name": "my_station",
"children": ["my_deck", "sub_device_1", "sub_device_2"],
"parent": null,
"type": "device",
"class": "my_workstation",
"position": {"x": 0, "y": 0, "z": 0},
"config": {
"api_host": "http://192.168.1.100:8080",
"api_key": "YOUR_KEY"
},
"deck": {
"data": {
"_resource_child_name": "my_deck",
"_resource_type": "unilabos.resources.my_module.decks:MyStation_Deck"
}
},
"size_x": 2700.0,
"size_y": 1080.0,
"size_z": 1500.0,
"protocol_type": [],
"data": {}
},
{
"id": "my_deck",
"name": "my_deck",
"children": [],
"parent": "my_station",
"type": "deck",
"class": "MyStation_Deck",
"position": {"x": 0, "y": 0, "z": 0},
"config": {
"type": "MyStation_Deck",
"setup": true,
"rotation": {"x": 0, "y": 0, "z": 0, "type": "Rotation"}
},
"data": {}
},
{
"id": "sub_device_1",
"name": "sub_device_1",
"children": [],
"parent": "my_station",
"type": "device",
"class": "sub_device_registry_name",
"position": {"x": 100, "y": 0, "z": 0},
"config": {},
"data": {}
}
]
}
```
### 图文件规则
| 字段 | 说明 |
|------|------|
| `id` | 节点唯一标识,与 `children` 数组中的引用一致 |
| `children` | 包含 deck ID 和所有子设备 ID |
| `parent` | 工作站节点为 `null`;子设备/deck 指向工作站 ID |
| `type` | 工作站和子设备为 `"device"`deck 为 `"deck"` |
| `class` | 对应注册表中的设备名 |
| `deck.data._resource_child_name` | 必须与 deck 节点的 `id` 一致 |
| `deck.data._resource_type` | Deck 工厂函数的完整 Python 路径 |
| `protocol_type` | Protocol 工作站填入协议名列表;否则为 `[]` |
| `config` | 传入驱动 `__init__``config` 参数 |
---
## 第八步:验证
```bash
# 1. 模块可导入
python -c "from unilabos.devices.workstation.<name>.<name> import <ClassName>"
# 2. 注册表补全
unilab -g <graph>.json --complete_registry
# 3. 启动测试
unilab -g <graph>.json
```
---
## 高级模式
实现外部系统对接型工作站时,详见 [reference.md](reference.md)RPC 客户端、HTTP 回调服务、连接监控、Config 结构模式material_type_mappings / warehouse_mapping / workflow_mappings、ResourceSynchronizer、update_resource、工作流序列、站间物料转移、post_init 完整模式。
---
## 关键规则
1. **`__init__` 必须接受 `deck``**kwargs`** — `WorkstationBase.__init__` 需要 `deck` 参数
2. **通过 `self._children` 访问子设备** — 不要自行维护子设备引用
3. **`post_init` 中启动后台服务** — 不要在 `__init__` 中启动网络连接
4. **异步方法使用 `await self._ros_node.sleep()`** — 禁止 `time.sleep()``asyncio.sleep()`
5. **子设备在图文件中声明** — 不在驱动代码中创建子设备实例
6. **`deck` 配置中的 `_resource_child_name` 必须与 deck 节点 ID 一致**
7. **Protocol 工作站优先使用 `ProtocolNode`** — 不需要自定义类
---
## 工作流清单
```
工作站接入进度:
- [ ] 1. 确定工作站类型Protocol / 外部系统 / 硬件控制)
- [ ] 2. 确认子设备组成和物料需求
- [ ] 3. 创建工作站驱动 unilabos/devices/workstation/<name>/<name>.py
- [ ] 4. 创建子设备驱动(如需要,按 add_device.md 流程)
- [ ] 5. 创建注册表 unilabos/registry/devices/<name>.yaml
- [ ] 6. 创建/选择 Deck 资源类(如需要)
- [ ] 7. 配置图文件 unilabos/test/experiments/<name>.json
- [ ] 8. 验证:可导入 + 注册表补全 + 启动测试
```
---
## 现有工作站参考
| 工作站 | 注册表名 | 驱动类 | 类型 |
|--------|----------|--------|------|
| Protocol 通用 | `workstation` | `ProtocolNode` | Protocol |
| Bioyond 反应站 | `reaction_station.bioyond` | `BioyondReactionStation` | 外部系统 |
| Bioyond 配液站 | `bioyond_dispensing_station` | `BioyondDispensingStation` | 外部系统 |
| 纽扣电池组装 | `coincellassemblyworkstation_device` | `CoinCellAssemblyWorkstation` | 硬件控制 |
### 参考文件路径
- 基类: `unilabos/devices/workstation/workstation_base.py`
- Bioyond 基类: `unilabos/devices/workstation/bioyond_studio/station.py`
- 反应站: `unilabos/devices/workstation/bioyond_studio/reaction_station/reaction_station.py`
- 配液站: `unilabos/devices/workstation/bioyond_studio/dispensing_station/dispensing_station.py`
- 纽扣电池: `unilabos/devices/workstation/coin_cell_assembly/coin_cell_assembly.py`
- ROS 节点: `unilabos/ros/nodes/presets/workstation.py`
- 图文件: `unilabos/test/experiments/reaction_station_bioyond.json`, `dispensing_station_bioyond.json`

View File

@@ -0,0 +1,371 @@
# 工作站高级模式参考
本文件是 SKILL.md 的补充,包含外部系统集成、物料同步、配置结构等高级模式。
Agent 在需要实现这些功能时按需阅读。
---
## 1. 外部系统集成模式
### 1.1 RPC 客户端
与外部 LIMS/MES 系统通信的标准模式。继承 `BaseRequest`,所有接口统一用 POST。
```python
from unilabos.device_comms.rpc import BaseRequest
class MySystemRPC(BaseRequest):
"""外部系统 RPC 客户端"""
def __init__(self, host: str, api_key: str):
super().__init__(host)
self.api_key = api_key
def _request(self, endpoint: str, data: dict = None) -> dict:
return self.post(
url=f"{self.host}/api/{endpoint}",
params={
"apiKey": self.api_key,
"requestTime": self.get_current_time_iso8601(),
"data": data or {},
},
)
def query_status(self) -> dict:
return self._request("status/query")
def create_order(self, order_data: dict) -> dict:
return self._request("order/create", order_data)
```
参考:`unilabos/devices/workstation/bioyond_studio/bioyond_rpc.py``BioyondV1RPC`
### 1.2 HTTP 回调服务
接收外部系统报送的标准模式。使用 `WorkstationHTTPService`,在 `post_init` 中启动。
```python
from unilabos.devices.workstation.workstation_http_service import WorkstationHTTPService
class MyWorkstation(WorkstationBase):
def __init__(self, config=None, deck=None, **kwargs):
super().__init__(deck=deck, **kwargs)
self.config = config or {}
http_cfg = self.config.get("http_service_config", {})
self._http_service_config = {
"host": http_cfg.get("http_service_host", "127.0.0.1"),
"port": http_cfg.get("http_service_port", 8080),
}
self.http_service = None
def post_init(self, ros_node):
super().post_init(ros_node)
self.http_service = WorkstationHTTPService(
workstation_instance=self,
host=self._http_service_config["host"],
port=self._http_service_config["port"],
)
self.http_service.start()
```
**HTTP 服务路由**(固定端点,由 `WorkstationHTTPHandler` 自动分发):
| 端点 | 调用的工作站方法 |
|------|-----------------|
| `/report/step_finish` | `process_step_finish_report(report_request)` |
| `/report/sample_finish` | `process_sample_finish_report(report_request)` |
| `/report/order_finish` | `process_order_finish_report(report_request, used_materials)` |
| `/report/material_change` | `process_material_change_report(report_data)` |
| `/report/error_handling` | `handle_external_error(error_data)` |
实现对应方法即可接收回调:
```python
def process_step_finish_report(self, report_request) -> Dict[str, Any]:
"""处理步骤完成报告"""
step_name = report_request.data.get("stepName")
return {"success": True, "message": f"步骤 {step_name} 已处理"}
def process_order_finish_report(self, report_request, used_materials) -> Dict[str, Any]:
"""处理订单完成报告"""
order_code = report_request.data.get("orderCode")
return {"success": True}
```
参考:`unilabos/devices/workstation/workstation_http_service.py`
### 1.3 连接监控
独立线程周期性检测外部系统连接状态,状态变化时发布 ROS 事件。
```python
class ConnectionMonitor:
def __init__(self, workstation, check_interval=30):
self.workstation = workstation
self.check_interval = check_interval
self._running = False
self._thread = None
def start(self):
self._running = True
self._thread = threading.Thread(target=self._monitor_loop, daemon=True)
self._thread.start()
def _monitor_loop(self):
while self._running:
try:
# 调用外部系统接口检测连接
self.workstation.hardware_interface.ping()
status = "online"
except Exception:
status = "offline"
time.sleep(self.check_interval)
```
参考:`unilabos/devices/workstation/bioyond_studio/station.py``ConnectionMonitor`
---
## 2. Config 结构模式
工作站的 `config` 在图文件中定义,传入 `__init__`。以下是常见字段模式:
### 2.1 外部系统连接
```json
{
"api_host": "http://192.168.1.100:8080",
"api_key": "YOUR_API_KEY"
}
```
### 2.2 HTTP 回调服务
```json
{
"http_service_config": {
"http_service_host": "127.0.0.1",
"http_service_port": 8080
}
}
```
### 2.3 物料类型映射
将 PLR 资源类名映射到外部系统的物料类型(名称 + UUID。用于双向物料转换。
```json
{
"material_type_mappings": {
"PLR_ResourceClassName": ["外部系统显示名", "external-type-uuid"],
"BIOYOND_PolymerStation_Reactor": ["反应器", "3a14233b-902d-0d7b-..."]
}
}
```
### 2.4 仓库映射
将仓库名映射到外部系统的仓库 UUID 和库位 UUID。用于入库/出库操作。
```json
{
"warehouse_mapping": {
"仓库名": {
"uuid": "warehouse-uuid",
"site_uuids": {
"A01": "site-uuid-A01",
"A02": "site-uuid-A02"
}
}
}
}
```
### 2.5 工作流映射
将内部工作流名映射到外部系统的工作流 ID。
```json
{
"workflow_mappings": {
"internal_workflow_name": "external-workflow-uuid"
}
}
```
### 2.6 物料默认参数
```json
{
"material_default_parameters": {
"NMP": {
"unit": "毫升",
"density": "1.03",
"densityUnit": "g/mL",
"description": "N-甲基吡咯烷酮"
}
}
}
```
---
## 3. 资源同步机制
### 3.1 ResourceSynchronizer
抽象基类,用于与外部物料系统双向同步。定义在 `workstation_base.py`
```python
from unilabos.devices.workstation.workstation_base import ResourceSynchronizer
class MyResourceSynchronizer(ResourceSynchronizer):
def __init__(self, workstation, api_client):
super().__init__(workstation)
self.api_client = api_client
def sync_from_external(self) -> bool:
"""从外部系统拉取物料到 deck"""
external_materials = self.api_client.list_materials()
for material in external_materials:
plr_resource = self._convert_to_plr(material)
self.workstation.deck.assign_child_resource(plr_resource, coordinate)
return True
def sync_to_external(self, plr_resource) -> bool:
"""将 deck 中的物料变更推送到外部系统"""
external_data = self._convert_from_plr(plr_resource)
self.api_client.update_material(external_data)
return True
def handle_external_change(self, change_info) -> bool:
"""处理外部系统推送的物料变更"""
return True
```
### 3.2 update_resource — 上传资源树到云端
将 PLR Deck 序列化后通过 ROS 服务上传。典型使用场景:
```python
# 在 post_init 中上传初始 deck
from unilabos.ros.nodes.base_device_node import ROS2DeviceNode
ROS2DeviceNode.run_async_func(
self._ros_node.update_resource, True,
**{"resources": [self.deck]}
)
# 在动作方法中更新特定资源
ROS2DeviceNode.run_async_func(
self._ros_node.update_resource, True,
**{"resources": [updated_plate]}
)
```
---
## 4. 工作流序列管理
工作站通过 `workflow_sequence` 属性管理任务队列JSON 字符串形式)。
```python
class MyWorkstation(WorkstationBase):
def __init__(self, **kwargs):
super().__init__(**kwargs)
self._workflow_sequence = []
@property
def workflow_sequence(self) -> str:
"""返回 JSON 字符串ROS 自动发布"""
import json
return json.dumps(self._workflow_sequence)
async def append_to_workflow_sequence(self, workflow_name: str) -> Dict[str, Any]:
"""添加工作流到队列"""
self._workflow_sequence.append({
"name": workflow_name,
"status": "pending",
"created_at": time.time(),
})
return {"success": True}
async def clear_workflows(self) -> Dict[str, Any]:
"""清空工作流队列"""
self._workflow_sequence = []
return {"success": True}
```
---
## 5. 站间物料转移
工作站之间转移物料的模式。通过 ROS ActionClient 调用目标站的动作。
```python
async def transfer_materials_to_another_station(
self,
target_device_id: str,
transfer_groups: list,
**kwargs,
) -> Dict[str, Any]:
"""将物料转移到另一个工作站"""
target_node = self._children.get(target_device_id)
if not target_node:
# 通过 ROS 节点查找非子设备的目标站
pass
for group in transfer_groups:
resource = self.find_resource_by_name(group["resource_name"])
# 从本站 deck 移除
resource.unassign()
# 调用目标站的接收方法
# ...
return {"success": True, "transferred": len(transfer_groups)}
```
参考:`BioyondDispensingStation.transfer_materials_to_reaction_station`
---
## 6. post_init 完整模式
`post_init` 是工作站初始化的关键阶段,此时 ROS 节点和子设备已就绪。
```python
def post_init(self, ros_node):
super().post_init(ros_node)
# 1. 初始化外部系统客户端(此时 config 已可用)
self.rpc_client = MySystemRPC(
host=self.config.get("api_host"),
api_key=self.config.get("api_key"),
)
self.hardware_interface = self.rpc_client
# 2. 启动连接监控
self.connection_monitor = ConnectionMonitor(self)
self.connection_monitor.start()
# 3. 启动 HTTP 回调服务
if hasattr(self, '_http_service_config'):
self.http_service = WorkstationHTTPService(
workstation_instance=self,
host=self._http_service_config["host"],
port=self._http_service_config["port"],
)
self.http_service.start()
# 4. 上传 deck 到云端
ROS2DeviceNode.run_async_func(
self._ros_node.update_resource, True,
**{"resources": [self.deck]}
)
# 5. 初始化资源同步器(可选)
self.resource_synchronizer = MyResourceSynchronizer(self, self.rpc_client)
```

View File

@@ -0,0 +1,381 @@
---
name: edit-experiment-graph
description: Guide for creating and editing experiment graph files in Uni-Lab-OS (创建/编辑实验组态图). Covers node types, link types, parent-child relationships, deck configuration, and common graph patterns. Use when the user wants to create a graph file, edit an experiment configuration, set up device topology, or mentions 图文件/graph/组态/拓扑/实验图/experiment JSON.
---
# 创建/编辑实验组态图
实验图Graph File定义设备拓扑、物理连接和物料配置。系统启动时加载图文件初始化所有设备和连接关系。
路径:`unilabos/test/experiments/<name>.json`
---
## 第一步:确认需求
向用户确认:
| 信息 | 说明 |
|------|------|
| 场景类型 | 单设备调试 / 多设备联调 / 工作站完整图 |
| 包含的设备 | 设备 ID、注册表 class 名、配置参数 |
| 连接关系 | 物理连接(管道)/ 通信连接(串口)/ 无连接 |
| 父子关系 | 是否有工作站包含子设备 |
| 物料需求 | 是否需要 Deck、容器、试剂瓶 |
---
## 第二步JSON 顶层结构
```json
{
"nodes": [],
"links": []
}
```
> `links` 也可写作 `edges`,加载时两者等效。
---
## 第三步:定义 Nodes
### 节点字段
| 字段 | 类型 | 必需 | 默认值 | 说明 |
|------|------|------|--------|------|
| `id` | string | **是** | — | 节点唯一标识links 和 children 中引用此值 |
| `class` | string | **是** | — | 对应注册表名(设备/资源 YAML 的 key容器可为 `null` |
| `name` | string | 否 | 同 `id` | 显示名称,缺省时自动用 `id` |
| `type` | string | 否 | `"device"` | 节点类型(见下表),缺省时自动设为 `"device"` |
| `children` | string[] | 否 | `[]` | 子节点 ID 列表 |
| `parent` | string\|null | 否 | `null` | 父节点 ID顶层设备为 `null` |
| `position` | object | 否 | `{x:0,y:0,z:0}` | 空间坐标 |
| `config` | object | 否 | `{}` | 传给驱动 `__init__` 的参数 |
| `data` | object | 否 | `{}` | 初始运行状态 |
| `size_x/y/z` | float | 否 | — | 节点物理尺寸(工作站节点常用) |
> 非标准字段(如 `api_host`)会自动移入 `config`。
### 节点类型
| `type` | 用途 | `class` 要求 |
|--------|------|-------------|
| `device` | 设备(默认) | 注册表中的设备名 |
| `deck` | 工作台面 | Deck 工厂函数/类名 |
| `container` | 容器(烧瓶、反应釜) | `null` 或具体容器类名 |
### 设备节点模板
```json
{
"id": "my_device",
"name": "我的设备",
"children": [],
"parent": null,
"type": "device",
"class": "registry_device_name",
"position": {"x": 0, "y": 0, "z": 0},
"config": {
"port": "/dev/ttyUSB0",
"baudrate": 115200
},
"data": {
"status": "Idle"
}
}
```
### 容器节点模板
容器用于协议系统中表示试剂瓶、反应釜等,`class` 通常为 `null`
```json
{
"id": "flask_DMF",
"name": "DMF试剂瓶",
"children": [],
"parent": "my_station",
"type": "container",
"class": null,
"position": {"x": 200, "y": 500, "z": 0},
"config": {"max_volume": 1000.0},
"data": {
"liquid": [{"liquid_type": "DMF", "liquid_volume": 800.0}]
}
}
```
### Deck 节点模板
```json
{
"id": "my_deck",
"name": "my_deck",
"children": [],
"parent": "my_station",
"type": "deck",
"class": "MyStation_Deck",
"position": {"x": 0, "y": 0, "z": 0},
"config": {
"type": "MyStation_Deck",
"setup": true,
"rotation": {"x": 0, "y": 0, "z": 0, "type": "Rotation"}
},
"data": {}
}
```
---
## 第四步:定义 Links
### Link 字段
| 字段 | 类型 | 说明 |
|------|------|------|
| `source` | string | 源节点 ID |
| `target` | string | 目标节点 ID |
| `type` | string | `"physical"` / `"fluid"` / `"communication"` |
| `port` | object | 端口映射 `{source_id: "port_name", target_id: "port_name"}` |
### 物理/流体连接
设备间的管道连接,协议系统用此查找路径:
```json
{
"source": "multiway_valve_1",
"target": "flask_DMF",
"type": "fluid",
"port": {
"multiway_valve_1": "2",
"flask_DMF": "outlet"
}
}
```
### 通信连接
设备间的串口/IO 通信代理,加载时自动将端口信息写入目标设备 config
```json
{
"source": "pump_1",
"target": "serial_device",
"type": "communication",
"port": {
"pump_1": "port",
"serial_device": "port"
}
}
```
---
## 第五步:父子关系与工作站配置
### 工作站 + 子设备
工作站节点的 `children` 列出所有子节点 ID子节点的 `parent` 指向工作站:
```json
{
"id": "my_station",
"children": ["my_deck", "pump_1", "valve_1", "reactor_1"],
"parent": null,
"type": "device",
"class": "workstation",
"config": {
"protocol_type": ["PumpTransferProtocol", "CleanProtocol"]
}
}
```
### 工作站 + Deck 引用
工作站节点中通过 `deck` 字段引用 Deck
```json
{
"id": "my_station",
"children": ["my_deck", "sub_device_1"],
"deck": {
"data": {
"_resource_child_name": "my_deck",
"_resource_type": "unilabos.resources.my_module.decks:MyDeck"
}
}
}
```
**关键约束:**
- `_resource_child_name` 必须与 Deck 节点的 `id` 一致
- `_resource_type` 为 Deck 类/工厂函数的完整 Python 路径
---
## 常见图模式
### 模式 A单设备调试
最简形式,一个设备节点,无连接:
```json
{
"nodes": [
{
"id": "my_device",
"name": "my_device",
"children": [],
"parent": null,
"type": "device",
"class": "motor.zdt_x42",
"position": {"x": 0, "y": 0, "z": 0},
"config": {"port": "/dev/ttyUSB0", "baudrate": 115200},
"data": {"status": "idle"}
}
],
"links": []
}
```
### 模式 BProtocol 工作站(泵+阀+容器)
工作站配合泵、阀、容器和物理连接,用于协议编译:
```json
{
"nodes": [
{
"id": "station", "name": "协议工作站",
"class": "workstation", "type": "device", "parent": null,
"children": ["pump", "valve", "flask_solvent", "reactor", "waste"],
"config": {"protocol_type": ["PumpTransferProtocol"]}
},
{"id": "pump", "name": "转移泵", "class": "virtual_transfer_pump",
"type": "device", "parent": "station",
"config": {"port": "VIRTUAL", "max_volume": 25.0},
"data": {"status": "Idle", "position": 0.0, "valve_position": "0"}},
{"id": "valve", "name": "多通阀", "class": "virtual_multiway_valve",
"type": "device", "parent": "station",
"config": {"port": "VIRTUAL", "positions": 8}},
{"id": "flask_solvent", "name": "溶剂瓶", "type": "container",
"class": null, "parent": "station",
"config": {"max_volume": 1000.0},
"data": {"liquid": [{"liquid_type": "DMF", "liquid_volume": 500}]}},
{"id": "reactor", "name": "反应器", "type": "container",
"class": null, "parent": "station"},
{"id": "waste", "name": "废液瓶", "type": "container",
"class": null, "parent": "station"}
],
"links": [
{"source": "pump", "target": "valve", "type": "fluid",
"port": {"pump": "transferpump", "valve": "transferpump"}},
{"source": "valve", "target": "flask_solvent", "type": "fluid",
"port": {"valve": "1", "flask_solvent": "outlet"}},
{"source": "valve", "target": "reactor", "type": "fluid",
"port": {"valve": "2", "reactor": "inlet"}},
{"source": "valve", "target": "waste", "type": "fluid",
"port": {"valve": "3", "waste": "inlet"}}
]
}
```
### 模式 C外部系统工作站 + Deck
```json
{
"nodes": [
{
"id": "bioyond_station", "class": "reaction_station.bioyond",
"parent": null, "children": ["bioyond_deck"],
"config": {
"api_host": "http://192.168.1.100:8080",
"api_key": "YOUR_KEY",
"material_type_mappings": {},
"warehouse_mapping": {}
},
"deck": {
"data": {
"_resource_child_name": "bioyond_deck",
"_resource_type": "unilabos.resources.bioyond.decks:BIOYOND_PolymerReactionStation_Deck"
}
}
},
{
"id": "bioyond_deck", "class": "BIOYOND_PolymerReactionStation_Deck",
"parent": "bioyond_station", "type": "deck",
"config": {"type": "BIOYOND_PolymerReactionStation_Deck", "setup": true}
}
],
"links": []
}
```
### 模式 D通信代理串口设备
泵通过串口设备通信,使用 `communication` 类型的 link。加载时系统会自动将串口端口信息写入泵的 `config`
```json
{
"nodes": [
{"id": "station", "name": "工作站", "type": "device",
"class": "workstation", "parent": null,
"children": ["serial_1", "pump_1"]},
{"id": "serial_1", "name": "串口", "type": "device",
"class": "serial", "parent": "station",
"config": {"port": "COM7", "baudrate": 9600}},
{"id": "pump_1", "name": "注射泵", "type": "device",
"class": "syringe_pump_with_valve.runze.SY03B-T08", "parent": "station"}
],
"links": [
{"source": "pump_1", "target": "serial_1", "type": "communication",
"port": {"pump_1": "port", "serial_1": "port"}}
]
}
```
---
## 验证
```bash
# 启动测试
unilab -g unilabos/test/experiments/<name>.json --complete_registry
# 仅检查注册表
python -m unilabos --check_mode --skip_env_check
```
---
## 高级模式
处理复杂图文件时,详见 [reference.md](reference.md)ResourceDict 完整字段 schema、Pose 标准化规则、Handle 验证机制、GraphML 格式支持、外部系统工作站完整 config 结构。
---
## 常见错误
| 错误 | 原因 | 修复 |
|------|------|------|
| `class` 找不到 | 注册表中无此设备名 | 在 `unilabos/registry/devices/``resources/` 中搜索正确名称 |
| children/parent 不一致 | 子节点 `parent` 与父节点 `children` 不匹配 | 确保双向一致 |
| `_resource_child_name` 不匹配 | Deck 引用名与 Deck 节点 `id` 不同 | 保持一致 |
| Link 端口错误 | `port` 中的 key 不是 source/target 的 `id` | key 必须是对应节点的 `id` |
| 重复 UUID | 多个节点有相同 `uuid` | 删除或修改 UUID |
---
## 参考路径
| 内容 | 路径 |
|------|------|
| 图文件目录 | `unilabos/test/experiments/` |
| 协议测试站 | `unilabos/test/experiments/Protocol_Test_Station/` |
| 图加载代码 | `unilabos/resources/graphio.py` |
| 节点模型 | `unilabos/resources/resource_tracker.py` |
| 设备注册表 | `unilabos/registry/devices/` |
| 资源注册表 | `unilabos/registry/resources/` |
| 用户文档 | `docs/user_guide/graph_files.md` |

View File

@@ -0,0 +1,255 @@
# 实验图高级参考
本文件是 SKILL.md 的补充,包含 ResourceDict 完整 schema、Handle 验证、GraphML 格式、Pose 标准化规则和复杂图文件结构。Agent 在需要处理这些场景时按需阅读。
---
## 1. ResourceDict 完整字段
`unilabos/resources/resource_tracker.py` 中定义的节点数据模型:
| 字段 | 类型 | 别名 | 说明 |
|------|------|------|------|
| `id` | `str` | — | 节点唯一标识 |
| `uuid` | `str` | — | 全局唯一标识 |
| `name` | `str` | — | 显示名称 |
| `description` | `str` | — | 描述(默认 `""` |
| `resource_schema` | `Dict[str, Any]` | `schema` | 资源 schema |
| `model` | `Dict[str, Any]` | — | 3D 模型信息 |
| `icon` | `str` | — | 图标(默认 `""` |
| `parent_uuid` | `Optional[str]` | — | 父节点 UUID |
| `parent` | `Optional[ResourceDict]` | — | 父节点引用(序列化时 exclude |
| `type` | `Union[Literal["device"], str]` | — | 节点类型 |
| `klass` | `str` | `class` | 注册表类名 |
| `pose` | `ResourceDictPosition` | — | 位姿信息 |
| `config` | `Dict[str, Any]` | — | 配置参数 |
| `data` | `Dict[str, Any]` | — | 运行时数据 |
| `extra` | `Dict[str, Any]` | — | 扩展数据 |
### Pose 完整结构ResourceDictPosition
| 字段 | 类型 | 默认值 | 说明 |
|------|------|--------|------|
| `size` | `{width, height, depth}` | `{0,0,0}` | 节点尺寸 |
| `scale` | `{x, y, z}` | `{1,1,1}` | 缩放比例 |
| `layout` | `"2d"/"x-y"/"z-y"/"x-z"` | `"x-y"` | 布局方向 |
| `position` | `{x, y, z}` | `{0,0,0}` | 2D 位置 |
| `position3d` | `{x, y, z}` | `{0,0,0}` | 3D 位置 |
| `rotation` | `{x, y, z}` | `{0,0,0}` | 旋转角度 |
| `cross_section_type` | `"rectangle"/"circle"/"rounded_rectangle"` | `"rectangle"` | 横截面形状 |
---
## 2. Position / Pose 标准化规则
图文件中的 `position` 有多种写法,加载时自动标准化。
### 输入格式兼容
```json
// 格式 A: 直接 {x, y, z}(最常用)
"position": {"x": 100, "y": 200, "z": 0}
// 格式 B: 嵌套 position
"position": {"position": {"x": 100, "y": 200, "z": 0}}
// 格式 C: 使用 pose 字段
"pose": {"position": {"x": 100, "y": 200, "z": 0}}
// 格式 D: 顶层 x, y, z无 position 字段)
"x": 100, "y": 200, "z": 0
```
### 标准化流程
1. **graphio.py `canonicalize_nodes_data`**:若 `position` 不是 dict从节点顶层提取 `x/y/z` 填入 `pose.position`
2. **resource_tracker.py `get_resource_instance_from_dict`**:若 `position.x` 存在(旧格式),转为 `{"position": {"x":..., "y":..., "z":...}}`
3. `pose.size``config.size_x/size_y/size_z` 自动填充
---
## 3. Handle 验证
启动时系统验证 link 中的 `sourceHandle` / `targetHandle` 是否在注册表的 `handles` 中定义。
```python
# unilabos/app/main.py (约 449-481 行)
source_handler_keys = [
h["handler_key"] for h in materials[source_node.klass]["handles"]
if h["io_type"] == "source"
]
target_handler_keys = [
h["handler_key"] for h in materials[target_node.klass]["handles"]
if h["io_type"] == "target"
]
if source_handle not in source_handler_keys:
print_status(f"节点 {source_node.id} 的source端点 {source_handle} 不存在", "error")
resource_edge_info.pop(...) # 移除非法 link
```
**Handle 定义在注册表 YAML 中:**
```yaml
my_device:
handles:
- handler_key: access
io_type: target
data_type: fluid
side: NORTH
label: access
```
> 大多数简单设备不定义 handles此验证仅对有 `sourceHandle`/`targetHandle` 的 link 生效。
---
## 4. GraphML 格式支持
除 JSON 外,系统也支持 GraphML 格式(`unilabos/resources/graphio.py::read_graphml`)。
### 与 JSON 的关键差异
| 特性 | JSON | GraphML |
|------|------|---------|
| 父子关系 | `parent`/`children` 字段 | `::` 分隔的节点 ID`station::pump_1` |
| 加载后 | 直接解析 | 先 `nx.read_graphml` 再转 JSON 格式 |
| 输出 | 不生成副本 | 自动生成等价的 `.json` 文件 |
### GraphML 转换流程
```
nx.read_graphml(file)
↓ 用 label 重映射节点名
↓ 从 "::" 推断 parent_relation
nx.relabel_nodes + nx.node_link_data
↓ canonicalize_nodes_data + canonicalize_links_ports
↓ 写出等价 JSON 文件
physical_setup_graph + handle_communications
```
---
## 5. 复杂图文件结构示例
### 外部系统工作站完整 config
`reaction_station_bioyond.json` 为例,工作站 `config` 中的关键字段:
```json
{
"config": {
"api_key": "DE9BDDA0",
"api_host": "http://172.21.103.36:45388",
"workflow_mappings": {
"scheduler_start": {"workflow": "start", "params": {}},
"create_order": {"workflow": "create_order", "params": {}}
},
"material_type_mappings": {
"BIOYOND_PolymerStation_Reactor": ["反应器", "type-uuid-here"],
"BIOYOND_PolymerStation_1BottleCarrier": ["试剂瓶", "type-uuid-here"]
},
"warehouse_mapping": {
"堆栈1左": {
"uuid": "warehouse-uuid-here",
"site_uuids": {
"A01": "site-uuid-1",
"A02": "site-uuid-2"
}
}
},
"http_service_config": {
"enabled": true,
"host": "0.0.0.0",
"port": 45399,
"routes": ["/callback/workflow", "/callback/material"]
},
"deck": {
"data": {
"_resource_child_name": "Bioyond_Deck",
"_resource_type": "unilabos.resources.bioyond.decks:BIOYOND_PolymerReactionStation_Deck"
}
},
"size_x": 2700.0,
"size_y": 1080.0,
"size_z": 2500.0,
"protocol_type": [],
"data": {}
}
}
```
### 子设备 Reactor 节点
```json
{
"id": "reactor_1",
"name": "reactor_1",
"parent": "reaction_station_bioyond",
"type": "device",
"class": "bioyond_reactor",
"position": {"x": 1150, "y": 300, "z": 0},
"config": {
"reactor_index": 0,
"bioyond_workflow_key": "reactor_1"
},
"data": {}
}
```
### Deck 节点
```json
{
"id": "Bioyond_Deck",
"name": "Bioyond_Deck",
"parent": "reaction_station_bioyond",
"type": "deck",
"class": "BIOYOND_PolymerReactionStation_Deck",
"position": {"x": 0, "y": 0, "z": 0},
"config": {
"type": "BIOYOND_PolymerReactionStation_Deck",
"setup": true,
"rotation": {"x": 0, "y": 0, "z": 0, "type": "Rotation"}
},
"data": {}
}
```
---
## 6. Link 端口标准化
`graphio.py::canonicalize_links_ports` 处理 `port` 字段的多种格式:
```python
# 输入: 字符串格式 "(A,B)"
"port": "(pump_1, valve_1)"
# 输出: 字典格式
"port": {"source_id": "pump_1", "target_id": "valve_1"}
# 输入: 已是字典
"port": {"pump_1": "port", "serial_1": "port"}
# 保持不变
# 输入: 无 port 字段
# 自动补充空 port
```
---
## 7. 关键路径
| 内容 | 路径 |
|------|------|
| ResourceDict 模型 | `unilabos/resources/resource_tracker.py` |
| 图加载 + 标准化 | `unilabos/resources/graphio.py` |
| Handle 验证 | `unilabos/app/main.py` (449-481 行) |
| 反应站图文件 | `unilabos/test/experiments/reaction_station_bioyond.json` |
| 配液站图文件 | `unilabos/test/experiments/dispensing_station_bioyond.json` |
| 用户文档 | `docs/user_guide/graph_files.md` |