Ready for open source (#47)

* Create app/main API

* create example device

* create ROS backend and example device SDK Wrapper

* Add ROS host and host starting from app.py

* Add gripper device and mock implementation

* add "status_types" & "action_types" to ROS device decorator

* add ActionServer debug example

* [bugfix] complete mock gripper example

* ROS Backend Host for Device action calling and Resource management

* add conda/mamba ENV file

* add host_node communication with app/main.py

* add action message value mappings and converters

* Update ilabos.yaml

* Update issue templates

* example devices.json and resources.json

* Fix Device wrapper to use async property and actions (#7)

* Fix Device wrapper to use async property and actions

* Resolve #1 : support async get methods and actions. Give new examples.

* add both sync/async GRBL controller SDK

* 2 call device actions from appmainpy api to ros hostpy (#8)

* feature: add job

* fix:node start

* feature:add get job status

* fix:get device

* clean

* Resolve #5 device connection diagram and workflow compilation support (#9)

* add syringe pump device and its compilation using device connection diagram

* add RunzeSyringePump real device with ROS2 example

* Prototype machine with 1 pump and 1 CNC

* add ROS2ProtocolNode and related functions

* add ilabos_msgs (to use PumpTransfer action)

* add example device connection graph

* refactor protocol_node code into separate file

* add ROS2SerialNode

* add SerialCommand srv in ilabos_msgs

* add pump_protocol example, and fix bugs

* [fix] serial service: avoid async service deadlock by directly call serial `send_command`

* use SendCmd instead of SingleJointPosition for valve control

* initialize device connection graph when server starts

* Fix #5: async workflow execution (#16)

* add rclpyx and protocol example for async-native workflow

* use async in ROS2ProtocolNode, and host initialization

* add examples of "ros-async" protocol implementation, and `run_in_event_loop` for using native async functions

* use "ros-async" in protocols and device nodes

* fix pump_protocol: push to 0.0 μL

* Envs, docs, and conda recipes (#19)

* update ENV: use python 3.11 and deprecate ros-humble-gazebo-ros

* add ilabos-msgs conda recipe

* Update ilabos.yaml

* fix recipe and env yaml

* Add sphinx docs

* add aichemeco

* add bioyong

* add bioyong

* Support XDL devices & protocols (#20)

* [Feature] support multiple protocols in a single ProtocolNode

* add Junjie's code

* Support "Clean" protocol

* Update Grignard_flow_batchreact_single_pumpvalve.json

* test_grignard_add

* add stir device node and example

* Update device_node.py

add print_publish flag to control the node's log output

* NH4Cl_add

* add "HeaterStirrer" device and "HeatChill" action

* add wait time after each pump action for equilibration

* fix stir

* add Separate protocol

* Refactor Separator device and Stir action

* add rotavap_node

* fix stir

* add chiller node

* Move rinsings into PumpTransfer

* Fix SeparateProtocol under refactored Separator device and Stir action

* Supports automatically add new protocol action_types

* fix PumpTransfer protocol because of rinsing

* Add Rotavap and Chiller devices

* fix SeparateProtocol

* add EvaporateProtocol

* add rotavap devices config

* fix HeaterStirrer and SeparatorController IO

* Fix automatically add new protocol action_types

* Add HeaterStirrer and SeparatorController device config

* fix pump protocols

* Fix Evaporate action

* Update evaporate_protocol.py

* add temp_sensor node and add function remap

* update docs

---------

Co-authored-by: 王俊杰 <1800011822@pku.edu.cn>
Co-authored-by: q434343 <554662886@qq.com>

* fix aichemeco

* [Bugfix] fix Windows conda packaging

* add file upload api

* update dependencies: force to use 3.11 and remove conflict in WIN64 and OSX64

* update dependencies: force to use 3.11 and remove conflict in WIN64 and OSX64

* Create aichemeco_simple.py

* fix

* update

* add aichemeco file

* MQTT [1/2]: action start (#25)

* add mq

* fix

* clean

* add class

* fix excel

* update bioyong

* add api

* fix

---------

Co-authored-by: caok@dp.tech <xiaoyeqiannian@163.com>

* motor & grasp

* Add Grasp motor support and enhance EleGripper class

- Introduced a new motor configuration for Grasp in sjtu.json.
- Updated EleGripper class to inherit from UniversalDriver and added status property.
- Implemented move_and_rotate method for coordinated movement.
- Adjusted threading logic in EleGripper initialization.
- Registered Grasp motor in ROS2 device node configuration.

This commit enhances the motor control capabilities by integrating the Grasp motor and improving the existing EleGripper functionality.

* fix read data lenth

* Update Grasp.py

* MQTT (2/2): publish Device Status, Action Feedback & Results (#27)

* Add bridges in HostNode for device_status publishing

* Add "bridges" selection (fastapi & mqtt) when app start

* add MQ feedback & result publisher, and fix message converter

* fix UUID converting between ROS and MQ

* lint api model.py

* Continuous controllers: PID, MPC, custom controllers etc. (#23)

* add controller config & wrapper

* add controller setup at app.main

* control loop example

* fix com port

* add agv , ur_arm and raman

* MQTT (3/4): Unified Resources and Sync when starting the server (#28)

* update http upload api

* generate uuid when init device

* example resource json

* fix

* add new example full-content json (device, resource, graph)

* fix full-content json and related reading code

* fix

* add json_schema when initialize resources

* fix

* update schema

* refactor heaterstirrer.dalong

* fix

* fix refactor heaterstirrer.dalong

* refactor syringepump.runze: use ml instead of μL

* Update ilabos/ros/host.py

Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com>

---------

Co-authored-by: 王俊杰 <1800011822@pku.edu.cn>
Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com>

* Distributed initialization with self-organizing network (#29)

* add distributed launching option "--without_host"

* fix

---------

Co-authored-by: 王俊杰 <1800011822@pku.edu.cn>

* Refactor Workstation: Add resource service and tracking (#30)

* move ilabos/ros/rpc to ilabos/device_comms/rpc, and merge bioyond/aichemeco files under /devices

* add Resource srv and message_converter

* move graphio to ilabos/resources

* refactor resources type conversion

* add resource clients in device_node

* add mock resources service

* pass Gripper1 resource test

* update http resource services

* add AGV compile function

* add AGV transfer protocol

* update full mock_gripper edit_id example

* update full mock_gripper edit_id example

* get and update resource also in protocol_node

* mock resource update in AichemecoHiwo

* Create HT_hiwo.json

* add children in resources

* bugfixes

* fix rpc

* add Revvity winprep

---------

Co-authored-by: wjjxxx <43375851+wjjxxx@users.noreply.github.com>
Co-authored-by: 3218923350 <105201755+3218923350@users.noreply.github.com>

* Distributed launch (2/2): distributed resource create (#32)

* add resource_add request to host for slave mode

* add AGV

* fix protocol resources

* optimize host callbacks

* bugfixes

* add revvity registry

---------

Co-authored-by: 王俊杰 <1800011822@pku.edu.cn>
Co-authored-by: wjjxxx <43375851+wjjxxx@users.noreply.github.com>

* Refactor Driver Files Structure (#33)

* Integration with pywinauto & recorder
Added execute run and initialize procdure

* 酶标仪状态检测、使用示例,整体流程待测试

* nivo ready version

* Add HPLC driver and example script

- Introduced HPLCDriver class for managing HPLC device status and operations.
- Implemented device status monitoring and command execution via ROS2 actions.
- Added example script (hplc.py) demonstrating how to run commands on the HPLC device.
- Created PlayerUtil and UniversalDriver classes for shared functionality across devices.
- Refactored NivoDriver to utilize the new UniversalDriver structure.
- Enhanced error handling and process management in the NivoDriver.

* 修复start的错误定位

* hplc tested ok

* relative path to build msgs

* template_driver & jiageng devices

* fetch correct status type and action type

* fix mtype fetch

* gpc bus integration

* ilab build

* remove chs

* recipe rename

* modbus update 1

* json available

* hplc & modbus rewrite

* Update AgilentHPLC.py

hplc datafile reader

* move ilabos/ros/rpc to ilabos/device_comms/rpc, and merge bioyond/aichemeco files under /devices

* modbus分设备

* gpc

* gpc 2

* fix address

* default register node

* fix MainScreenGPC

* add Resource srv and message_converter

* move graphio to ilabos/resources

* refactor resources type conversion

* add resource clients in device_node

* add mock resources service

* pass Gripper1 resource test

* update http resource services

* add AGV compile function

* add AGV transfer protocol

* update recipe.yaml

* update full mock_gripper edit_id example

* update full mock_gripper edit_id example

* get and update resource also in protocol_node

* mock resource update in AichemecoHiwo

* feat: add other jiageng PLC device code

* ilabos compile

* correct format

* correct recipe format

* correct setup.py format

* remove unnecessary files

* remove unnecessary files

* Create HT_hiwo.json

* add children in resources

* hplc support sample_id

* correct hplc sample_id

* correct hplc sample_id

* hplc upload

* fix type hint

* oss upload tested ver

* recipe yaml fix for linux

* update installation yaml

* refactor: moved all driver files according to its feat

* merge main to dev

---------

Co-authored-by: 王俊杰 <2201110460@stu.pku.edu.cn>
Co-authored-by: Junhan Chang <changjh@pku.edu.cn>
Co-authored-by: jiawei <miaojiawei@dp.tech>

* add: NMR LH and RU device control (#34)

* Add Registry for device drivers and Support GraphML (#35)

* read chemputer graphml

* read graphml in app/main

* add devices in ros/devices

* add schema for devices

* read registry directory and initialize when entry from main

* Delete devices.py

* Update add_protocol.md

* delete unecessary files

* feat: 2278 devices registry yaml (#36)

* read chemputer graphml

* read graphml in app/main

* add devices in ros/devices

* add schema for devices

* read registry directory and initialize when entry from main

* Delete devices.py

* add: NMR LH and RU device control

* fix: modify jiageng devices registry

---------

Co-authored-by: Junhan Chang <changjh@pku.edu.cn>

* Device/Resource Registry and GraphML support (#37)

* add resource type conversion to PLR

* add resource registry and test

* add docs

* fix registry

* add solenoid_valve_mock, its registry and test

* fix registry for directly using examples

* add EvacuateAndRefillProtocol and testcases

* allow function sequence call in ACTION

* add read & write & extra_info for hardware_interface

* Update device_node.py

* add solenoid valve

* add doc developer guide yaml

* fixes for starting IK station

* add graphml grouping parser

* fix graphml grouping parser

* add communication edge parser

* fix io solenoid valve

* Update .gitignore

* Update plates.yaml

---------

Co-authored-by: ColumbiaCC <2100011801@stu.pku.edu.cn>

* Uni-Lab Doc v0.2 (#39)

* add Uni-Lab docs

* change doc name

* Dev (#41)

* Integration with pywinauto & recorder
Added execute run and initialize procdure

* 酶标仪状态检测、使用示例,整体流程待测试

* nivo ready version

* Add HPLC driver and example script

- Introduced HPLCDriver class for managing HPLC device status and operations.
- Implemented device status monitoring and command execution via ROS2 actions.
- Added example script (hplc.py) demonstrating how to run commands on the HPLC device.
- Created PlayerUtil and UniversalDriver classes for shared functionality across devices.
- Refactored NivoDriver to utilize the new UniversalDriver structure.
- Enhanced error handling and process management in the NivoDriver.

* 修复start的错误定位

* hplc tested ok

* relative path to build msgs

* template_driver & jiageng devices

* fetch correct status type and action type

* fix mtype fetch

* gpc bus integration

* ilab build

* remove chs

* recipe rename

* modbus update 1

* json available

* hplc & modbus rewrite

* Update AgilentHPLC.py

hplc datafile reader

* move ilabos/ros/rpc to ilabos/device_comms/rpc, and merge bioyond/aichemeco files under /devices

* modbus分设备

* gpc

* gpc 2

* fix address

* default register node

* fix MainScreenGPC

* add Resource srv and message_converter

* move graphio to ilabos/resources

* refactor resources type conversion

* add resource clients in device_node

* add mock resources service

* pass Gripper1 resource test

* update http resource services

* add AGV compile function

* add AGV transfer protocol

* update recipe.yaml

* update full mock_gripper edit_id example

* update full mock_gripper edit_id example

* get and update resource also in protocol_node

* mock resource update in AichemecoHiwo

* feat: add other jiageng PLC device code

* ilabos compile

* correct format

* correct recipe format

* correct setup.py format

* remove unnecessary files

* remove unnecessary files

* Create HT_hiwo.json

* add children in resources

* hplc support sample_id

* correct hplc sample_id

* correct hplc sample_id

* hplc upload

* fix type hint

* oss upload tested ver

* recipe yaml fix for linux

* update installation yaml

* refactor: moved all driver files according to its feat

* merge main to dev

* add HPLC registry and json

* 升级 ros2-distro-mutex 依赖版本至 0.6

* 修改 ros2-distro-mutex 依赖版本为通配符匹配

* 更新 ros-humble-ilabos-msgs 依赖为 robostack-humble 命名空间

* add resource type conversion to PLR

* add resource registry and test

* feat: 更新oss上传

* fix device id

* add docs

* fix registry

* add solenoid_valve_mock, its registry and test

* fix registry for directly using examples

* add EvacuateAndRefillProtocol and testcases

* allow function sequence call in ACTION

* add read & write & extra_info for hardware_interface

* Update device_node.py

* add solenoid valve

* add doc developer guide yaml

* use robostack-staging

* rclpy version test

* lower rclpy

* ensure 0.6* env

* fixes for starting IK station

* add graphml grouping parser

* fix graphml grouping parser

* add communication edge parser

* fix io solenoid valve

* Update .gitignore

* Update plates.yaml

* Feature/device node later init (#40)

* 修改config路径,方便后续打包
增加device_node打印

* 支持plr序列化/init创建

* 统一命名

* import mgr
logger optimize
banner print

* 日志OK

* fix unicorn frame

* banner print

* correct import format

* file path changes

* 取消后补全,在加载设备的时候直接替换

* converter update

* web page update

* 在线device更新,node继承替换

* 修复动作、状态的类型缺失 和 命令提示

* web功能实现结束

* host节点更改完成
新增status时间戳管理
新增每10s动态发现其他node

* ros2类型的节点也应该被包一次

* 修复类型提示

* websocket 动态显示状态

* add workflow & book theme for docs

* add workflow & book theme for docs

* fix workflow build

* fix workflow build

* 理清启动关系

* stm32 example

* mac . name

* device_instance device_cls

* 新增config添加方式
更新mqtt提示

* plr test

* 移动is_host_mode
新增slave_no_host

* 确保config优先修改生效

* fix graph io

* 支持带参数传入

* 支持物料解析

* 支持物料解析

* device为空的时候不进行绑定或初始化

* protocol node new

* protocol node runnable

* protocol node runnable

---------

Co-authored-by: 王俊杰 <2201110460@stu.pku.edu.cn>
Co-authored-by: Junhan Chang <changjh@pku.edu.cn>
Co-authored-by: jiawei <miaojiawei@dp.tech>
Co-authored-by: ColumbiaCC <2100011801@stu.pku.edu.cn>

* Dev (#45)

* Integration with pywinauto & recorder
Added execute run and initialize procdure

* 酶标仪状态检测、使用示例,整体流程待测试

* nivo ready version

* Add HPLC driver and example script

- Introduced HPLCDriver class for managing HPLC device status and operations.
- Implemented device status monitoring and command execution via ROS2 actions.
- Added example script (hplc.py) demonstrating how to run commands on the HPLC device.
- Created PlayerUtil and UniversalDriver classes for shared functionality across devices.
- Refactored NivoDriver to utilize the new UniversalDriver structure.
- Enhanced error handling and process management in the NivoDriver.

* 修复start的错误定位

* hplc tested ok

* relative path to build msgs

* template_driver & jiageng devices

* fetch correct status type and action type

* fix mtype fetch

* gpc bus integration

* ilab build

* remove chs

* recipe rename

* modbus update 1

* json available

* hplc & modbus rewrite

* Update AgilentHPLC.py

hplc datafile reader

* move ilabos/ros/rpc to ilabos/device_comms/rpc, and merge bioyond/aichemeco files under /devices

* modbus分设备

* gpc

* gpc 2

* fix address

* default register node

* fix MainScreenGPC

* add Resource srv and message_converter

* move graphio to ilabos/resources

* refactor resources type conversion

* add resource clients in device_node

* add mock resources service

* pass Gripper1 resource test

* update http resource services

* add AGV compile function

* add AGV transfer protocol

* update recipe.yaml

* update full mock_gripper edit_id example

* update full mock_gripper edit_id example

* get and update resource also in protocol_node

* mock resource update in AichemecoHiwo

* feat: add other jiageng PLC device code

* ilabos compile

* correct format

* correct recipe format

* correct setup.py format

* remove unnecessary files

* remove unnecessary files

* Create HT_hiwo.json

* add children in resources

* hplc support sample_id

* correct hplc sample_id

* correct hplc sample_id

* hplc upload

* fix type hint

* oss upload tested ver

* recipe yaml fix for linux

* update installation yaml

* refactor: moved all driver files according to its feat

* merge main to dev

* add HPLC registry and json

* 升级 ros2-distro-mutex 依赖版本至 0.6

* 修改 ros2-distro-mutex 依赖版本为通配符匹配

* 更新 ros-humble-ilabos-msgs 依赖为 robostack-humble 命名空间

* add resource type conversion to PLR

* add resource registry and test

* feat: 更新oss上传

* fix device id

* add docs

* fix registry

* add solenoid_valve_mock, its registry and test

* fix registry for directly using examples

* add EvacuateAndRefillProtocol and testcases

* allow function sequence call in ACTION

* add read & write & extra_info for hardware_interface

* Update device_node.py

* add solenoid valve

* add doc developer guide yaml

* use robostack-staging

* rclpy version test

* lower rclpy

* ensure 0.6* env

* fixes for starting IK station

* add graphml grouping parser

* fix graphml grouping parser

* add communication edge parser

* fix io solenoid valve

* Update .gitignore

* Update plates.yaml

* Feature/device node later init (#40)

* 修改config路径,方便后续打包
增加device_node打印

* 支持plr序列化/init创建

* 统一命名

* import mgr
logger optimize
banner print

* 日志OK

* fix unicorn frame

* banner print

* correct import format

* file path changes

* 取消后补全,在加载设备的时候直接替换

* converter update

* web page update

* 在线device更新,node继承替换

* 修复动作、状态的类型缺失 和 命令提示

* web功能实现结束

* host节点更改完成
新增status时间戳管理
新增每10s动态发现其他node

* ros2类型的节点也应该被包一次

* 修复类型提示

* websocket 动态显示状态

* add workflow & book theme for docs

* add workflow & book theme for docs

* fix workflow build

* fix workflow build

* 理清启动关系

* stm32 example

* mac . name

* device_instance device_cls

* 新增config添加方式
更新mqtt提示

* plr test

* 移动is_host_mode
新增slave_no_host

* 确保config优先修改生效

* fix graph io

* 支持带参数传入

* 支持物料解析

* 支持物料解析

* device为空的时候不进行绑定或初始化

* protocol node new

* protocol node runnable

* protocol node runnable

* Feature/device node later init (#42)

* 修改config路径,方便后续打包
增加device_node打印

* 支持plr序列化/init创建

* 统一命名

* import mgr
logger optimize
banner print

* 日志OK

* fix unicorn frame

* banner print

* correct import format

* file path changes

* 取消后补全,在加载设备的时候直接替换

* converter update

* web page update

* 在线device更新,node继承替换

* 修复动作、状态的类型缺失 和 命令提示

* web功能实现结束

* host节点更改完成
新增status时间戳管理
新增每10s动态发现其他node

* ros2类型的节点也应该被包一次

* 修复类型提示

* websocket 动态显示状态

* add workflow & book theme for docs

* add workflow & book theme for docs

* fix workflow build

* fix workflow build

* 理清启动关系

* stm32 example

* mac . name

* device_instance device_cls

* 新增config添加方式
更新mqtt提示

* plr test

* 移动is_host_mode
新增slave_no_host

* 确保config优先修改生效

* fix graph io

* 支持带参数传入

* 支持物料解析

* 支持物料解析

* device为空的时候不进行绑定或初始化

* protocol node new

* protocol node runnable

* protocol node runnable

* action

* plr suc

* plr suc!!

* plr suc!!

* plr suc!!

* plr msgs

* Feature/device node later init (#43)

* 修改config路径,方便后续打包
增加device_node打印

* 支持plr序列化/init创建

* 统一命名

* import mgr
logger optimize
banner print

* 日志OK

* fix unicorn frame

* banner print

* correct import format

* file path changes

* 取消后补全,在加载设备的时候直接替换

* converter update

* web page update

* 在线device更新,node继承替换

* 修复动作、状态的类型缺失 和 命令提示

* web功能实现结束

* host节点更改完成
新增status时间戳管理
新增每10s动态发现其他node

* ros2类型的节点也应该被包一次

* 修复类型提示

* websocket 动态显示状态

* add workflow & book theme for docs

* add workflow & book theme for docs

* fix workflow build

* fix workflow build

* 理清启动关系

* stm32 example

* mac . name

* device_instance device_cls

* 新增config添加方式
更新mqtt提示

* plr test

* 移动is_host_mode
新增slave_no_host

* 确保config优先修改生效

* fix graph io

* 支持带参数传入

* 支持物料解析

* 支持物料解析

* device为空的时候不进行绑定或初始化

* protocol node new

* protocol node runnable

* protocol node runnable

* action

* plr suc

* plr suc!!

* plr suc!!

* plr suc!!

* plr msgs

* plr

* action

* plr reg fix

* Feature/device node later init (#44)

* 修改config路径,方便后续打包
增加device_node打印

* 支持plr序列化/init创建

* 统一命名

* import mgr
logger optimize
banner print

* 日志OK

* fix unicorn frame

* banner print

* correct import format

* file path changes

* 取消后补全,在加载设备的时候直接替换

* converter update

* web page update

* 在线device更新,node继承替换

* 修复动作、状态的类型缺失 和 命令提示

* web功能实现结束

* host节点更改完成
新增status时间戳管理
新增每10s动态发现其他node

* ros2类型的节点也应该被包一次

* 修复类型提示

* websocket 动态显示状态

* add workflow & book theme for docs

* add workflow & book theme for docs

* fix workflow build

* fix workflow build

* 理清启动关系

* stm32 example

* mac . name

* device_instance device_cls

* 新增config添加方式
更新mqtt提示

* plr test

* 移动is_host_mode
新增slave_no_host

* 确保config优先修改生效

* fix graph io

* 支持带参数传入

* 支持物料解析

* 支持物料解析

* device为空的时候不进行绑定或初始化

* protocol node new

* protocol node runnable

* protocol node runnable

* action

* plr suc

* plr suc!!

* plr suc!!

* plr suc!!

* plr msgs

* plr

* fix convert error
fix async logic error
added async error print

* new device test

* test resource add

* test resource add

* test resource add

* test resource add

* local env setup

* node type fix
temp fix root_node error
fix convert res from type error

* resource tracker

* fix bug from qhh

* fix bug from qhh

* fix bug from qhh

* fix bug from qhh

* refactor MQTT client logging and connection handling; update group ID in config

* driver_params allow empty

* allow other init param

* fix driver param and enhance type hint

* refactor MQConfig to use double quotes for string literals

* fix wrong function calling

* fix wrong function calling

* fix log for mac

* fix networkx compatibility

* add mqtt loggers

* add action to jsonschema converter

* random client id

* type converter & registry

* correct conversion

* fix action publish only from discovered devices

* add "Bio" tag for action doc generation

* 改进module提示

* Fix doc

* mqtt不连接也可用
性价样例提示

* add docs

* 更新plr test案例

* Update intro.md

* 更新有机案例

* skip

---------

Co-authored-by: Harvey Que <Q-Query@outlook.com>
Co-authored-by: Junhan Chang <1700011741@pku.edu.cn>

---------

Co-authored-by: 王俊杰 <2201110460@stu.pku.edu.cn>
Co-authored-by: Junhan Chang <changjh@pku.edu.cn>
Co-authored-by: jiawei <miaojiawei@dp.tech>
Co-authored-by: ColumbiaCC <2100011801@stu.pku.edu.cn>
Co-authored-by: Harvey Que <Q-Query@outlook.com>
Co-authored-by: Junhan Chang <1700011741@pku.edu.cn>

* Canonicalize before Open Source (#46)

* big big refactor try01

* refactor 02

---------

Co-authored-by: ck <xiaoyeqiannian@163.com>
Co-authored-by: 王俊杰 <1800011822@pku.edu.cn>
Co-authored-by: q434343 <554662886@qq.com>
Co-authored-by: Xuwznln <xuwznln@gmail.com>
Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com>
Co-authored-by: wjjxxx <43375851+wjjxxx@users.noreply.github.com>
Co-authored-by: 3218923350 <105201755+3218923350@users.noreply.github.com>
Co-authored-by: Xuwznln <1023025701@qq.com>
Co-authored-by: 王俊杰 <2201110460@stu.pku.edu.cn>
Co-authored-by: jiawei <miaojiawei@dp.tech>
Co-authored-by: Jiawei <91898272+jiawei723@users.noreply.github.com>
Co-authored-by: ColumbiaCC <2100011801@stu.pku.edu.cn>
Co-authored-by: Harvey Que <Q-Query@outlook.com>
This commit is contained in:
Junhan Chang
2025-04-17 15:09:58 +08:00
committed by GitHub
parent 7ccb425e39
commit a62a695812
266 changed files with 40772 additions and 2 deletions

View File

@@ -0,0 +1,672 @@
import threading
import time
import traceback
import uuid
from typing import get_type_hints, TypeVar, Generic, Dict, Any, Type, TypedDict, Optional
from concurrent.futures import ThreadPoolExecutor
import asyncio
import rclpy
from rclpy.node import Node
from rclpy.action import ActionServer
from rclpy.action.server import ServerGoalHandle
from rclpy.client import Client
from rclpy.callback_groups import ReentrantCallbackGroup
from unilabos.resources.graphio import convert_resources_to_type, convert_resources_from_type
from unilabos.ros.msgs.message_converter import (
convert_to_ros_msg,
convert_from_ros_msg,
convert_from_ros_msg_with_mapping,
convert_to_ros_msg_with_mapping,
)
from unilabos_msgs.srv import ResourceAdd, ResourceGet, ResourceDelete, ResourceUpdate, ResourceList # type: ignore
from unilabos_msgs.msg import Resource # type: ignore
from unilabos.ros.nodes.resource_tracker import DeviceNodeResourceTracker
from unilabos.ros.x.rclpyx import get_event_loop
from unilabos.ros.utils.driver_creator import ProtocolNodeCreator, PyLabRobotCreator, DeviceClassCreator
from unilabos.utils.async_util import run_async_func
from unilabos.utils.log import info, debug, warning, error, critical, logger
from unilabos.utils.type_check import get_type_class
T = TypeVar("T")
# 在线设备注册表
registered_devices: Dict[str, "DeviceInfoType"] = {}
# 实现同时记录自定义日志和ROS2日志的适配器
class ROSLoggerAdapter:
"""同时向自定义日志和ROS2日志发送消息的适配器"""
@property
def identifier(self):
return f"{self.namespace}/{self.node_name}"
def __init__(self, ros_logger, node_name, namespace):
"""
初始化日志适配器
Args:
ros_logger: ROS2日志记录器
node_name: 节点名称
namespace: 命名空间
"""
self.ros_logger = ros_logger
self.node_name = node_name
self.namespace = namespace
self.level_2_logger_func = {
"info": info,
"debug": debug,
"warning": warning,
"error": error,
"critical": critical,
}
def _log(self, level, msg, *args, **kwargs):
"""实际执行日志记录的内部方法"""
# 添加前缀,使日志更易识别
msg = f"[{self.identifier}] {msg}"
# 向ROS2日志发送消息标准库logging不支持stack_level参数
ros_log_func = getattr(self.ros_logger, "debug") # 默认发送debug这样不会显示在控制台
ros_log_func(msg)
self.level_2_logger_func[level](msg, *args, stack_level=1, **kwargs)
def debug(self, msg, *args, **kwargs):
"""记录DEBUG级别日志"""
self._log("debug", msg, *args, **kwargs)
def info(self, msg, *args, **kwargs):
"""记录INFO级别日志"""
self._log("info", msg, *args, **kwargs)
def warning(self, msg, *args, **kwargs):
"""记录WARNING级别日志"""
self._log("warning", msg, *args, **kwargs)
def error(self, msg, *args, **kwargs):
"""记录ERROR级别日志"""
self._log("error", msg, *args, **kwargs)
def critical(self, msg, *args, **kwargs):
"""记录CRITICAL级别日志"""
self._log("critical", msg, *args, **kwargs)
def init_wrapper(
self,
device_id: str,
driver_class: type[T],
status_types: Dict[str, Any],
action_value_mappings: Dict[str, Any],
hardware_interface: Dict[str, Any],
print_publish: bool,
children: Optional[list] = None,
driver_params: Optional[Dict[str, Any]] = None,
driver_is_ros: bool = False,
*args,
**kwargs,
):
"""初始化设备节点的包装函数和ROS2DeviceNode初始化保持一致"""
if driver_params is None:
driver_params = kwargs.copy()
if children is None:
children = []
kwargs["device_id"] = device_id
kwargs["driver_class"] = driver_class
kwargs["driver_params"] = driver_params
kwargs["status_types"] = status_types
kwargs["action_value_mappings"] = action_value_mappings
kwargs["hardware_interface"] = hardware_interface
kwargs["children"] = children
kwargs["print_publish"] = print_publish
kwargs["driver_is_ros"] = driver_is_ros
super(type(self), self).__init__(*args, **kwargs)
class PropertyPublisher:
def __init__(
self,
node: "BaseROS2DeviceNode",
name: str,
get_method,
msg_type,
initial_period: float = 5.0,
print_publish=True,
):
self.node = node
self.name = name
self.msg_type = msg_type
self.get_method = get_method
self.timer_period = initial_period
self.print_publish = print_publish
self._value = None
self.publisher_ = node.create_publisher(msg_type, f"{name}", 10)
self.timer = node.create_timer(self.timer_period, self.publish_property)
self.__loop = get_event_loop()
str_msg_type = str(msg_type)[8:-2]
self.node.lab_logger().debug(f"发布属性: {name}, 类型: {str_msg_type}, 周期: {initial_period}")
def get_property(self):
if asyncio.iscoroutinefunction(self.get_method):
# 如果是异步函数,运行事件循环并等待结果
self.node.get_logger().debug(f"【PropertyPublisher.get_property】获取异步属性: {self.name}")
loop = self.__loop
if loop:
future = asyncio.run_coroutine_threadsafe(self.get_method(), loop)
self._value = future.result()
return self._value
else:
self.node.get_logger().error(f"【PropertyPublisher.get_property】事件循环未初始化")
return None
else:
# 如果是同步函数,直接调用并返回结果
self.node.get_logger().debug(f"【PropertyPublisher.get_property】获取同步属性: {self.name}")
self._value = self.get_method()
return self._value
async def get_property_async(self):
try:
# 获取异步属性值
self.node.get_logger().debug(f"【PropertyPublisher.get_property_async】异步获取属性: {self.name}")
self._value = await self.get_method()
except Exception as e:
self.node.get_logger().error(f"【PropertyPublisher.get_property_async】获取异步属性出错: {str(e)}")
def publish_property(self):
try:
self.node.get_logger().debug(f"【PropertyPublisher.publish_property】开始发布属性: {self.name}")
value = self.get_property()
if self.print_publish:
self.node.get_logger().info(f"【PropertyPublisher.publish_property】发布 {self.msg_type}: {value}")
if value is not None:
msg = convert_to_ros_msg(self.msg_type, value)
self.publisher_.publish(msg)
self.node.get_logger().debug(f"【PropertyPublisher.publish_property】属性 {self.name} 发布成功")
except Exception as e:
traceback.print_exc()
self.node.get_logger().error(f"【PropertyPublisher.publish_property】发布属性出错: {str(e)}")
def change_frequency(self, period):
# 动态改变定时器频率
self.timer_period = period
self.node.get_logger().info(
f"【PropertyPublisher.change_frequency】修改 {self.name} 定时器周期为: {self.timer_period}"
)
# 重置定时器
self.timer.cancel()
self.timer = self.node.create_timer(self.timer_period, self.publish_property)
class BaseROS2DeviceNode(Node, Generic[T]):
"""
ROS2设备节点基类
这个类提供了ROS2设备节点的基本功能包括属性发布、动作服务等。
通过泛型参数T来指定具体的设备类型。
"""
@property
def identifier(self):
return f"{self.namespace}/{self.device_id}"
node_name: str
namespace: str
# TODO 要删除,添加时间相关的属性,避免动态添加属性的警告
time_spent = 0.0
time_remaining = 0.0
create_action_server = True
def __init__(
self,
driver_instance: T,
device_id: str,
status_types: Dict[str, Any],
action_value_mappings: Dict[str, Any],
hardware_interface: Dict[str, Any],
print_publish=True,
resource_tracker: Optional["DeviceNodeResourceTracker"] = None,
):
"""
初始化ROS2设备节点
Args:
driver_instance: 设备实例
device_id: 设备标识符
status_types: 需要发布的状态和传感器信息
action_value_mappings: 设备动作
hardware_interface: 硬件接口配置
print_publish: 是否打印发布信息
"""
self.driver_instance = driver_instance
self.device_id = device_id
self.uuid = str(uuid.uuid4())
self.publish_high_frequency = False
self.callback_group = ReentrantCallbackGroup()
self.resource_tracker = resource_tracker
# 初始化ROS节点
self.node_name = f'{device_id.split("/")[-1]}'
self.namespace = f"/devices/{device_id}"
Node.__init__(self, self.node_name, namespace=self.namespace) # type: ignore
if self.resource_tracker is None:
self.lab_logger().critical("资源跟踪器未初始化,请检查")
# 创建自定义日志记录器
self._lab_logger = ROSLoggerAdapter(self.get_logger(), self.node_name, self.namespace)
self._action_servers = {}
self._property_publishers = {}
self._status_types = status_types
self._action_value_mappings = action_value_mappings
self._hardware_interface = hardware_interface
self._print_publish = print_publish
# 创建属性发布者
for attr_name, msg_type in self._status_types.items():
if isinstance(attr_name, (int, float)):
if "param" in msg_type.keys():
pass
else:
for k, v in msg_type.items():
self.create_ros_publisher(k, v, initial_period=5.0)
else:
self.create_ros_publisher(attr_name, msg_type)
# 创建动作服务
if self.create_action_server:
for action_name, action_value_mapping in self._action_value_mappings.items():
self.create_ros_action_server(action_name, action_value_mapping)
# 创建线程池执行器
self._executor = ThreadPoolExecutor(max_workers=max(len(action_value_mappings), 1))
# 创建资源管理客户端
self._resource_clients: Dict[str, Client] = {
"resource_add": self.create_client(ResourceAdd, "/resources/add"),
"resource_get": self.create_client(ResourceGet, "/resources/get"),
"resource_delete": self.create_client(ResourceDelete, "/resources/delete"),
"resource_update": self.create_client(ResourceUpdate, "/resources/update"),
"resource_list": self.create_client(ResourceList, "/resources/list"),
}
# 向全局在线设备注册表添加设备信息
self.register_device()
rclpy.get_global_executor().add_node(self)
self.lab_logger().debug(f"ROS节点初始化完成")
def register_device(self):
"""向注册表中注册设备信息"""
topics_info = self._property_publishers.copy()
actions_info = self._action_servers.copy()
# 创建设备信息
device_info = DeviceInfoType(
id=self.device_id,
uuid=self.uuid,
node_name=self.node_name,
namespace=self.namespace,
driver_instance=self.driver_instance,
status_publishers=topics_info,
actions=actions_info,
hardware_interface=self._hardware_interface,
base_node_instance=self,
)
# 加入全局注册表
registered_devices[self.device_id] = device_info
def lab_logger(self):
"""
获取实验室自定义日志记录器
这个日志记录器会同时向ROS2日志和自定义日志发送消息
并使用node_name和namespace作为标识。
Returns:
日志记录器实例
"""
return self._lab_logger
def create_ros_publisher(self, attr_name, msg_type, initial_period=5.0):
"""创建ROS发布者"""
# 获取属性值的方法
def get_device_attr():
try:
if hasattr(self.driver_instance, f"get_{attr_name}"):
return getattr(self.driver_instance, f"get_{attr_name}")()
else:
return getattr(self.driver_instance, attr_name)
except AttributeError as ex:
self.lab_logger().error(
f"publish error, {str(type(self.driver_instance))[8:-2]} has no attribute '{attr_name}'"
)
self._property_publishers[attr_name] = PropertyPublisher(
self, attr_name, get_device_attr, msg_type, initial_period, self._print_publish
)
def create_ros_action_server(self, action_name, action_value_mapping):
"""创建ROS动作服务器"""
action_type = action_value_mapping["type"]
str_action_type = str(action_type)[8:-2]
self._action_servers[action_name] = ActionServer(
self,
action_type,
action_name,
execute_callback=self._create_execute_callback(action_name, action_value_mapping),
callback_group=ReentrantCallbackGroup(),
)
self.lab_logger().debug(f"发布动作: {action_name}, 类型: {str_action_type}")
def _create_execute_callback(self, action_name, action_value_mapping):
"""创建动作执行回调函数"""
async def execute_callback(goal_handle: ServerGoalHandle):
self.lab_logger().info(f"执行动作: {action_name}")
goal = goal_handle.request
# 从目标消息中提取参数, 并调用对应的方法
if "sequence" in self._action_value_mappings:
# 如果一个指令对应函数的连续调用,如启动和等待结果,默认参数应该属于第一个函数调用
def ACTION(**kwargs):
for i, action in enumerate(self._action_value_mappings["sequence"]):
if i == 0:
self.lab_logger().info(f"执行序列动作第一步: {action}")
getattr(self.driver_instance, action)(**kwargs)
else:
self.lab_logger().info(f"执行序列动作后续步骤: {action}")
getattr(self.driver_instance, action)()
action_paramtypes = get_type_hints(
getattr(self.driver_instance, self._action_value_mappings["sequence"][0])
)
else:
ACTION = getattr(self.driver_instance, action_name)
action_paramtypes = get_type_hints(ACTION)
action_kwargs = convert_from_ros_msg_with_mapping(goal, action_value_mapping["goal"])
self.lab_logger().debug(f"接收到原始目标: {action_kwargs}")
# 向Host查询物料当前状态
for k, v in goal.get_fields_and_field_types().items():
if v in ["unilabos_msgs/Resource", "sequence<unilabos_msgs/Resource>"]:
self.lab_logger().info(f"查询资源状态: Key: {k} Type: {v}")
try:
r = ResourceGet.Request()
r.id = action_kwargs[k]["id"] if v == "unilabos_msgs/Resource" else action_kwargs[k][0]["id"]
r.with_children = True
response = await self._resource_clients["resource_get"].call_async(r)
except Exception:
logger.error(f"资源查询失败,默认使用本地资源")
# 删除对response.resources的检查因为它总是存在
resources_list = [convert_from_ros_msg(rs) for rs in response.resources] # type: ignore # FIXME
self.lab_logger().debug(f"资源查询结果: {len(resources_list)} 个资源")
type_hint = action_paramtypes[k]
final_type = get_type_class(type_hint)
# 判断 ACTION 是否需要特殊的物料类型如 pylabrobot.resources.Resource并做转换
final_resource = convert_resources_to_type(resources_list, final_type)
action_kwargs[k] = self.resource_tracker.figure_resource(final_resource)
self.lab_logger().info(f"准备执行: {action_kwargs}, 函数: {ACTION.__name__}")
time_start = time.time()
time_overall = 100
# 将阻塞操作放入线程池执行
if asyncio.iscoroutinefunction(ACTION):
try:
self.lab_logger().info(f"异步执行动作 {ACTION}")
future = ROS2DeviceNode.run_async_func(ACTION, **action_kwargs)
except Exception as e:
self.lab_logger().error(f"创建异步任务失败: {traceback.format_exc()}")
raise e
else:
self.lab_logger().info(f"同步执行动作 {ACTION}")
future = self._executor.submit(ACTION, **action_kwargs)
action_type = action_value_mapping["type"]
feedback_msg_types = action_type.Feedback.get_fields_and_field_types()
result_msg_types = action_type.Result.get_fields_and_field_types()
while not future.done():
if goal_handle.is_cancel_requested:
self.lab_logger().info(f"取消动作: {action_name}")
future.cancel() # 尝试取消线程池中的任务
goal_handle.canceled()
return action_type.Result()
self.time_spent = time.time() - time_start
self.time_remaining = time_overall - self.time_spent
# 发布反馈
feedback_values = {}
for msg_name, attr_name in action_value_mapping["feedback"].items():
if hasattr(self.driver_instance, f"get_{attr_name}"):
method = getattr(self.driver_instance, f"get_{attr_name}")
if not asyncio.iscoroutinefunction(method):
feedback_values[msg_name] = method()
elif hasattr(self.driver_instance, attr_name):
feedback_values[msg_name] = getattr(self.driver_instance, attr_name)
if self._print_publish:
self.lab_logger().info(f"反馈: {feedback_values}")
feedback_msg = convert_to_ros_msg_with_mapping(
ros_msg_type=action_type.Feedback(),
obj=feedback_values,
value_mapping=action_value_mapping["feedback"],
)
goal_handle.publish_feedback(feedback_msg)
time.sleep(0.5)
if future.cancelled():
self.lab_logger().info(f"动作 {action_name} 已取消")
return action_type.Result()
self.lab_logger().info(f"动作执行完成: {action_name}")
del future
# 向Host更新物料当前状态
for k, v in goal.get_fields_and_field_types().items():
if v not in ["unilabos_msgs/Resource", "sequence<unilabos_msgs/Resource>"]:
continue
self.lab_logger().info(f"更新资源状态: {k}")
r = ResourceUpdate.Request()
# 仅当action_kwargs[k]不为None时尝试转换
akv = action_kwargs[k]
apv = action_paramtypes[k]
final_type = get_type_class(apv)
if final_type is None:
continue
try:
r.resources = [
convert_to_ros_msg(Resource, self.resource_tracker.root_resource(rs))
for rs in convert_resources_from_type(akv, final_type) # type: ignore # FIXME # 考虑反查到最大的
]
response = await self._resource_clients["resource_update"].call_async(r)
self.lab_logger().debug(f"资源更新结果: {response}")
except Exception as e:
self.lab_logger().error(f"资源更新失败: {e}")
self.lab_logger().error(traceback.format_exc())
# 发布结果
goal_handle.succeed()
self.lab_logger().info(f"设置动作成功: {action_name}")
result_values = {}
for msg_name, attr_name in action_value_mapping["result"].items():
if hasattr(self.driver_instance, f"get_{attr_name}"):
result_values[msg_name] = getattr(self.driver_instance, f"get_{attr_name}")()
elif hasattr(self.driver_instance, attr_name):
result_values[msg_name] = getattr(self.driver_instance, attr_name)
result_msg = convert_to_ros_msg_with_mapping(
ros_msg_type=action_type.Result(), obj=result_values, value_mapping=action_value_mapping["result"]
)
for attr_name in result_msg_types.keys():
if attr_name in ["success", "reached_goal"]:
setattr(result_msg, attr_name, True)
self.lab_logger().info(f"动作 {action_name} 完成并返回结果")
return result_msg
return execute_callback
# 异步上下文管理方法
async def __aenter__(self):
"""进入异步上下文"""
self.lab_logger().info(f"进入异步上下文: {self.device_id}")
if hasattr(self.driver_instance, "__aenter__"):
await self.driver_instance.__aenter__() # type: ignore
self.lab_logger().info(f"异步上下文初始化完成: {self.device_id}")
return self
async def __aexit__(self, exc_type, exc_val, exc_tb):
"""退出异步上下文"""
self.lab_logger().info(f"退出异步上下文: {self.device_id}")
if hasattr(self.driver_instance, "__aexit__"):
await self.driver_instance.__aexit__(exc_type, exc_val, exc_tb) # type: ignore
self.lab_logger().info(f"异步上下文清理完成: {self.device_id}")
class DeviceInitError(Exception):
pass
class ROS2DeviceNode:
"""
ROS2设备节点类
这个类封装了设备类实例和ROS2节点的功能提供ROS2接口。
它不继承设备类,而是通过代理模式访问设备类的属性和方法。
"""
# 类变量,用于循环管理
_loop = None
_loop_running = False
_loop_thread = None
@classmethod
def get_loop(cls):
return cls._loop
@classmethod
def run_async_func(cls, func, **kwargs):
return run_async_func(func, loop=cls._loop, **kwargs)
@property
def driver_instance(self):
return self._driver_instance
@property
def ros_node_instance(self):
return self._ros_node
def __init__(
self,
device_id: str,
driver_class: Type[T],
driver_params: Dict[str, Any],
status_types: Dict[str, Any],
action_value_mappings: Dict[str, Any],
hardware_interface: Dict[str, Any],
children: Dict[str, Any],
print_publish: bool = True,
driver_is_ros: bool = False,
):
"""
初始化ROS2设备节点
Args:
device_id: 设备标识符
driver_class: 设备类
status_types: 状态类型映射
action_value_mappings: 动作值映射
hardware_interface: 硬件接口配置
children:
print_publish: 是否打印发布信息
driver_is_ros:
"""
# 在初始化时检查循环状态
if ROS2DeviceNode._loop_running and ROS2DeviceNode._loop_thread is not None:
pass
elif ROS2DeviceNode._loop_thread is None:
self._start_loop()
# 保存设备类是否支持异步上下文
self._has_async_context = hasattr(driver_class, "__aenter__") and hasattr(driver_class, "__aexit__")
self._driver_class = driver_class
self.driver_is_ros = driver_is_ros
self.resource_tracker = DeviceNodeResourceTracker()
# use_pylabrobot_creator 使用 cls的包路径检测
use_pylabrobot_creator = driver_class.__module__.startswith("pylabrobot")
# TODO: 要在创建之前预先请求服务器是否有当前id的物料放到resource_tracker中让pylabrobot进行创建
# 创建设备类实例
if use_pylabrobot_creator:
self._driver_creator = PyLabRobotCreator(
driver_class, children=children, resource_tracker=self.resource_tracker
)
else:
from unilabos.ros.nodes.presets.protocol_node import ROS2ProtocolNode
if self._driver_class is ROS2ProtocolNode:
self._driver_creator = ProtocolNodeCreator(driver_class, children=children)
else:
self._driver_creator = DeviceClassCreator(driver_class)
if driver_is_ros:
driver_params["device_id"] = device_id
driver_params["resource_tracker"] = self.resource_tracker
self._driver_instance = self._driver_creator.create_instance(driver_params)
if self._driver_instance is None:
logger.critical(f"设备实例创建失败 {driver_class}, params: {driver_params}")
raise DeviceInitError("错误: 设备实例创建失败")
# 创建ROS2节点
if driver_is_ros:
self._ros_node = self._driver_instance # type: ignore
else:
self._ros_node = BaseROS2DeviceNode(
driver_instance=self._driver_instance,
device_id=device_id,
status_types=status_types,
action_value_mappings=action_value_mappings,
hardware_interface=hardware_interface,
print_publish=print_publish,
resource_tracker=self.resource_tracker,
)
self._ros_node: BaseROS2DeviceNode
self._ros_node.lab_logger().info(f"初始化完成 {self._ros_node.uuid} {self.driver_is_ros}")
def _start_loop(self):
def run_event_loop():
loop = asyncio.new_event_loop()
ROS2DeviceNode._loop = loop
asyncio.set_event_loop(loop)
loop.run_forever()
ROS2DeviceNode._loop_thread = threading.Thread(target=run_event_loop, daemon=True, name="ROS2DeviceNode")
ROS2DeviceNode._loop_thread.start()
logger.info(f"循环线程已启动")
class DeviceInfoType(TypedDict):
id: str
uuid: str
node_name: str
namespace: str
driver_instance: Any
status_publishers: Dict[str, PropertyPublisher]
actions: Dict[str, ActionServer]
hardware_interface: Dict[str, Any]
base_node_instance: BaseROS2DeviceNode