Update Tkinter label during threading in Python

Question

I am currently trying to program a robot arm which is controlled by a Raspberry Pi.

Everything works fine so far, except for one thing and I already googled and tried everything for many hours but can't find a working solution.

For the movement of the robot arm it is necessary to run all motors "simultaneously" with threads (works fine).

The problem I have is that I need to update a label which shows the current angle of an axis (motor) as soon as it finished its movement but other motors are still running (threads).

After a lot of research I thought I found the solution by using a queue and Tkinters after-method. But it still doesn't work as the labels text only gets updated after all threads terminated.

I wrote an example code where I want to get a label update for motor "one" which will finish its for-loop (100 iterations) before motor "two" (500 iterations). I expected the label to get updated as soon as motor one reached its target while motor two is still runing.

But although I used the after-method it still waits till motor two finished before updating the label.

Hope you can help me!

from tkinter import *
import threading
import time
from queue import * 


class StepperMotors:

    def __init__(self, root):
        self.root = root

        self.start_btn = Button(root, text="Start", command=lambda:self.start_movement())
        self.start_btn.config(width = 10)
        self.start_btn.grid(row=1,column=1)

        self.label_one = Label(root, text='')
        self.label_one.config(width = 10)
        self.label_one.grid(row=2, column=1)

        self.label_two = Label(root, text='')
        self.label_two.config(width = 10)
        self.label_two.grid(row=3, column=1)

    def start_movement(self):
        self.thread_queue = Queue()
        self.root.after(100, self.wait_for_finish) 

        thread_one = threading.Thread(target=self.motor_actuation, args=(1,100))
        thread_two = threading.Thread(target=self.motor_actuation, args=(2,500))

        thread_one.start()
        thread_two.start()

        thread_one.join()
        thread_two.join()

    def motor_actuation(self, motor, iterations):  
        for i in range(iterations):
            i = i+1  
            update_text = str(motor) + " " + str(i) + "\n"
            print(update_text)
            time.sleep(0.01)

        self.thread_queue.put(update_text)

    def wait_for_finish(self):
        try:      
            self.text = self.thread_queue.get()  
            self.label_one.config(text=self.text)  

        except self.thread_queue.empty():
            self.root.after(100, self.wait_for_finish)

if __name__ == "__main__":
    root = Tk()
    root.title("test")
    stepper = StepperMotors(root)
    root.mainloop()

Answer 1

It's better to use daemon thread which are non-blocking.

Also, It's better to have a separation of concerns : a robot (or robot arm) can be a object which has its own life time: a daemon thread. Idem, you can define a "LabelUpdater" which read the state of a robot and update a label.

Let's define a robot:

• it is created on application initialisation and is run when the user click the "Start" button,
• The robot moves and reports its angle in a app-level multithreading-queue,

class Robot(threading.Thread):

    def __init__(self, name: str, label_queue: queue.Queue, end_pos: int):
        super().__init__(name=name)
        self.daemon = True
        self.label_queue = label_queue
        self.end_pos = end_pos

    def run(self) -> None:
        for angle in range(self.end_pos):
            self.label_queue.put(angle)
            time.sleep(0.01)

Let's define a LabelUpdater:

• It is created on application initialisation and run forever (it can observe a robot even if it is not running).
• It reads the robot queue (for instance each second to avoid blinking) and update the label

class LabelUpdater(threading.Thread):
    def __init__(self, name: str, label_queue: queue.Queue, root_app: tkinter.Tk, variable: tkinter.Variable):
        super().__init__(name=name)
        self.daemon = True
        self.label_queue = label_queue
        self.root_app = root_app
        self.variable = variable

    def run(self) -> None:
        # run forever
        while True:
            # wait a second please
            time.sleep(1)
            # consume all the queue and keep only the last message
            last_msg = None
            while True:
                try:
                    msg = self.label_queue.get(block=False)
                except queue.Empty:
                    break
                last_msg = msg
                self.label_queue.task_done()
            if last_msg:
                self.variable.set(last_msg)

Then, the main application should define:

• 2 multithreading queues: one for each label,
• 2 tkinter.StringVar variable which will be updated,
• the robots and the label updaters,
• the two updaters are started, and will run forever.

class StepperMotors:
    def __init__(self, root):
        self.root = root
        self.label_one_queue = queue.Queue()
        self.label_two_queue = queue.Queue()

        self.start_btn = tkinter.Button(root, text="Start", command=lambda: self.start_movement())
        self.start_btn.config(width=10)
        self.start_btn.grid(row=1, column=1)

        self.text_one = tkinter.StringVar()
        self.text_one.set("one")
        self.label_one = tkinter.Label(root, textvariable=self.text_one)
        self.label_one.config(width=10)
        self.label_one.grid(row=2, column=1)

        self.text_two = tkinter.StringVar()
        self.text_two.set("two")
        self.label_two = tkinter.Label(root, textvariable=self.text_two)
        self.label_two.config(width=10)
        self.label_two.grid(row=3, column=1)

        self.robot_one = Robot("robot_one", self.label_one_queue, 100)
        self.robot_two = Robot("robot_two", self.label_two_queue, 500)

        self.updater_one = LabelUpdater("updater_one", self.label_one_queue, self.root, self.text_one)
        self.updater_two = LabelUpdater("updater_two", self.label_two_queue, self.root, self.text_two)
        self.updater_one.start()
        self.updater_two.start()

    def start_movement(self):
        self.robot_one.start()
        self.robot_two.start()

Of course, you need a flag or something to check that each robot is not already running.