I have a series of python scripts that connect together (one .py closes and runs a separate .py). This works completely fine when running it through the terminal in VS Code or cmd line. Once it is in a .exe by pyinstaller, only the first code works and the program closes once it tries to execute a separate .py file.
All of the separate python files are saved in the same directory. The first one to open, 'Main.py', has a tkinter interface that allows the user to select which .py script they want to run. The code then closes the Main window and opens the selected python script using exec(open('chosen .py').read()). (This is a simplified version of the initial code but I am having the same issues)
import tkinter as tk
from tkinter import ttk
from tkinter.constants import W
from tkinter import messagebox as mb
""" Open a window to select which separate script to run"""
root = tk.Tk()
root.title('Selection Window')
root.geometry('300x200')
frame_1 = tk.LabelFrame(root, text='Choose Program')
frame_1.pack()
# Using this function to update on radio button select
def radio_button_get():
global program_int
choice = radio_ID.get()
if(choice == 1):
program_int = 1
elif(choice == 2):
program_int = 2
# Display confirmation popup
def run_script():
if(program_int == 1):
select = mb.askokcancel("Confirm", "Run choice 1?")
if(select == 1):
root.destroy()
else:
return
if(program_int == 2):
select = mb.askokcancel("Confirm", "No selection")
if(select == 1):
root.destroy()
else:
return
# Create radio buttons to select program
radio_ID = tk.IntVar()
radio_ID.set(2)
program_int = 2 # Set default selection
choice_1 = tk.Radiobutton(frame_1, text='Execute Script 1', variable=radio_ID, value=1, command=radio_button_get)
choice_1.pack()
no_choice = tk.Radiobutton(frame_1, text='No Selection', variable=radio_ID, value=2, command=radio_button_get)
no_choice.pack()
# Button to run the selected code
run_button = ttk.Button(root, text='Run', command=run_script)
run_button.pack()
root.mainloop()
# Execute the other python script
if(program_int == 1):
exec(open('Script1.py').read())
The next code is the 'Script1.py' file which 'Main.py' runs at the end. This is the step which works fine in VS Code and cmd line, but causes the .exe from pyinstaller to close.
import tkinter as tk
from tkinter import ttk
""" Create this programs GUI window"""
root = tk.Tk()
root.title('Script 1')
def run():
root.destroy()
label = ttk.Label(root, text='Close to run')
label.pack()
button = ttk.Button(root, text='Close', command=run)
button.pack()
root.mainloop()
""" Do some code stuff here"""
# When above code is done, want to return to the Main.py window
exec(open('Main.py').read())
Each independent .py file have been successfully turned into .exe files with pyinstaller previously. The cmd line command that I am using to execute pyinstaller is pyinstaller 'Main.py' This successfully creates a Main.exe in the dist folder and also includes a build folder.
I have read through pyinstallers documentation, but have not found anything that I believe would be useful in this case. The nearest issue I could find was importing python scripts as modules in the .spec file options but since the code executes the python script as a separate entity, I don't think this is the fix.
Would the issue be in how the scripts are coded and referencing each other, or with the installation process with pyinstaller? If I missed something in the documentation that would explain this issue, please let me know and I will look there!
Any help is greatly appreciated, thank you
We must avoid using the .exec command. It is hacky but unsafe. Ref: Running a Python script from another
Instead use import :
# Execute the other python script
if(program_int == 1):
import Script1
And here too:
# When above code is done, want to return to the Main.py window
import Main
That's it, now use pyinstaller.
EDIT:
Why .exe file fails to execute another script, and why exec() is the problem:
According to the documentation:
Pyinstaller analyzes your code to discover every other module and library your script needs in order to execute. Then it collects copies of all those files – including the active Python interpreter! – and puts them with your script in a single folder, or optionally in a single executable file.
So, when pyinstaller is analyzing & creating the .exe file, it only executes the exec() function that time (so no error thrown while pyinstaller runs), pyinstaller does not import it or copies it to your .exe. file, and then after the .exe file is created, upon running it throws error that no such script file exists because it was never compiled into that .exe file.
Thus, using import will actually import the script as module, when pyinstaller is executed, and now your .exe file will give no error.