I'll have to turn off forced-homing so LCNC can run an M-code after machine-on but before homing... but that should be fairly simple.
Ah, forgot about that. Here is a solution with a python component that does not rely on MDI:
1. Save this as 'py2bash.py' to a folder called 'python' in your config folder and mark executable (You should not need to modify this):
Code:
#!/usr/bin/env python3
import os
import hal
import linuxcnc
h = hal.component("py2bash")
h.newpin("run-script", hal.HAL_BIT, hal.HAL_IN)
h.ready()
# create a connection to the status channel
s = linuxcnc.stat()
print("py2bash: Started")
script_has_run = False
try:
while 1:
if h['run-script'] == True and script_has_run == False:
print("py2bash: Running bash script now.")
os.system('bash ./script.sh')
script_has_run = True
except KeyboardInterrupt:
raise SystemExit
2. Save this as 'script.sh' to your config folder and mark as executable (you will need to change 'halcmd loadusr -W hal_input -K +Keyboard:0 ' to load your component)
Code:
#!/bin/bash
halcmd loadusr -W hal_input -K +Keyboard:0
halcmd source myhal.hal
exit 0
3. Add this to [HAL] section in your ini (No need to change this):
Code:
HALCMD = loadusr -W ./python/py2bash.py
4. Run your config and check for a pin 'py2bash.start-script':
Attachment not found
5. If you force this pin HIGH the python component should run 'script.sh'
