[SOLVED] Laptop lid (monitor/screen)

kj6loh · 04-19-2017, 07:07 PM

Is there a way to catch the lid opening event? or do I just have to poll /proc/acpi/button/lid/LID/state? I would like to do something when it immediately goes from closed to open. I do have the acpi and acpi-support packages installed. I am using Debian 9 (stretch).

What I want to do is on the lid open I want to move a program (kodi) to xfce workspace 1 (wmctrl -r kodi -t 1).

Alternatively, if I could make kodi persistent in workspace 1, that would be ok too. This will achieve the same desired effect.

What is happening now is that every time the laptop (lid) is closed, kodi goes to workspace 0 (my main workspace).

I do have a startup script that waits for kodi to start and then puts it in workspace 1.

TenTenths · 04-21-2017, 07:45 AM

I wonder if you could set up incron to monitor that file for changes (the IN_MODIFY event) and then run a script appropriately:

https://www.howtoforge.com/tutorial/...s-with-incron/

BW-userx · 04-21-2017, 11:08 AM

if that dang thing knows when to go to sleep whenever ones closes it then yeah. I do not see why not. because it got a know when you open it up again and hit that space bar to wake it up too.

so on open - it got a get a good slap first to wake it up.

so maybe you got a look into that event instead, or the two events put together.

Laserbeak · 04-23-2017, 01:09 AM

It would be OS and/or hardware dependent. I did some basic searches, and didn't find anything in the MacOS X apis that were able to tell a piece of code that the lid was opened or closed. That doesn't mean it doesn't exist, I just couldn't easily find it.

sundialsvcs · 04-23-2017, 10:08 AM

See e.g. https://unix.stackexchange.com/quest...ose-laptop-lid

Pearlseattle · 04-23-2017, 03:58 PM

Quote:

or do I just have to poll /proc/acpi/button/lid/LID/state?

I guess so? (EDIT: meaning that yes that might work - but what I show below is to just exec commands or scripts as a reaction to the acpi event)

I did a small test with my "/etc/acpi/default.sh", which now looks like this...

Code:

#!/bin/sh

set $*

group=${1%%/*}
action=${1#*/}
device=$2
id=$3
value=$4

log_unhandled() {
        logger "MYCUSTOMLOG-unhandled acpi event: group $group, action $action, device $device, id $id"
}

case "$group" in
        button)
                case "$action" in
                        lid)
                                case "$id" in
                                        open)
                                                logger "MYCUSTOMLOG-Hey you just opened the lid"
                                                ;;
                                        *)
                                                log_unhandled $*
                                                ;;
                                esac
                                ;;
                        *)
                                log_unhandled $*
                                ;;
                esac
                ;;
        *)
                log_unhandled $*
                ;;
esac
;;

...and now after having restarted "/etc/init.d/acpid" and having closed & reopened the lid I see the following popping up in "/var/log/messages":

Code:

Apr 23 22:56:26 asus nonroot[19683]: MYCUSTOMLOG-unhandled acpi event: group button, action lid, device LID, id close
Apr 23 22:56:29 asus nonroot[19686]: MYCUSTOMLOG-Hey you just opened the lid

Once you manage to make the message appear in the log you can replace that with any script you would like to run... .

kj6loh · 04-24-2017, 09:31 AM

Thanks guys!

ondoho · 04-24-2017, 02:42 PM

Quote:

Originally Posted by kj6loh

Thanks guys!

please share your solution.
a forum is a 2-way street.

suicidaleggroll · 04-24-2017, 03:29 PM

Not the OP, but I had to do something similar on my laptop a while back. My touchpad was going NUTS whenever I would close the lid, I'm guessing electrical activity in or behind the screen was somehow registering as taps/drags on the touchpad, as soon as I closed the lid my mouse would start going everywhere, "clicking" on things on the desktop, etc.

I ended up making a small ACPI rule to disable the touchpad whenever the lid was closed, and re-enable it when it was opened, that would probably work for the OP or anyone else who finds this thread.

/etc/acpi/lid.sh:

Code:

#!/bin/bash
DEVICE="SynPS/2 Synaptics TouchPad"
grep -q closed /proc/acpi/button/lid/LID/state
su user -c "export DISPLAY=:0.0; /usr/bin/xinput set-int-prop '$DEVICE' 'Device Enabled' 8 $?"

/etc/acpi/events/lm_lid:

Code:

event=button/lid.*
action=/etc/acpi/lid.sh

get it running:

Code:

chkconfig acpid on
service acpid restart

ssh in from another system and run:

Code:

export DISPLAY=:0.0
xinput --watch-props "SynPS/2 Synaptics TouchPad"

as you open/close the lid you'll see the touchpad switch on and off, or replace the lid.sh script with whatever else you want to run when the lid is opened/closed.

kj6loh · 04-26-2017, 02:14 PM

So my original question was basically how execute something on laptop lid open. I didn't really want to write a daemon that polled /proc/acpi/button/lid/LID/state, which is easy enough to do (I will leave this as an exercise to the reader).

So what I did was, which I think pearlseattle was getting at, was use the acpi package: I'm not sure but you may have to add the acpi-support package.

I basically followed this guide: https://help.ubuntu.com/community/La...FClose_Scripts

which initially did not work for me. Ultimately what I did was put a delay in the open script, and it worked.

Pearlseattle · 04-26-2017, 05:42 PM

Ah-ha, a delay, that's interesting!
Does this mean that when you close the lid your notebook goes to sleep (suspend-to-ram/suspend-to-disk - doesn't matter)?
If yes, then this might mean that the ACPI-event of "lid open" fires too early/quickly in relation to the "readiness" of the desktop environment (when the desktop env. is not yet fully restored respectively ready to accept hints)?

kj6loh · 04-26-2017, 07:24 PM

I guess, I put in a logger entry which did come up, and followed it with the command I wanted, but that didn't execute. So I guess the program itself has to wake up. And BTW I just switch the display off, no suspend to ram, or anything like that.