The switch doesn't work yet but "uci get switch-button.@main[0].func"
gets the currently active configured action for button.
I guess that /etc/rc.button/switch is called on every press of the
button.
I don't know whether this works. On the v4 firmware, we can see files in
/etc/gl-switch.d but I don't know yet how the system determines which
file to choose. I assume it's dependend on the "page" of the display.
But I haven't figured out yet how or rather where those pages are
organised.
I have rsynced the whole device before associating with a new device and
after. The only file that got modified was /etc/oui-tertf/client.db.
We intend to have it stored in memory rather than on flash. This should
be okay since the kernel also holds the MAC addresses in memory.
Before this change, uninstalling the package would complain about
/tmp/sim_change_start not existing:
root@GL-E750:/tmp# opkg remove blue-merle
Removing package blue-merle from root...
The /tmp/ directory exists.
The /etc/ directory exists.
killall: gltertf: no process killed
No file found within /tmp/tertf. No shredding to be done there.
No file found within /etc/tertf. No shredding to be done there.
Looks like /tmp/ is clean!
Looks like /etc/ is clean!
rm: can't remove '/tmp/sim_change_start': No such file or directory
rm: can't remove '/tmp/sim_change_switch': No such file or directory
No packages removed.
Collected errors:
* pkg_run_script: package "blue-merle" postrm script returned status 1.
root@GL-E750:/tmp#
So we use -f to not raise an error if the file does not exist.
We assume that the version 3 series is compatible, but still prompt the
user if they are using a version that we have not yet tested.
We know that version 4 is not compatible, so we bail out directly.
The double-dollar is for escaping the Makefile.
