105 + timeout=10
106 + # loop 10 times and then exit, if the property service
107 + # isnt up after 80 sec # it is likely not starting at all
108 + while [ ! -e /dev/socket/property_service ]; do
109 + sleep 8
110 + if [ "$timeout" -le 0 ]; then
111 + stop
112 + exit 0
113 + fi
114 + timeout=$(($timeout - 1))
115 + done
116 +
117 + if [ -x /usr/bin/setprop ]; then
118 + /usr/bin/setprop sys.usb.config mtp,adb
119 + fi
You don't need to go over the timeout logic if setprop is not available.
121 + start mtp-server
Why do we need to manually start the mtp-server if the other upstart job is already watching for android_usb-device-changed udev events?
337 + readFiles("/home/phablet");
How hard would it be to not make it hardcoded to phablet? Guess the server would be per user anyway, right?
105 + timeout=10 property_ service ]; do
106 + # loop 10 times and then exit, if the property service
107 + # isnt up after 80 sec # it is likely not starting at all
108 + while [ ! -e /dev/socket/
109 + sleep 8
110 + if [ "$timeout" -le 0 ]; then
111 + stop
112 + exit 0
113 + fi
114 + timeout=$(($timeout - 1))
115 + done
116 +
117 + if [ -x /usr/bin/setprop ]; then
118 + /usr/bin/setprop sys.usb.config mtp,adb
119 + fi
You don't need to go over the timeout logic if setprop is not available.
121 + start mtp-server
Why do we need to manually start the mtp-server if the other upstart job is already watching for android_ usb-device- changed udev events?
337 + readFiles( "/home/ phablet" );
How hard would it be to not make it hardcoded to phablet? Guess the server would be per user anyway, right?
Rest looks fine, need testing.