This page is part of archived documentation for openHAB 4.3. Go to the current stable version

# Exec Binding

This binding integrates the possibility to execute arbitrary shell commands.

# Supported Things

Currently, the binding supports a single type of Thing, being the command Thing.

# Binding Configuration

For security reasons all commands need to be whitelisted. Allowed commands need to be added to the misc/exec.whitelist file in the configuration directory. Every command needs to be on a separate line.

Example:

/bin/echo "Hello world!"
/usr/local/bin/apcaccess status
php ./configurations/scripts/script.php %2$s

Linux: Note that the commands are executed in the context and with the privileges of the process running the Java Virtual Machine. On a Linux system the system user openhab needs to have the privileges needed to execute your intended command. It is advised to test the correct operation of the command in the scope of the openhab user on the command line first:

sudo -u openhab <YOUR COMMAND>

It is not advised to run the virtual machine as superuser/root.

# Thing Configuration

The "command" Thing requires the command to execute on the shell. Optionally one can specify:

  • transform - A transformation (opens new window) to apply on the execution result string.
  • interval - An interval, in seconds, the command will be repeatedly executed. Default is 60 seconds, set to 0 to avoid automatic repetition.
  • timeout - A time-out, in seconds, the execution of the command will time out, and lastly,
  • autorun - A boolean parameter to make the command execute immediately every time the input channel is sent a different openHAB command. If choosing autorun, you may wish to also set interval=0. Note that sending the same command a second time will not trigger execution.

For each shell command, a separate Thing has to be defined.

Thing exec:command:uniquename [command="/command/to/execute here", interval=15, timeout=5, autorun=false]

The command itself can be enhanced using the well known syntax of the Java formatter class syntax (opens new window). The following parameters are automatically added:

  • the current date (as java.util.Date, example: %1$tY-%1$tm-%1$td)
  • the current (or last) command to the input channel (see below, example: %2$s)

note - if you trigger execution using interval or the run channel, the %2 substitution will use the most recent command (if there has been one) sent to the input channel. The state of the Item linked to input channel is ignored.

# Channels

All Things support the following channels:

Channel Type ID Item Type Description
input String Input parameter to provide to the command
output String Output of the last execution of the command
exit Number The exit value of the last execution of the command
run Switch Send ON to execute the command, the current state tells whether it is running or not
lastexecution DateTime Time/Date the command was last executed, in yyyy-MM-dd'T'HH:mm:ss.SSSZ format

Attention: Linking input to any other item type than String will result in erroneous behavior. If needed, please use a rule to convert your item's state to a string. Also note that only commands (e.g. sendCommand) to the input channel are recognized, updating the item's state will not work (e.g. postUpdate).

# Minimal Example

# demo.things

Thing exec:command:apc [command="/usr/local/bin/apcaccess status", interval=15, timeout=5]
Thing exec:command:myscript [command="php ./configurations/scripts/script.php %2$s", transform="REGEX((.*?))"]

# demo.items

String APCRaw "[%s]" (All) {channel="exec:command:apc:output"}
Switch APCRunning {channel="exec:command:apc:run"}
Number APCExitValue "[%d]" {channel="exec:command:apc:exit"}
DateTime APCLastExecution {channel="exec:command:apc:lastexecution"}

# Full Example

Following is an example how to set up an exec command thing, pass it a parameter, debug it with a rule and set the returned string to a Number Item.

# demo.things

// "%2$s" will be replace by the input channel command, this makes it possible to use one command line with different arguments.
// e.g: "ls" as <YOUR COMMAND> and "-a" or "-l" as additional argument sent to the input channel in the rule.
Thing exec:command:yourcommand [ command="<YOUR COMMAND> %2$s", interval=0, autorun=false ]

# demo.items

Switch YourTrigger "External trigger [%s]"
Number YourNumber "Your Number [%.1f °C]"

// state of the execution, is running or finished
Switch yourcommand_Run {channel="exec:command:yourcommand:run", autoupdate="false"}
// Arguments to be placed for '%2$s' in command line
String yourcommand_Args {channel="exec:command:yourcommand:input"}
// Output of command line execution 
String yourcommand_Out {channel="exec:command:yourcommand:output"}

# demo.sitemap

// Name of file and name of sitemap has to be the same
sitemap demo label="Your Value"
{
        Frame {
            Switch item=YourTrigger
            Text item=YourNumber
        }
}

# demo.rules

rule "Set up your parameters"
when
   Item YourTrigger changed
then
      // here we can take different actions according to source Item
   if(YourTrigger.state == ON){
      yourcommand_Args.sendCommand("Additional Argument to command line for ON")
   }else{
      yourcommand_Args.sendCommand("Different Argument to command line for OFF")
   }
      // Caution : openHAB bus is asynchronous
      // we must let the command work before triggering execution (if autorun false)
end

rule "begin your execution"
when
   Item yourcommand_Args received command
then
      // Separately triggering RUN allows subsequent executions with unchanged parameter %2
      // which autorun does not.
   if (yourcommand_Run.state != ON) {
      yourcommand_Run.sendCommand(ON)
         // the Run indicator state will go ON shortly, and return OFF when script finished
   }else{
      logInfo("Your command exec", "Script already in use, skipping execution.")
   }
end

rule "script complete"
when
   Item yourcommand_Run changed from ON to OFF
then
      // Logging of ending
   logInfo("Your command exec", "Script has completed.")
      // Caution : openHAB bus is asynchronous
      // there is no guarantee the output Item will get updated before the run channel triggers rules
end

rule "process your results"
when
   Item yourcommand_Out received update
then
      // Logging of raw command line result
   logInfo("Your command exec", "Raw result:" + yourcommand_Out.state )
      // If the returned string is just a number it can be parsed
      // If not, a regex or another transformation could be used
   YourNumber.postUpdate(Integer::parseInt(yourcommand_Out.state.toString) as Number)
end

# Source

openHAB community thread with a detailed example. (opens new window)