about_Functions_Advanced_Methods

Short description

Describes how functions that specify the CmdletBinding attribute can use the methods and properties that are available to compiled cmdlets.

Long description

Functions that specify the CmdletBinding attribute can access additional methods and properties through the $PSCmdlet variable. These methods include the following methods:

  • The same input-processing methods available to all functions.
  • The ShouldProcess and ShouldContinue methods that are used to get user feedback before an action is performed.
  • The ThrowTerminatingError method for generating error records.
  • Several Write methods that return different types of output.

All the methods and properties of the PSCmdlet class are available to advanced functions. For more information, see System.Management.Automation.PSCmdlet.

For more information about the CmdletBinding attribute, see about_Functions_CmdletBindingAttribute. For the CmdletBindingAttribute class, see System.Management.Automation.Cmdlet.CmdletBindingAttribute.

Input processing methods

The methods described in this section are referred to as the input processing methods. For functions, these three methods are represented by the begin, process, and end blocks of the function.

You aren't required to use any of these blocks in your functions. If you don't use a named block, then PowerShell puts the code in the end block of the function. However, if you use any of these named blocks, or define a dynamicparam block, you must put all code in a named block.

The following example shows the outline of a function that contains a begin block for one-time preprocessing, a process block for multiple record processing, and an end block for one-time post-processing.

Function Test-ScriptCmdlet
{
[CmdletBinding(SupportsShouldProcess=$True)]
    Param ($Parameter1)
    begin{}
    process{}
    end{}
}

Note

These blocks apply to all functions, not just functions that use the CmdletBinding attribute.

begin

This block is used to provide optional one-time preprocessing for the function. The PowerShell runtime uses the code in this block once for each instance of the function in the pipeline.

process

This block is used to provide record-by-record processing for the function. You can use a process block without defining the other blocks. The number of process block executions depends on how you use the function and what input the function receives.

The automatic variable $_ or $PSItem contains the current object in the pipeline for use in the process block. The $input automatic variable contains an enumerator that's only available to functions and script blocks. For more information, see about_Automatic_Variables.

  • Calling the function at the beginning, or outside of a pipeline, executes the process block once.
  • Within a pipeline, the process block executes once for each input object that reaches the function.
  • If the pipeline input that reaches the function is empty, the process block does not execute.
    • The begin and end blocks still execute.

Important

If a function parameter is set to accept pipeline input, and a process block isn't defined, record-by-record processing will fail. In this case, your function will only execute once, regardless of the input.

When you create a function that accepts pipeline input and uses CmdletBinding, the process block should use the parameter variable you defined for pipeline input instead of $_ or $PSItem. For example:

function Get-SumOfNumbers {
    [CmdletBinding()]
    param (
        [Parameter(Mandatory, Position=0, ValueFromPipeline)]
        [int[]]$Numbers
    )

    begin { $retValue = 0 }

    process {
       foreach ($n in $Numbers) {
           $retValue += $n
       }
    }

    end { $retValue }
}

PS> Get-SumOfNumbers 1,2,3,4
10
PS> 1,2,3,4 | Get-SumOfNumbers
10

end

This block is used to provide optional one-time post-processing for the function.

Confirmation methods

ShouldProcess

This method is called to request confirmation from the user before the function performs an action that would change the system. The function can continue based on the Boolean value returned by the method. This method can only be called only from within the Process{} block of the function. The CmdletBinding attribute must also declare that the function supports ShouldProcess (as shown in the previous example).

For more information about this method, see System.Management.Automation.Cmdlet.ShouldProcess.

For more information about how to request confirmation, see Requesting Confirmation.

ShouldContinue

This method is called to request a second confirmation message. It should be called when the ShouldProcess method returns $true. For more information about this method, see System.Management.Automation.Cmdlet.ShouldContinue.

Error methods

Functions can call two different methods when errors occur. When a non-terminating error occurs, the function should call the WriteError method, which is described in the Write methods section. When a terminating error occurs and the function can't continue, it should call the ThrowTerminatingError method. You can also use the Throw statement for terminating errors and the Write-Error cmdlet for non-terminating errors.

For more information, see System.Management.Automation.Cmdlet.ThrowTerminatingError.

Write methods

A function can call the following methods to return different types of output. Notice that not all the output goes to the next command in the pipeline. You can also use the various Write cmdlets, such as Write-Error.

WriteCommandDetail

For information about the WriteCommandDetails method, see System.Management.Automation.Cmdlet.WriteCommandDetail.

WriteDebug

To provide information that can be used to troubleshoot a function, make the function call the WriteDebug method. The WriteDebug method displays debug messages to the user. For more information, see System.Management.Automation.Cmdlet.WriteDebug.

WriteError

Functions should call this method when non-terminating errors occur and the function is designed to continue processing records. For more information, see System.Management.Automation.Cmdlet.WriteError.

Note

If a terminating error occurs, the function should call the ThrowTerminatingError method.

WriteObject

The WriteObject method allows the function to send an object to the next command in the pipeline. In most cases, WriteObject is the method to use when the function returns data. For more information, see System.Management.Automation.PSCmdlet.WriteObject.

WriteProgress

For functions with actions that take a long time to complete, this method allows the function to call the WriteProgress method so that progress information is displayed. For example, you can display the percent completed. For more information, see System.Management.Automation.PSCmdlet.WriteProgress.

WriteVerbose

To provide detailed information about what the function is doing, make the function call the WriteVerbose method to display verbose messages to the user. By default, verbose messages aren't displayed. For more information, see System.Management.Automation.PSCmdlet.WriteVerbose.

WriteWarning

To provide information about conditions that may cause unexpected results, make the function call the WriteWarning method to display warning messages to the user. By default, warning messages are displayed. For more information, see System.Management.Automation.PSCmdlet.WriteWarning.

Note

You can also display warning messages by configuring the $WarningPreference variable or by using the Verbose and Debug command-line options. for more information about the $WarningPreference variable, see about_Preference_Variables.

Other methods and properties

For information about the other methods and properties that can be accessed through the $PSCmdlet variable, see System.Management.Automation.PSCmdlet.

For example, the ParameterSetName property allows you to see the parameter set that's being used. Parameter sets allow you to create a function that performs different tasks based on the parameters that are specified when the function is run.

See also