Error passing through catch block powershell

Question

I wrote a script to check for the file path for firefox in Windows x64 Host Key. It has a try catch block that has a fully qualified error id in the brackets.

Despite the error message being the same that is in the code, it does not catch the error.

 $program = "FireFox"

 $filepath = "HKLM:\Software\Wow6432Node\Microsoft\Windows\CurrentVersion\Uninstall"

 $FPF = $filepath + "\" + $program 

 try { Get-ChildItem $FPF} 

     catch [PathNotFound,Microsoft.PowerShell.Commands.GetChildItemCommand] {

         if ($_.Exception.Message -match "Get-ChildItem : Cannot find path*") {

             Write-Host "false"}
     }

The full error that comes out is

Get-ChildItem : Cannot find path 'HKEY_LOCAL_MACHINE\Software\Wow6432Node\Microsoft\Windows\CurrentVersion\Uninstall\FireFox' because it does not exist. At line:7 char:7 + try { Get-ChildItem $FPF} + ~~~~~~~~~~~~~~~~~~ + CategoryInfo : ObjectNotFound: (HKEY_LOCAL_MACH...install\FireFox:String) [Get-ChildItem], ItemNotFoundException + FullyQualifiedErrorId :PathNotFound,Microsoft.PowerShell.Commands.GetChildItemCommand

What could could be the issue?

Does your registry path HKEY_LOCAL_MACHINE\Software\Wow6432Node\Microsoft\Windows\CurrentVersion\Uninstall\FireFox really exist? — TobyU
– TobyU, Commented Nov 9, 2018 at 16:59
No it does not. Ultimately, I am going to put this on a script that will install Firefox after that check is done — AJSKRILLA
– AJSKRILLA, Commented Nov 9, 2018 at 17:02
There is a difference between a terminating error (what try is for) and a non terminating error (what get-childitem it doing). either way you should be using if(Test-Path $FPF) for things like this. — Matt
– Matt, Commented Nov 9, 2018 at 17:02
That seemed to do it. Test-Path must have been what I needed to do originally. Thank you — AJSKRILLA
– AJSKRILLA, Commented Nov 9, 2018 at 17:05

mklement0 · Accepted Answer · 2018-11-10 02:49:12Z

As you've discovered, you can bypass the need for your attempt at exception handling by using the Test-Path cmdlet up front to determine whether a given path exists - and if the existence is all you care about, that is enough: if (Test-Path $FPF) { ... }

Generally speaking, however, even if the path exists, you may still run into errors on traversal.

To address your original attempt: there are two fundamental problems:

As Matt points out in a comment, Get-ChildItem not finding a given path results in a non-terminating error, whereas try / catch only applies to terminating errors.
- You can, however, promote non-terminating errors to terminating ones, by adding common parameter -ErrorAction Stop to the command.
- For a summary of PowerShell's error types and their handling, see this GitHub post.
To qualify conditional catch handlers, you must use exception-type literals, e.g., [System.Management.Automation.ItemNotFoundException], not the value of an error record's .FullyQualifiedErrorId property (e.g., PathNotFound,Microsoft.PowerShell.Commands.GetChildItemCommand)
- To determine the full name of an error record's exception type, run the following after the error occurred:
  - $Error[0].Exception.GetType().FullName

Here's an amended version of your code, based on the above:

try {
    Get-ChildItem $FPF -ErrorAction Stop
} catch [System.Management.Automation.ItemNotFoundException] { # input path not found
    Write-Error "Not found: $FPF"
} catch {  # any other error
    Throw "An unexpected error occurred: $_"
}

Collectives™ on Stack Overflow

Error passing through catch block powershell

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related