Richard Siddaway's Blog

How can you check if a PowerShell string contains a character or substring?

You might be tempted to try this:

PS> $s = 'abcdefghijk'
PS> $s -contains 'f'
False

But –contains is for working with the contents of arrays. So you could do this:

PS> ($s.ToCharArray()) -contains 'f'
True

You’re implicitly converting the string ‘f’ to [char] to make the comparison. Your comparison is actually this

PS> ($s.ToCharArray()) -contains [char]'f'
True

That’s fine for a single character but if you want to test a substring

PS> $s -contains 'def'
False
PS> ($s.ToCharArray()) -contains 'def'
False

That approach won’t work.

You need to use the Indexof method

PS> $s.Indexof('f')
5
PS> $s.Indexof('def')
3

The value returned is the position of the FIRST character in the substring.

You can also test an an array of characters

PS> $a = 'g','j','a'
PS> $s.IndexOfAny($a)
0

Again you get the FIRST character in this case the ‘a’

Remember PowerShell is .NET based so the first index is 0

Lets get some repetition into our target

PS> $s = $s * 3

PS> $s
abcdefghijkabcdefghijkabcdefghijk

You also have the option of picking the LAST occurrence of the substring

PS> $s.LastIndexOf('f')
27

PS> $s.LastIndexOfAny($a)
31

This last one is the last ‘j’ in the string – its the last occurrence of any of the characters you wanted to match.

If there isn’t a match you get –1 returned

PS> $s.IndexOf('z')
-1
PS> $s.LastIndexOf('z')
-1

There’s been a Powershell Day UK Agenda Update.

Looks like there’ll be three tracks for most of the day.

Full agenda and tickets from - https://psday.uk/

An interesting problem around deleting folders based on month name.

You have a folder structure that looks like this

C:\Testdata\2011\08-Aug
C:\Testdata\2011\09-Sep
C:\Testdata\2011\10-Oct
C:\Testdata\2011\11-Nov
C:\Testdata\2011\12-Dec
C:\Testdata\2012\01-Jan
C:\Testdata\2012\02-Feb
C:\Testdata\2012\03-Mar
C:\Testdata\2012\04-Apr
C:\Testdata\2012\05-May
C:\Testdata\2012\06-Jun
C:\Testdata\2012\07-Jul
C:\Testdata\2012\08-Aug
C:\Testdata\2012\09-Sep
C:\Testdata\2012\10-Oct
C:\Testdata\2012\11-Nov
C:\Testdata\2012\12-Dec
C:\Testdata\2013\01-Jan
C:\Testdata\2013\02-Feb
C:\Testdata\2013\03-Mar
C:\Testdata\2013\04-Apr
C:\Testdata\2013\05-May
C:\Testdata\2013\06-Jun
C:\Testdata\2013\07-Jul
C:\Testdata\2013\08-Aug

The structure extends forward and backward in time. Each year folder has 12 subfolders based on the number and name of the month.

You want to delete all folder, subfolders and contents that are more than 7 years old BUT you have to use the folder name for month and its parent for the year. You can’t assume that the create date or last write date will work in this instance.

If you want to paly around with this the code to create the folder structure looks like this:

##
## define enum for months
##
enum month {
Jan = 1
Feb = 2
Mar = 3
Apr = 4
May = 5
Jun = 6
Jul = 7
Aug = 8
Sep = 9
Oct = 10
Nov = 11
Dec = 12
}

##
## create folder structure
##

New-Item -Path c:\ -Name Testdata -ItemType Directory

2018..2005 |
foreach {
New-Item -Path c:\Testdata -Name $psitem -ItemType Directory

$path = "c:\Testdata\$psitem"

1..12 | 
foreach {
$name = "{0:00}-{1}" -f $psitem, [month]$psitem
New-Item -Path $path -Name $name -ItemType Directory
}
}

Create a top level folder. Then for the years 2018 back to 2005 create a year folder using the year as the name. In each year folder create 12 month folders. I’ve used the string formatting –f operator to put a leading zero on the month number and to get the appropriate value from the enum.

Now to delete the folders

enum month {
Jan = 1
Feb = 2
Mar = 3
Apr = 4
May = 5
Jun = 6
Jul = 7
Aug = 8
Sep = 9
Oct = 10
Nov = 11
Dec = 12
}

$date = Get-Date
$year = $date.Year - 8

##
## delete everything 8 years or older
##
Get-ChildItem -Path C:\Testdata -Directory |
where Name -le $year |
foreach {
Remove-Item -Path $psitem.Fullname -Recurse -Force -Confirm:$false
}

##
## if Month -ne January
## need to delete some months
##

if ($date.Month -gt 1){
$path = "C:\testdata\$($year+1)"
$month = $date.Month -1

1..$month | 
foreach {
$mpath = "$path\{0:00}-{1}" -f $psitem, [month]$psitem
Remove-Item -Path $mpath -Recurse -Force -Confirm:$false
}
}

The code gets the current date and gets the year minus 8. It loops through your top level folder and gets the folders that are less than or equal to the year you've defined. They, and their contents, are force deleted. Only thing that could stop the deletion is if you have one of the files pinned open.

If the current month is January there's nothing else to do. Otherwise create the path to the -7 year folder and calculate the last month you want to delete. Loop through the months, build the path and force the deletion of the folder and its contents.

The bulk of the work is done at the year level with a quick clean up of the months.

Usually when you create a variable you implicitly set the type by the value you use. Sometimes though you may want to explicitly set the variable type.

if you don’t give the variable a type you can do this:

PS> $x = 35
PS> $x
35
PS> $x = 'now a string'
PS> $x
now a string

If you give the variable an explicit type

PS> [int]$x = 35
PS> $x
35
PS> $x = 'now a string'
Cannot convert value "now a string" to type "System.Int32". Error: "Input string was not in a correct format."
At line:1 char:1
+ $x = 'now a string'
+ ~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : MetadataError: (:) [], ArgumentTransformationMetadataException
+ FullyQualifiedErrorId : RuntimeException

The variable expects an integer or something that can be converted to an integer

PS> $x = '123'
PS> $x
123

You can’t supply a type when you use New-Variable so if you need a read only or constant variable then create it as shown above then use Set-Variable to make it read only or constant.

When you create a variable you usually want to be able to change its value – the clue is in the term variable. Sometimes though you might want to use a value that doesn’t change – you want read only and constant variables.

Standard variables are changeable:

PS> $x = 10
PS> $x
10
PS> $x += 5
PS> $x
15
PS> $x -= 3
PS> $x
12

The variable holds a value that can be changed to match your needs. Another example is the variable used in a for loop:

PS> for ($i=1; $i -lt 6; $i++){$i}
1
2
3
4
5

If you want to use a value in your code that is pre-set and doesn’t change you have a couple of options – read only variables and constants.

To create a read only variable use New-variable with the ReadOnly option

PS> New-Variable -Name x -Value 10 -Option ReadOnly
PS> $x
10

Now if you try to change the value

PS> $x += 5
Cannot overwrite variable x because it is read-only or constant.
At line:1 char:1
+ $x += 5
+ ~~~~~~~
+ CategoryInfo : WriteError: (x:String) [], SessionStateUnauthorizedAccessException
+ FullyQualifiedErrorId : VariableNotWritable

You can also change an existing variable to be read only

PS> $y = 20
PS> Set-Variable -Name y -Option ReadOnly
PS> $y
20
PS> $y -= 5
Cannot overwrite variable y because it is read-only or constant.
At line:1 char:1
+ $y -= 5
+ ~~~~~~~
+ CategoryInfo : WriteError: (y:String) [], SessionStateUnauthorizedAccessException
+ FullyQualifiedErrorId : VariableNotWritable

A read only variable is protected from deletion

PS> Remove-Variable -Name x
Remove-Variable : Cannot remove variable x because it is constant or read-only. If the variable is read-only, try the
operation again specifying the Force option.
At line:1 char:1
+ Remove-Variable -Name x
+ ~~~~~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : WriteError: (x:String) [Remove-Variable], SessionStateUnauthorizedAccessException
+ FullyQualifiedErrorId : VariableNotRemovable,Microsoft.PowerShell.Commands.RemoveVariableCommand

Time to be a PowerShell Jedi and use the force:

PS> Remove-Variable -Name x -Force

the variable is removed.

But what if you want to prevent the variable from being removed. Then you have to make it a constant:

PS> New-Variable -Name x -Value 10 -Option Constant
PS> $x
10
PS> $x += 5
Cannot overwrite variable x because it is read-only or constant.
At line:1 char:1
+ $x += 5
+ ~~~~~~~
+ CategoryInfo : WriteError: (x:String) [], SessionStateUnauthorizedAccessException
+ FullyQualifiedErrorId : VariableNotWritable

PS> Remove-Variable -Name x
Remove-Variable : Cannot remove variable x because it is constant or read-only. If the variable is read-only, try the
operation again specifying the Force option.
At line:1 char:1
+ Remove-Variable -Name x
+ ~~~~~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : WriteError: (x:String) [Remove-Variable], SessionStateUnauthorizedAccessException
+ FullyQualifiedErrorId : VariableNotRemovable,Microsoft.PowerShell.Commands.RemoveVariableCommand

PS> Remove-Variable -Name x -Force
Remove-Variable : Cannot remove variable x because it is constant or read-only. If the variable is read-only, try the
operation again specifying the Force option.
At line:1 char:1
+ Remove-Variable -Name x -Force
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : WriteError: (x:String) [Remove-Variable], SessionStateUnauthorizedAccessException
+ FullyQualifiedErrorId : VariableNotRemovable,Microsoft.PowerShell.Commands.RemoveVariableCommand

You can’t modify the variable’s value and you can’t delete it.

The variable will disappear when you close the PowerShell console.

Splitting strings is usually easy – you use the –split operator or the Split() method. Occasionally you may hit a problem – for instance splitting on a \ character.

Let me demonstrate with an example.

PS> Get-CimInstance -ClassName Win32_ComputerSystem | select Username

Username
--------
W510W10\Richard

I get the domain (in this case the machine as its a local account) and the user name. What I want is just the user name.

PS> (Get-CimInstance -ClassName Win32_ComputerSystem | select -ExpandProperty Username) -split '\'
parsing "\" - Illegal \ at end of pattern.
At line:1 char:1
+ (Get-CimInstance -ClassName Win32_ComputerSystem | select -ExpandProp ...
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : OperationStopped: (:) [], ArgumentException
+ FullyQualifiedErrorId : System.ArgumentException

I’ve seen possible solutions like this

(Get-CimInstance -ClassName Win32_ComputerSystem | select username | out-string).split('\')

But Out-String isn’t necessary in this case.

The problem is that \ is the PowerShell escape character so you have to escape it so that its treated as a literal character

PS> ((Get-CimInstance -ClassName Win32_ComputerSystem | select -ExpandProperty Username) -split '\\')[1]
Richard

This also works

PS> (Get-CimInstance -ClassName Win32_ComputerSystem | select -ExpandProperty Username).Split('\')[1]
Richard

Keep it simple