PowerShell Scripting Cheatsheet
This is the companion to PowerShell Cheatsheet, which focuses on writing PowerShell scripts.
According to Microsoft PowerShell
Is a cross-platform task automation and configuration management framework, consisting of a command-line shell and
scripting language that is built on top of the .NET Common Language Runtime (CLR), accepts and returns .NET objects.
This brings entirely new tools and methods for automation.
This means learning new skills and thinking differently which can be frustrating while learning.
To simplify maintenance I write PowerShell scripts as standalone utilities deployed in a single file, this means I have to copy-and-paste my favourite frequently used functions, such as dumpArrayList, dumpHashTable because there is no mechanism to textually include your favourite functions into the source when writing and testing.
It is possible to split your script into multiple files, create libraries of your favourite utilities etc.
I do not cover this topic, the example script shows where/how to source you library files, and if you wish to create your
own modules, see How to Write a PowerShell Script Module.
Introduction
PowerShell is very powerful scripting language, often used to automate routine tasks, which makes it an ideal
target for would-be hackers. To mitigate this Microsoft limits PowerShell execution, even though the
individual cmdlets can always be run.
If your Get-ExecutionPolicy is like this
PS> Get-ExecutionPolicy
Restricted
PS> Get-ExecutionPolicy -List
Scope ExecutionPolicy
----- ---------------
MachinePolicy Undefined
UserPolicy Undefined
Process Undefined
CurrentUser Undefined
LocalMachine Restricted
The PowerShell script will not execute!
PS1> .\hello-world.ps1
.\hello-world.ps1 : File C:\Users\sjfke\hello-world.ps1 cannot be loaded because running scripts is disabled on this
system. For more information, see about_Execution_Policies at https:/go.microsoft.com/fwlink/?LinkID=135170.
At line:1 char:1
+ .\hello-world.ps1
+ ~~~~~~~~~~~~~~~~~
+ CategoryInfo : SecurityError: (:) [], PSSecurityException
+ FullyQualifiedErrorId : UnauthorizedAccess
Many articles on the internet suggest the disabling or trying to work around this security feature… PLEASE AVOID DOING THIS!
Many Windows distributions provide Developer section in Settings, which allows local PowerShell scripts to be
executed by the CurrentUser by setting the ExecutionPolicy to RemoteSigned.
Alternatively run a PowerShell as Administrator set the following, choosing the [A] option.
ADM-PS1> set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser
Execution Policy Change
The execution policy helps protect you from scripts that you do not trust. Changing the execution policy might expose you to the security risks
described in the about_Execution_Policies help topic at https:/go.microsoft.com/fwlink/?LinkID=135170. Do you want to change the execution policy?
[Y] Yes [A] Yes to All [N] No [L] No to All [S] Suspend [?] Help (default is "N"): A
Suggested Laptop settings
Locally developed PowerShell scripts will be executed but those from any other source will need to be signed.
PS C:\WINDOWS\system32> Get-ExecutionPolicy -List
Scope ExecutionPolicy
----- ---------------
MachinePolicy Undefined
UserPolicy Undefined
Process Undefined
CurrentUser RemoteSigned
LocalMachine Undefined
PS C:\WINDOWS\system32> Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser
Suggested Server settings
All PowerShell scripts should be signed, if too restrictive for your environment use
Suggested Laptop settings
PS C:\WINDOWS\system32> Get-ExecutionPolicy -List
Scope ExecutionPolicy
----- ---------------
MachinePolicy Undefined
UserPolicy Undefined
Process Undefined
CurrentUser AllSigned
LocalMachine AllSigned
PS C:\WINDOWS\system32> Set-ExecutionPolicy -ExecutionPolicy AllSigned -Scope CurrentUser
PS C:\WINDOWS\system32> Set-ExecutionPolicy -ExecutionPolicy AllSigned -Scope LocalMachine
Language Overview
The language makes use of .Net Framework and is built on top of the .NET Common Language Runtime (CLR) , and manipulates .NET objects. If the language itself does not provide what you need, there may be a Popular PowerShell Module you can download or you can access the .Net APIs directly, a good example being ArrayLists which are dynamic in size unlike a PowerShell Array.
In common with other object oriented languages, PowerShell has features such inheritance, subclasses, getters, setters, modules etc.
Functions and methods support both named and positional arguments, that can be mixed liberally, but done inappropriately can make the intention
more confusing. Often it is clearer to use splatting rather
than individual name or positional parameters.
Useful starting points when learning about the language:
Unlike most texts on programming languages, let us starts with a simple but realistic PowerShell example, with many of the language details being covered in subsequent sections.
Example Script
This is a contrived but realistic PowerShell script to illustrate several important points. It is based on a gist template from 9to5IT, which is augmented to force the syntax version and to be more strict on the use of uninitialized variables.
#requires -version 4
<#
.SYNOPSIS
9to5IT Template for PowerShell scripts.
.DESCRIPTION
Displays the names and ages of the flintstones.
.PARAMETER names
List the names only
.PARAMETER ages
List the ages only
.PARAMETER person <name>
List person's age
.INPUTS
None
.OUTPUTS
The Requested text.
.NOTES
Version: 1.0
Author: sjfke
Creation Date: 2021.01.03
Purpose/Change: Initial script development
.EXAMPLE
families.ps1 -names
.EXAMPLE
families.ps1 -person fred
#>
param(
[switch]$names = $false,
[switch]$ages = $false,
[string]$person = $null,
[switch]$stackTrace = $false
)
Set-StrictMode -Version 2
#---------------------------------------------------------[Initialisations]--------------------------------------------------------
# Set Error Action to Silently Continue
# $ErrorActionPreference = "SilentlyContinue"
# Dot Source required Function Libraries
# . "C:\Scripts\Functions\Logging_Functions.ps1"
#----------------------------------------------------------[Declarations]----------------------------------------------------------
$scriptName = "flintstones.ps1"
$scriptVersion = "1.0"
#Log File Info
# $sLogPath = "C:\Windows\Temp"
# $sLogName = "<script_name>.log"
# $sLogFile = Join-Path -Path $sLogPath -ChildPath $sLogName
$hash = $null
#-----------------------------------------------------------[Functions]------------------------------------------------------------
function initializeHash {
return @{ Fred = 30; Wilma = 25; Pebbles = 1; Dino = 5 }
}
function getNames {
return $hash.keys
}
function getAges {
return $hash.values
}
function getPerson {
param(
[string]$name = ''
)
return $hash[$name]
}
#-----------------------------------------------------------[Execution]------------------------------------------------------------
$hash = initializeHash
if ($names) {
getNames
}
elseif ($ages) {
getAges
}
elseif (($person -ne '') -and ($person -ne $null)) {
$arguments = @{
name = $person
}
getPerson @arguments
}
else {
if ($stackTrace) {
write-error("invalid or missing argument") # stack-trace like error message
}
else {
write-warning("{0} v{1}: invalid or missing argument" -f $scriptName, $scriptVersion)
exit(1)
}
}
Things to note:
The #requires -version 4 PowerShell version 4 syntax, (use version 2, if windows is very old);
Initial comment block
.SYNOPSIS...provides theget-helptext, note line-spacing is important;The param() block must be the first non-comment line for command-line arguments;
The Set-StrictMode -Version 2 checks the usage of uninitialized variables;
Language Keypoints
Variables
Powershell variables can be any of the Basic DataTypes such as integers, characters, strings, arrays, and hash-tables, but also .Net objects that represent such things as
processes, services, event-logs, and even computers.
PS> $age = 5 # System.Int32
PS> [int]$age = "5" # System.Int32, cast System.String + System.Int32
PS> $name = "Dino" # System.String
PS> $name + $age # Fails; System.String + System.Int32
PS> $name + [string]$age # Dino5; System.String + System.String
PS> $a = (5, 30, 25, 1) # array of System.Int32
PS> $a = (5, "Dino") # array of (System.Int32, System.String)
PS> $h = @{ Fred = 30; Wilma = 25; Pebbles = 1; Dino = 5 } # hash table
PS> $d = Get-ChildItem C:\Windows # directory listing, FileInfo and DirectoryInfo types,
PS> $d | get-member # FileInfo, DirectoryInfo Properties and Methods
PS> $p = Get-Process # System.Diagnostics.Process type
PS> set-variable -name age 5 # same as $age = 5
PS> set-variable -name name Dino # same as $name = "Dino" (variable's name is *name*)
PS> clear-variable -name age # clear $age; $age = $null
PS> clear-variable -name name # clear $name; $name = $null
PS> remove-variable -name age # delete variable $age
PS> remove-item -path variable:\name # delete variable $name
PS> set-variable -name pi -option Constant 3.14159 # constant variable
PS> $pi = 42 # Fails $pi is a constant
Basic DataTypes
Data Type |
Definition |
|---|---|
Boolean |
True or False Condition |
Byte |
An 8-bit unsigned whole number from 0 to 255 |
Char |
A 16-bit unsigned whole number from 0 to 65,535 |
Date |
A calendar date |
Decimal |
A 128-bit decimal value, such as 3.14159 |
Double |
A double-precision 64-bit floating point number, narrower range than Decimal |
Integer |
A 32-bit signed whole number from -2,147,483,648 to 2,147,483,647 |
Long |
A 64-bit signed whole number, very big integer, 9,233,372,036,854,775,807 |
Object |
|
Short |
A 16-bit unsigned whole number, -32,768 to 32,767 |
Single |
A single-precision 32-bit floating point number |
String |
Text, a character string |
Array Variables
Array variables are a fixed size, can have mixed values and can be multi-dimensional.
PS> $a = 1, 2, 3 # array of integers
PS> $a = (1, 2, 3) # array of integers (my personal preference)
PS> $a = ('a','b','c')
PS> $a = (1, 2, 3, 'x') # array of System.Int32's, System.String
PS> [int[]]$a = (1, 2, 3, 'x') # will fail 'x', array of System.Int32 only
PS> $a = ('fred','wilma','pebbles')
PS> $a[0] # fred
PS> $[2] # pebbles
PS> $a.length # 3
PS> $a[0] = 'freddie' # fred becomes freddie
PS> $a[3] = 'dino' # Error: Index was outside the bounds of the array.
PS> $a += 'dino' # correct way to add 'dino' (note does an array copy)
PS> $a[1,3,2] # wilma, dino, pebbles
PS> $a[1..3] # wilma, pebbles, dino
PS> $a = $a[0..2] # dino ran away (note does an array copy)
PS> $b = ('barbey', 'betty', 'bamm-bamm')
PS> $a = ($a, $b) # [0]:fred [1]:wilma [2]:pebbles [3]:barney [4]:betty [5]:bamm-bamm
PS> $a.length # 6
PS> $a = ($a, ($b)) # [0]:fred [1]:wilma [2]:pebbles [3][0]:barney [3][1]:betty [3][2]:bamm-bamm
PS> $a.length # 4
PS> $ages = (30, 25, 1, 5) # flintstones ages
PS> $names = ('fred','wilma','pebbles', 'dino') # flintstones names
PS> $a = ($names),($ages)) # multi-dimensional array example
PS> $a.length # 4
PS> $a[0] # fred wilma pebbles dino
PS> $a[1] # 30 25 1 5
PS> $a[0][0] # fred
PS> $a[0][1] # 30
Useful references:
HashTables
A HashTable is an unordered collection of key:value pairs, synonymous with an object and its properties.
Later versions support known/fixed order hash elements, $hash = [ordered]@{}.
PS> $h = @{} # empty hash
PS> $key = 'Fred' # set key name
PS> $value = 30 # set key value
PS> $h.add($key, $value) # add key:value ('fred':30) to the hash-table
PS> $h.add('Wilma', 25 ) # add 'Wilma':25
PS> $h['Pebbles'] = 1 # add 'Pebbles':1
PS> $h.Dino = 5 # add 'Dino':5
PS> $h # actual hash-table, printed if on command-line
PS> $h['Fred'] # how old is Fred? 30
PS> $h[$key] # how old is Fred? 30
PS> $h.fred # how old is Fred? 30
# creating a populated hash, multi-line.
PS> $h = @{
Fred = 30
Wilma = 25
Pebbles = 1
Dino = 5
}
# creating the same populated hash, on single-line
PS> $h = @{ Fred = 30; Wilma = 25; Pebbles = 1; Dino = 5 }
PS> $h.keys # unordered: Dino, Pebbles, Fred, Wilma
PS> $h.values # unordered: 5, 1, 30, 25 (but same as $h.keys order)
# later PowerShell versions allow the order to be fixed.
PS> $h = [ordered]@{ Fred = 30; Wilma = 25; Pebbles = 1; Dino = 5 }
PS> $h.keys # ordered: Fred, Wilma, Pebbles, Dino
PS> $h.values # ordered: 30, 25, 1, 5
# key order is random, unless [ordered] was used in the declaration
PS> foreach ($key in $h.keys) {
write-output ('{0} Flintstone is {1:D} years old' -f $key, $h[$key])
}
# ascending alphabetic order (Dino, Fred, Pebbles, Wilma)
PS> foreach ($key in $h.keys | sort) {
write-output ('{0} Flintstone is {1:D} years old' -f $key, $h[$key])
}
# descending alphabetic order (Wilma, Pebbles, Fred, Dino)
PS> foreach ($key in $h.keys | sort -descending) {
write-output ('{0} Flintstone is {1:D} years old' -f $key, $h[$key])
}
# specfific order (Fred, Wilma, Pebbles, Dino)
PS> $keys = ('fred', 'wilma', 'pebbles', 'dino')
for ($i = 0; $i -lt $keys.length; $i++) {
write-output ('{0} Flintstone is {1:D} years old' -f $keys[$i], $h[$keys[$i]])
}
PS> if ($h.ContainsKey('fred')) { ... } # true
PS> if ($h.ContainsKey('barney')) { ... } # false
PS> if ($h.fred) { ... } # avoid, works most of the time.
PS> if ($h['barney']) { ... } # avoid, works most of the time.
PS> $h.remove('Dino') # remove Dino, because he ran away :-)
PS> $h.clear() # flintstone family deceased
For more details read the excellent review by Kevin Marquette:
Objects
If you cannot create what you need from Arrays, HashTables, ArrayLists, Queues, Stacks etc., then it is possible to create custom PowerShell objects, but to date I have never needed to do this. For more details, read:
Functions
Function arguments and responses are passed by reference, so an argument can be changed inside the function and remains unchanged outside the function, but this is considered “bad programming practice”, so better to avoid doing this. Functions return references to objects, as illustrated in the Example Script where references to HashTable and Array objects are returned.
While each function call returns a reference to a new (different) object, be careful about the scope of the variable you assign this reference too, it is easy to create multiple references to the same object.
While mixing named (order independent) and positional (order dependent) arguments is permitted it can cause strange errors, so unless you are only
supplying one or two arguments, a better approach is to use splatting.
The following contrived example illustrates the basics but the param ( ... ) section has many options not shown here.
#requires -version 4
Set-StrictMode -Version 2
function createPerson {
param (
[string]$name = '',
[int]$age = 0,
[switch]$verbose = $false,
[switch]$debug = $false
)
if (($name -eq $null) -or ($name.length -eq 0)) {
if ($verbose) {
write-warning("createPerson - name is missing")
return $null
}
elseif ($debug) {
write-error("createPerson - name is missing")
exit(1)
}
else {
return $null
}
}
if (($age -le 0) -or ($age -gt 130)) {
if ($verbose) {
write-warning("createPerson - age, {0:D}, is incorrect" -f $age)
return $null
}
elseif ($debug) {
write-error("createPerson - age, {0:D}, is incorrect" -f $age)
exit(1)
}
else {
return $null
}
}
$hash = @{}
$hash[$name] = $age
return $hash
}
createPerson 'fred' 30 -verbose # positional arguments
createPerson 30 'fred' -verbose # positional arguments, breaks name=30
createPerson -name 'fred' -age 30 -verbose # named arguments
createPerson -age 30 'fred' -verbose # mixed arguments, be careful, no-named taken param order
$arguments = @{ # splatting
name = 'fred'
age = 30
verbose = $true
}
createPerson @arguments
$arguments = @{name = 'wilma'; age = 25; verbose = $true} # splatting one-line
createPerson @arguments
$arguments = @{
name = 'fred'
verbose = $true
debug = $false
}
createPerson @arguments # fails, age default is 0
$arguments = @{
age = 21
verbose = $true
debug = $false
}
createPerson @arguments # fails, name default is an empty string
Further reading:
Microsoft Docs, Chapter 9 - Functions
Microsoft Docs, About Functions Advanced Parameters.
ArrayList
PS> $names = New-Object -TypeName System.Collections.ArrayList
PS> $names = [System.Collections.ArrayList]::new()
PS> $names.gettype() # ArrayList
PS> $index = $names.Add('fred') # returns array-list index: i.e. 0
PS> [void]$names.Add('wilma') # discard array-list index
PS> [void]$names.Add('pebbles')
PS> [void]$names.Add('dino')
# one-line creation, empty or populated
PS> [System.Collections.ArrayList]$names = @()
PS> [System.Collections.ArrayList]$names = @('fred','wilma','pebbles', 'dino')
PS> $names.Count # returns 4
PS> $names[1] # wilma
PS> $names.remove(3) # dino ran away or did he?
PS> $names.Count # 4, no dino is still there
PS> $names.[3] # dino
PS> $names.RemoveAt(3) # dino, has really gone this time
PS> [void]$names.Add('dino') # dino found
PS> $names.Remove('dino') # dino, escaped again
PS> [void]$names.Add('dino') # dino found ... again
PS> 'fred' -in $names # True (not supported in PowerShell 2)
PS> 'barney' -in $names # False (not supported in PowerShell 2)
PS> $names -contains 'fred' # True
PS> $names -contains 'barney' # False
PS> [void]$names.Insert(3,'fido')
PS> $names # 0:fred, 1:wilma, 2:pebbles, 3:fido, 4:dino
PS> $names.remove('fido')
PS> $names # 0:fred, 1:wilma, 2:pebbles, 3:dino
# Generic List are ArrayList's of a fixed type
PS> [System.Collections.Generic.List[string]]$names = @()
PS> [System.Collections.Generic.List[string]]$names = @('fred','wilma','pebbles', 'dino')
PS> [System.Collections.Generic.List[int]]$ages = @()
PS> [System.Collections.Generic.List[int]]$ages = (30, 25, 1, 5)
$names.add(30) # 0:fred, 1:wilma, 2:pebbles, 3:dino, 4:30
$ages.add('fred') # fails, throws conversion exception
Further reading:
IF/Switch commands
The conditions that can be tested in an if statement are very extensive:
Equality/inequality:
-eq|-ieq|-ceq / -ne|-ine|-cne;Greater/less than:
-gt|-igt|-cgt|-ge|-ige / -lt|-ilt|-clt|-le|-ile|-cle;Wildcard:
-like|-ilike|-clike|-notlike|-inotlike|-cnotlike;Regular Expressions:
-match|-imatch|-cmatch|-notmatch|-inotmatch|-cnotmatch;Object type check:
-is|-isnot;Array <op> value:
-contains|-icontains|-ccontains|-notcontains|-inotcontains|-cnotcontains;Value <op> array:
-in|-iin|-cin|-notin|-inotin|-cnotinLogical operators:
-not|!|-and|-or|-xorBitwise operators:
-band|-bor|-bxor|-bnot|-shl|-shr;PowerShell expressions:
Test-Path|Get-Process;PowerShell pipeline:
(Get-Process | Where Name -eq Notepad);Null checking:
($null -eq $value);
There is also a switch statement for comparing against multiple values.
#requires -version 2
Set-StrictMode -Version 2
$apple = 10
$pear = 20
if ( $apple -gt $pear ) {
write-host('apple is higher than pear')
}
elseif ( $apple -lt $pear ) {
write-host('apple is lower than pear')
}
else {
write-host('apple and pear are equal')
}
$path = 'file.txt'
$alternatePath = 'folder1'
if ( Test-Path -Path $path -PathType Leaf ) {
Move-Item -Path $path -Destination $alternatePath
}
elseif ( Test-Path -Path $path ) {
Write-Warning "A file is required but a folder was given."
}
else {
Write-Warning "$path could not be found."
}
$fruit = 10
switch ( $fruit ) {
10 {
write-host('fruit is an apple')
}
20 {
write-host('fruit is an apple')
}
Default {
write-host('unknown fruit')
}
}
Further reading:
Try/Catch
Exception handling uses Try/Catch, but the Catch block is only invoked on terminating errors.
#requires -version 4
Set-StrictMode -Version 2
$error.clear()
# $Error is an array of recent errors, index 0 being the latest
# $Error[0] | get-member # what does an error return
# $Error[0].tostring() # error text message
# $Error[0].Exception | get-member # method, properties of the exception
# $Error[0].Exception.GetType().FullName # how to catch-it :-)
$cwd = get-childitem variable:pwd
$filename = 'cannot-readme.txt'
$path = Join-Path -path $cwd.value -childpath $filename
try {
$content = get-content -path $path -ErrorAction Stop
}
catch [System.Management.Automation.ItemNotFoundException] {
write-warning $Error[0].ToString()
exit(1)
}
catch {
write-warning $Error[0].ToString()
write-warning $Error[0].Exception.GetType().FullName # exception message type
exit(1)
}
finally {
write-warning("Resetting the Error Array")
$error.clear()
}
write-host("Fetched the content of {0}" -f $path)
exit(0)
Note the following two points in the example:
Addition of
-ErrorAction Stoptoget-contentto make it a terminating error;The
finallyblock is always executed, whether an exception is thrown or not!
Further reading:
Loops
There are several loop constructirs for, foreach, while and do .. while.
#requires -version 4
Set-StrictMode -Version 2
$names = ('Fred', 'Wilma', 'Pebbles', 'Dino')
for ($index = 0; $index -lt $names.length; $index++) {
write-host ('{0} Flintstone' -f $names[$index])
}
# Index often written as $i, $j, $k
for ($i = 0; $i -lt $names.length; $i++) {
write-host ('{0} Flintstone' -f $names[$i])
}
foreach ($name in $names) {
write-host ('{0} Flintstone' -f $name)
}
$hash = @{ Fred = 30; Wilma = 25; Pebbles = 1; Dino = 5 }
foreach ($key in $hash.keys) {
write-host ('{0} Flintstone is {1:D} years old' -f $key, $hash[$key])
}
$index = 0;
while ($index -lt $names.length){
write-host ('{0} Flintstone' -f $names[$index])
$index += 1
}
$index = 0;
do {
write-host ('{0} Flintstone' -f $names[$index])
$index += 1
} while($index -lt $names.length)
Operators
PowerShell supports the almost all the common programming language operators, with parenthesis to alter operator precedence.
#requires -version 4
Set-StrictMode -Version 2
$a = 20
$b = 10
$c = 2
# Arithmetic
$a + $b + $c # addition = 32
$a - $b - $c # subtraction = 8
$a - $b + $c # subtraction, addition = 12
$a + $b - $c # addition, subtraction = 28
$a * $b * $c # multiplication = 400
$a + $b * $c # addition, multiplication = 40
$a * $b + $c # multiplication, addition = 202
$a * ($b + $c) # multiplication, addition = 240
$a / $b / $c # division = 1
$a + $b / $c # addition, division = 15
$a / $b + $c # division, addition = 4
$a / ($b + $c) # division, addition = 1.66666666666667
$a % $b # modulus = 0
$b % $a # modulus = 10
$c % $b # modulus = 2
# Comparison
$a -eq $b # equals = False
$a -ne $b # not equals = True
$a -gt $b # greater than = True
$a -ge $a # greater than or equal = True
$a -lt $b # less than = False
$a -le $a # less than or equal = True
# Assignment
$d = $a + $b # assignment = 30
$d += $c # addition, assignment = 32
$d -= $c # subtraction, assiginment = 30
$a = $true
$b = $false
# Logical
$a -and $b # and = False
$a -or $b # or = True
-not $a # not = False
-not $a -and $b # not, and = False
$a -and -not $b # and, not = True
Backtick Operator
The ` is used for line continuation and to identify a “tab” and “new line” character.
Word-wrap operator
`Newline
`nTab
`t
Regular Expressions
PowerShell supports regular expressions in much the same was as Perl or Python.
Table taken from TutorialsPoint.com - Regular Expression
Subquery |
Match description |
|---|---|
^ |
The beginning of the line. |
$ |
The end of the line. |
. |
Any single character except newline. Using m option it to matches the newline as well. |
[…] |
Any single character in brackets. |
[^…] |
Any single character not in brackets. |
\A |
Beginning of the entire string. |
\z |
End of the entire string. |
\Z |
End of the entire string except allowable final line terminator. |
re* |
0 or more occurrences of the preceding expression. |
re+ |
1 or more of the previous thing. |
re? |
0 or 1 occurrence of the preceding expression. |
re{ n} |
Exactly n number of occurrences of the preceding expression. |
re{ n,} |
n or more occurrences of the preceding expression. |
re{ n, m} |
At least n and at most m occurrences of the preceding expression. |
a¦b |
Either a or b. |
(re) |
Groups regular expressions and remembers the matched text. |
(?: re) |
Groups regular expressions without remembering the matched text. |
(?> re) |
Matches the independent pattern without backtracking. |
\w |
The word characters. |
\W |
The nonword characters. |
\s |
The whitespace. Equivalent to [tnrf]. |
\S |
The nonwhitespace. |
\d |
The digits. Equivalent to [0-9]. |
\D |
The nondigits. |
\A |
The beginning of the string. |
\Z |
The end of the string. If a newline exists, it matches just before newline. |
\z |
The end of the string. |
\G |
The point where the last match finished. |
\n |
Back-reference to capture group number “n”. |
\b |
The word boundaries. Matches the backspace (0x08) when inside the brackets. |
\B |
The nonword boundaries. |
\n,\t,\r |
Newlines, carriage returns, tabs, etc. |
\Q |
Escape (quote) all characters up to E. |
\E |
Ends quoting begun with Q. |
Examples
#requires -version 4
Set-StrictMode -Version 2
"fred" -match "f..d" # True (same as imatch)
"fred" -imatch "F..d" # True
"fred" -cmatch "F..d" # False
"fred" -notmatch "W..ma" # True
"fred" -match "re" # (match 're') True
"dog" -match "d[iou]g" # (dig, dug) True
"ant" -match "[a-e]nt" # (bnt, cnt, dnt, ent) True
"ant" -match "[^brt]nt" # True
"fred" -match "^fr" # (starts with 'fr') True
"fred" -match "ed$" # (ends with 'ed') True
"doggy" -match "g*" # True
"doggy" -match "g?" # True
"Fred Flintstone" -match "\w+" # (matches word Fred) True
"FredFlintstone" -match "\w+" # (matches word Fred) True
"Fred Flintstone" -match "\W+" # (matches >= 1 non-word) True
"FredFlintstone" -match "\W+" # (matches >= 1 non-word) False
"Fred Flintstone" -match "\s+" # (matches >= 1 white-space) True
"FredFlintstone" -match "\s+" # (matches >= 1 white-space) False
"Fred Flintstone" -match "\S+" # (matches >= 1 non white-space) True
"FredFlintstone" -match "\S+" # (matches >= 1 non white-space) True
"Fred Flintstone" -match "\d+" # (matches >= 1 digit 0..9) False
"Fred is 30" -match "\d+" # (matches >= 1 digit 0..9) True
"Fred Flintstone" -match "\D+" # (matches >= 1 non-digit 0..9) True
"Fred is 30" -match "\D+" # (matches >= 1 non-digit 0..9) True
"Fred Flintstone" -match "\w?" # (match >= 0 preceding pattern) True
"Fred Flintstone" -match "\w{2}" # (match 2 preceding pattern) True
"Fred Flintstone" -match "\W{2}" # (match 2 preceding pattern) False
"Fred Flintstone" -match "\w{2,}" # (match >2 preceding pattern) True
"Fred Flintstone" -match "\W{2,}" # (match >2 preceding pattern) False
"Fred Flintstone" -match "\w{2,3}" # (match >2 <=3 preceding pattern) True
"Fred Flintstone" -match "\W{2,3}" # (match >2 <=3 preceding pattern) False
'Fred Flinstone' -replace '(\w+) (\w+)', 'Wilma $2' # Wilma Flinstone
'fred Flinstone' -ireplace 'Fred (\w+)', 'Wilma $1' # Wilma Flinstone
'fred Flinstone' -replace 'Fred (\w+)', 'Wilma $1' # Wilma Flinstone
'fred Flinstone' -creplace 'Fred (\w+)', 'Wilma $1' # fred Flinstone
Entire technical books are dedicated to Regular Expressions, the above is very brief. For more details see:
Typical Usage Examples
Reading Files
Simple example, with the filename specified in the script.
#requires -version 4
Set-StrictMode -Version 2
$filename = 'file.txt'
$addCWD = $false
$path = $filename
if ($addCWD) {
$path = Join-Path -path $cwd.value -childpath $filename
}
write-host("if...then...else")
if (-not (Test-Path -path $path -pathtype leaf) ) {
write-warning("Filename, {0}, does not exist" -f $path)
exit(1)
}
else {
$count = 1
foreach ($line in get-content $path) {
write-host("{0:D3}:{1}" -f $count, $line)
$count += 1
}
$fh = get-childitem $path # get file attributes
}
write-host("try...catch")
try {
$count = 1
foreach ($line in get-content $path -ErrorAction Stop) {
write-host("{0:D3}:{1}" -f $count, $line)
$count += 1
}
$fh = get-childitem $path # get file attributes
}
catch {
write-warning $Error[0].ToString()
write-warning $Error[0].Exception.GetType().FullName # exception message type
exit(1)
}
exit(0)
If the filename(s) are supplied on the command line, then globbing (file pattern matching) will treat several files as one file.
This following accepts a single file name argument and expands the glob before processing so the name can be displayed.
#requires -version 4
Set-StrictMode -Version 2
$pattern = $Args[0] # 'file*'
if ($Args[0] -eq $null) {
write-warning("Missing file pattern argument")
exit(1)
}
$filenames = get-childitem -Name $pattern
write-host("Simple file pattern")
foreach ($filename in $filenames) {
$addCWD = $false
$path = $filename
if ($addCWD) {
$path = Join-Path -path $cwd.value -childpath $filename
}
if (-not (Test-Path -path $path -pathtype leaf) ) {
write-warning("Filename, {0}, does not exist" -f $path)
exit(1)
}
else {
$count = 1
write-host("filename: {0}" -f $filename)
foreach ($line in get-content $path) {
write-host(" {0:D3}:{1}" -f $count, $line)
$count += 1
}
$fh = get-childitem $path # get file attributes
}
}
This example accepts all commandline arguments as file names and does not consider any globbing (file pattern matching).
#requires -version 4
Set-StrictMode -Version 2
write-host("All file arguments")
foreach ($filename in $Args) {
$addCWD = $false
$path = $filename
if ($addCWD) {
$path = Join-Path -path $cwd.value -childpath $filename
}
if (-not (Test-Path -path $path -pathtype leaf) ) {
write-warning("Filename, {0}, does not exist" -f $path)
exit(1)
}
else {
$count = 1
write-host("filename: {0}" -f $filename)
foreach ($line in get-content $path) {
write-host(" {0:D3}:{1}" -f $count, $line)
$count += 1
}
$fh = get-childitem $path # get file attributes
}
}
Writing Files
Simplest approach is to use set-content, add-content and clear-content cmd-lets, which have many options not covered here.
#requires -version 4
Set-StrictMode -Version 2
$h = @{ Fred = 30; Wilma = 25; Pebbles = 1; Dino = 5 }
set-content -path "file.obj" -value $h # writes hash-table object
$path = "file.txt"
# add one line at a time, note no need to close the file
set-content -path $path -value $null # creates and closes an empty file
foreach ($key in $h.keys) {
add-content -path $path -value ("{0}:{1:D}" -f $key, $h[$key]) # adds content and closes
# ("{0}:{1:D}" -f $key, $h[$key]) | add-content -path $path # same, less intuative
}
clear-content -path $path # clear the file contents
# string with line continuation characters.
$text = "Fred:30`
Wilma:25`
Pebbles:1`
Dino:5"
$text | set-content -path $path
clear-content -path $path # clear the file contents
# string containing new-line characters.
$text = "Fred:30`nWilma:25`nPebbles:1`nDino:5"
$text | set-content -path $path
clear-content -path $path # clear the file contents
# string containing new-line characters using out-file
$text | Out-File -FilePath $path
See also:
Displaying CSV Files
Powershell provides cmdlets for handling these which avoid importing into Excel and MS Access.
The out-gridview renders the output the data in an interactive table.
PS> import-csv -Path file.csv -Delimeter "`t" | out-gridview # load and display a <TAB> separated file.
PS> import-csv -Path file.csv -Delimeter ";" | out-gridview # load and display a ';' separated file.
PS> get-content file.csv
Name;Age
Fred;30
Wilma;25
Pebbles;1
Dino;5
PS> $f = import-csv -delimiter ';' file.csv
PS> $f.Name # Fred Wilma Pebbles Dino
PS> $f[1].Name # Wilma
PS> $f.Age # 30 25 1 5
PS> $f[3].Age # 5
PS> for ($i =0; $i -lt $f.length; $i++) {
write-output("{0,-7} is {1:D} years" -f $f[$i].Name, $f[$i].Age)
}
PS> import-csv -delimiter ';' file.csv | out-gridview
Reading JSON files
PowerShell requires that ConvertTo-Json and ConvertFrom-Json modules are installed.
PS> get-content file2.json
{
"family":"flintstone",
"members":
[
{"Name":"Fred", "Age":"30"},
{"Name":"Wilma", "Age":"25"},
{"Name":"Pebbles", "Age":"1"},
{"Name":"Dino", "Age":"5"}
]
}
PS> get-content file2.json | ConvertFrom-Json
family members
------ -------
flintstone {@{Name=Fred; Age=30}, @{Name=Wilma; Age=25}, @{Name=Pebbles; Age=1}, @{Name=Dino; Age=5}}
PS> $obj = get-content file2.json | convertfrom-json
PS> $obj
family members
------ -------
flintstone {@{Name=Fred; Age=30}, @{Name=Wilma; Age=25}, @{Name=Pebbles; Age=1}, @{Name=Dino; Age=5}}
PS> $obj.family # returns flintstone
PS> $obj.members[0].name # returns Fred
PS> $obj.members[0].age # returns 30
PS> $obj.members[0].age = 35 # set Fred's age to 35
PS> $obj.members[0].age # now returns 35
PS> $obj | convertto-json | add-content newfile.json # save as JSON
PS> $obj.members.name # returns: Fred Wilma Pebbles Dino
PS> $obj.members.age # returns: 35 25 1 5
PS> $obj.members.age[0] # returns 35
PS> $obj.members.age[0] = 37 # immutable, silently fails, no error
PS> $obj.members.age[0] # returns 35
PS> remove-variable -name obj # cleanup
PS> get-content newfile.json
{
"family": "flintstone",
"members": [
{
"Name": "Fred",
"Age": 35
},
{
"Name": "Wilma",
"Age": "25"
},
{
"Name": "Pebbles",
"Age": "1"
},
{
"Name": "Dino",
"Age": "5"
}
]
}
Further reading:
Reading XML files
Powershell supports full manipulation of the XML DOM, read the Introduction to XML
and .NET XmlDocument Class for more detailed information. The examples shown
are very rudimentary, and only show a few of the manipulations you can perform on XML objects.
Note, cmdlets Export-Clixml and
Import-Clixml provide a simplified way to save
and reload your PowerShell objects and are Microsoft specific.
PS> get-content .\file2.xml
<?xml version="1.0" encoding="UTF-8"?>
<family surname = "Flintstone">
<member>
<name>Fred</name>
<age>30</age>
</member>
<member>
<name>Wilma</name>
<age>25</age>
</member>
<member>
<name>Pebbles</name>
<age>1</age>
</member>
<member>
<name>Dino</name>
<age>5</age>
</member>
</family>
PS> $obj = [XML] (get-content .\file2.xml) # returns a System.Xml.XmlDocument object
PS> $obj.childnodes # returns all the child nodes
PS> $obj.xml # returns version="1.0" encoding="UTF-8"
PS> $obj.childnodes.surname # Flintstone
PS> $obj.childnodes.member.name # returns Fred Wilma Pebbles Dino
PS> $obj.childnodes.member.age # returns 30 25 1 5
PS> $obj.ChildNodes[0].NextSibling
surname member
------- ------
Flintstone {Fred, Wilma, Pebbles, Dino}
PS> $obj.GetElementsByTagName("member");
name age
---- ---
Fred 30
Wilma 25
Pebbles 1
Dino 5
PS> $obj.GetElementsByTagName("member")[0].name # returns Fred
PS> $obj.GetElementsByTagName("member")[0].age # returns 30
PS> $obj.GetElementsByTagName("member")[0].age = 35 # Errors, only strings can be used.
PS> $obj.GetElementsByTagName("member")[0].age = "35" # Fred is now older
PS> $obj.GetElementsByTagName("member")[0].age # returns 35
PS> $obj.Save("$PWD\newfile.xml") # needs a full pathname
PS> get-content newfile.xml
<?xml version="1.0" encoding="UTF-8"?>
<family surname="Flintstone">
<member>
<name>Fred</name>
<age>35</age>
</member>
<member>
<name>Wilma</name>
<age>25</age>
</member>
<member>
<name>Pebbles</name>
<age>1</age>
</member>
<member>
<name>Dino</name>
<age>5</age>
</member>
</family>
Writing XML files
To generate an XML file, use the XmlTextWriter Class
Note: cmdlets Export-Clixml and
Import-Clixml provide a simplified way to save
and reload your PowerShell objects and are Microsoft specific.
$settings = New-Object System.Xml.XmlWriterSettings # to update XmlWriterSettings
$settings.Indent = $true # indented XML
$settings.IndentChars = "`t" # <TAB> indents
$settings.Encoding = [System.Text.Encoding]::UTF8 # force the default UTF8 encoding; others ASCII, Unicode...
$obj = [System.XML.XmlWriter]::Create("C:\users\geoff\bedrock.xml", $settings) # note full-pathname
# Simpler approach but no encoding is specified in XML header and again note full-pathname
# $obj = New-Object System.XMl.XmlTextWriter('C:\users\geoff\bedrock.xml', $null)
# $obj.Formatting = 'Indented'
# $obj.Indentation = 1
# $obj.IndentChar = "`t"
$obj.WriteStartDocument() # start xml document, <?xml version="1.0"?>
$obj.WriteComment('Bedrock Families') # add a comment, <!-- Bedrock Families -->
$obj.WriteStartElement('family') # start element <family>
$obj.WriteAttributeString('surname', 'Flintstone') # add surname attribute
$obj.WriteStartElement('member') # start element <member>
$obj.WriteElementString('name','Fred') # add <name>Fred</name>
$obj.WriteElementString('age','30') # add <age>30</age>
$obj.WriteEndElement() # end element </member>
$obj.WriteStartElement('member') # start element <member>
$obj.WriteElementString('name','Wilma') # add <name>Wilma</name>
$obj.WriteElementString('age','25') # add <age>25</age>
$obj.WriteEndElement() # end element </member>
$obj.WriteStartElement('member') # start element <member>
$obj.WriteElementString('name','Pebbles') # add <name>Pebbles</name>
$obj.WriteElementString('age','1') # add <age>1</age>
$obj.WriteEndElement() # end element </member>
$obj.WriteStartElement('member') # start element <member>
$obj.WriteElementString('name','Dino') # add <name>Dino</name>
$obj.WriteElementString('age','5') # add <age>5</age>
$obj.WriteEndElement() # end element </member>
$obj.WriteEndElement() # end element <family>
$obj.WriteEndDocument() # end document
$obj.Flush() # flush
$obj.Close() # close, writes the file
PS> get-content C:\users\geoff\bedrock.xml
<?xml version="1.0" encoding="utf-8"?>
<!--Bedrock Families-->
<family surname="Flintstone">
<member>
<name>Fred</name>
<age>30</age>
</member>
<member>
<name>Wilma</name>
<age>25</age>
</member>
<member>
<name>Pebbles</name>
<age>1</age>
</member>
<member>
<name>Dino</name>
<age>5</age>
</member>
</family>
PS> remove-variable -name settings
PS> remove-variable -name obj
PS> remove-item C:\users\geoff\bedrock.xml
Log files: tail, write time-stamped message
# tailing a log file
PS> get-content -wait -last 10 "application.log"
PS> get-content -wait "application.log" | out-host -paging
# writing a time-stamped log message
PS> $LogFile = "application.log"
PS> $DateTime = "[{0:MM/dd/yy} {0:HH:mm:ss}]" -f (Get-Date) # [03/22/21 21:07:06]
PS> $LogMessage = "$Datetime: $LogString"
PS> add-content $LogFile -value $LogMessage
Formatting Variables
Very similar to Python -f operator, examples use write-host but can be used with other cmdlets, such as assignment.
Specified as {<index>, <alignment><width>:<format_spec>}
PS> $shortText = "Align me"
PS> $longerText = "Please Align me, but I am very wide"
PS> write-host("{0,-20}" -f $shortText) # Left-align; no overflow.
PS> write-host("{0,20}" -f $shortText) # Right-align; no overflow.
PS> write-host("{0,-20}" -f $longerText) # Left-align; data overflows width.
PS> write-host("Room: {0:D}" -f 232) # Room: 232
PS> write-host("Invoice No.: {0:D8}" -f 17) # Invoice No.: 00000017
PS> $invoice = "{0}-{1}" -f 00017, 007 # (integers) so invoice = 17-7
PS> $invoice = "{0}-{1}" -f '00017', '007' # (strings) so invoice = 00017-007
PS> write-host("Temp: {0:F}°C" -f 18.456) # Temp: 18.46°C
PS> write-host("Grade: {0:p}" -f 0.875) # Grade: 87.50%
PS> write-host('Grade: {0:p0}' -f 0.875) # Grade: 88%
PS> write-host('{1}: {0:p0}' -f 0.875, 'Maths') # Maths: 88%
# Custom formats
PS> write-output('{1:00000}' -f 'x', 1234) # 01234
PS> write-output('{0:0.000}' -f [Math]::Pi) # 3.142
PS> write-output('{0:00.0000}' -f 1.23) # 01.2300
PS> write-host('{0:####}' -f 1234.567) # 1235
PS> write-host('{0:####.##}' -f 1234.567) # 1234.57
PS> write-host('{0:#,#}' -f 1234567) # 1,234,567
PS> write-host('{0:#,#.##}' -f 1234567.891) # 1,234,567.89
PS> write-host('{0:000}:{1}' -f 7, 'Bond') # 007:Bond
PS> get-date -Format 'yyyy-MM-dd:hh:mm:ss' # 2020-04-27T07:19:05
PS> get-date -Format 'yyyy-MM-dd:HH:mm:ss' # 2020-04-27T19:19:05
PS> get-date -UFormat "%A %m/%d/%Y %R %Z" # Monday 04/27/2020 19:19 +02
More detailed formatting examples:
Output methods:
Security Considerations
Running PowerShell scripts
PowerShell is very powerful scripting language, often used to automate routine tasks, which makes it an ideal
target for would-be hackers. To mitigate this Microsoft limits PowerShell execution, even though the
individual cmdlets can always be run.
Many articles on the internet suggest the disabling or trying to work around this security feature… PLEASE AVOID DOING THIS!
Many Windows distributions provide Developer section in Settings, which allows local PowerShell scripts to be
executed by the CurrentUser by setting the ExecutionPolicy to RemoteSigned.
Alternatively this can also be done manually by running PowerShell as Administrator
ADM-PS1> set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser
Execution Policy Change
The execution policy helps protect you from scripts that you do not trust. Changing the execution policy might expose you to the security risks
described in the about_Execution_Policies help topic at https:/go.microsoft.com/fwlink/?LinkID=135170. Do you want to change the execution policy?
[Y] Yes [A] Yes to All [N] No [L] No to All [S] Suspend [?] Help (default is "N"): A
A sensible working setup for your personal laptop
PS> Get-ExecutionPolicy -list
Scope ExecutionPolicy
----- ---------------
MachinePolicy Undefined
UserPolicy Undefined
Process Undefined
CurrentUser RemoteSigned
LocalMachine Undefined
A sensible working setup for a typical windows server installation
PS> Get-ExecutionPolicy -list
Scope ExecutionPolicy
----- ---------------
MachinePolicy Undefined
UserPolicy Undefined
Process Undefined
CurrentUser AllSigned
LocalMachine AllSigned
PowerShell Execution Policies
Execution policies
are a safety feature to control the conditions under which PowerShell loads configuration
files and runs scripts, with the intention to prevent the execution of malicious scripts. This is augmented with the notion
of a Execution Policy Scope, conditions under which the Execution Policy is applied
Execution policies (highest to lowest):
Restricteddoes not permit any scripts to run (.ps1xml, .psm1, .ps1)AllSignedprevents running scripts that do not have a digital signatureRemoteSignedprevents running downloaded scripts that do not have a digital signatureUnrestrictedruns without a digital signature, warns about non-local intranet zone scriptsBypassallows running of scripts without any digital signature, and without any warningsUndefinedno execution policy is defined
Execution Policy Scope (highest to lowest):
MachinePolicyset by a Group Policy for all users of the computerUserPolicyset by a Group Policy for the current user of the computerProcesscurrent PowerShell session, environment variable$env:PSExecutionPolicyPreferenceCurrentUseraffects only the current user,HKEY_CURRENT_USERregistry subkeyLocalMachineall users on the current computer,HKEY_LOCAL_MACHINEregistry subkey
In a commercial or industrial environment this is usually managed by your local Windows Administrators, hence
MachinePolicy and UserPolicy in Execution Policy Scope and you maybe prevented from changing anything.
Example Set-ExecutionPolicy commands, these need to be executed in PowerShell running as Administrator
ADM-PS1> Set-ExecutionPolicy -ExecutionPolicy RemoteSigned # sets: LocalMachine RemoteSigned
ADM-PS1> Set-ExecutionPolicy -ExecutionPolicy Restricted # sets: LocalMachine Restricted
ADM-PS1> Set-ExecutionPolicy -ExecutionPolicy Undefined # sets: LocalMachine Undefined
ADM-PS1> Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser
ADM-PS1> Set-ExecutionPolicy -ExecutionPolicy AllSigned # mandate AllSigned for LocalMachine
ADM-PS1> Set-ExecutionPolicy -ExecutionPolicy Default # restore: LocalMachine defaults
PowerShell Code Signing
Microsoft uses a proprietary technique called Authenticode for code signing PowerShell
Apart from the proprietary nature, which impacts its generation, it is an asymmetric keypair, signed by an approved Certificate Authority (CA), installed in the Windows certificate stores, and so involves creating a code signing request (CSR) with an associated keypair and having it signed by an approved Certificate Authority.
Within a commercial organization there is probably an existing process that needs to be followed to generate the CSR and have it approved by the internally Certificate Authority.
Externally available applications or product should probably use an external commercially available service, the following guides may be useful.
For an internal development it is possible to use Self-Signed Authenticode Certificates, the generation of which is covered in the following section.
Self-Signed Authenticode Certificates
PowerShell Generating, Installing and Using a Self-Signed Certificate
This section stolen from Adam the Automator articles below, demonstrates
using PowerShell New-SelfSignedCertificate, which supports stores cert:CurrentUserMy or cert:LocalMachineMy.
Self-Signed Certificates Setup
Requires creating the following certificates using a PowerShell in Administrative mode.
LocalMachine\My Personal - public/private key and certificate for signing;
LocalMachine\Root - certificate for authentication;
LocalMachine\TrustedPublisher - certificate for authentication;
# Certificate Manager tools
C:\Windows\system32\certmgr.msc # Current User
C:\Windows\system32\certlm.msc # Local Machine
C:\Windows\system32\mmc.exe # MMC tool
ADM-PS> Get-ExecutionPolicy -List
Scope ExecutionPolicy
----- ---------------
MachinePolicy Undefined
UserPolicy Undefined
Process Undefined
CurrentUser RemoteSigned
LocalMachine Undefined
ADM-PS> $authenticode = New-SelfSignedCertificate -Subject "ATA Authenticode" -CertStoreLocation Cert:\LocalMachine\My -Type CodeSigningCert
# Add the self-signed Authenticode to LocalMachine\Root certificate store
ADM-PS> $rootStore = [System.Security.Cryptography.X509Certificates.X509Store]::new("Root","LocalMachine")
ADM-PS> $rootStore.Open("ReadWrite") ## Open LocalMachine\Root certificate store for read/write
ADM-PS> $rootStore.Add($authenticode) ## Add the certificate stored in the $authenticode variable.
ADM-PS> $rootStore.Close() ## Close the root certificate store.
# Add the self-signed Authenticode to LocalMachine\TrustedPublisher certificate store.
ADM-PS> $publisherStore = [System.Security.Cryptography.X509Certificates.X509Store]::new("TrustedPublisher","LocalMachine")
ADM-PS> $publisherStore.Open("ReadWrite") ## Open LocalMachine\TrustedPublisher certificate store for read/write
ADM-PS> $publisherStore.Add($authenticode) ## Add the certificate stored in the $authenticode variable.
ADM-PS> $publisherStore.Close() ## Close the TrustedPublisher certificate store.
# Verify all certificates are created and the Thumbprint same
ADM-PS> Get-ChildItem Cert:\LocalMachine\My | Where-Object {$_.Subject -eq "CN=ATA Authenticode"}
PSParentPath: Microsoft.PowerShell.Security\Certificate::LocalMachine\My
Thumbprint Subject
---------- -------
F71A096EFCDC99DFAC109A228565B427B66DF49F CN=ATA Authenticode
ADM-PS> Get-ChildItem Cert:\LocalMachine\Root | Where-Object {$_.Subject -eq "CN=ATA Authenticode"}
PSParentPath: Microsoft.PowerShell.Security\Certificate::LocalMachine\Root
Thumbprint Subject
---------- -------
F71A096EFCDC99DFAC109A228565B427B66DF49F CN=ATA Authenticode
ADM-PS> Get-ChildItem Cert:\LocalMachine\TrustedPublisher | Where-Object {$_.Subject -eq "CN=ATA Authenticode"}
PSParentPath: Microsoft.PowerShell.Security\Certificate::LocalMachine\TrustedPublisher
Thumbprint Subject
---------- -------
F71A096EFCDC99DFAC109A228565B427B66DF49F CN=ATA Authenticode
Using the Authenticode, Signing and Running
# Enforce AllSigned, select '[A] Yes to All' option
ADM-PS> set-ExecutionPolicy -ExecutionPolicy AllSigned -Scope CurrentUser
ADM-PS> set-ExecutionPolicy -ExecutionPolicy AllSigned -Scope LocalMachine
ADM-PS> PS C:\Users\geoff> Get-ExecutionPolicy -List
Scope ExecutionPolicy
----- ---------------
MachinePolicy Undefined
UserPolicy Undefined
Process Undefined
CurrentUser AllSigned
LocalMachine AllSigned
# Get the *ATA Authenticode*
ADM-PS> $codeCertificate = Get-ChildItem Cert:\LocalMachine\My | Where-Object {$_.Subject -eq "CN=ATA Authenticode"}
PS C:\> Get-Content C:\Users\sjfke\hello-world.ps1
#requires -version 4
Set-StrictMode -Version 2
write-host 'host: hello world!'
write-output 'output: hello world!'
exit(0)
ADM-PS> Set-AuthenticodeSignature -FilePath C:\Users\sjfke\hello-world.ps1 -Certificate $codeCertificate
# Appends a signature, makes it immutable, any changes require Set-AuthenticodeSignature again.
PS C:\> Get-Content C:\Users\sjfke\hello-world.ps1
#requires -version 4
Set-StrictMode -Version 2
write-host 'host: hello world!'
write-output 'output: hello world!'
exit(0)
# SIG # Begin signature block
<-- text-removed -->
# SIG # End signature block
ADM-PS> C:\Users\sjfke\hello-world.ps1
host: hello world!
output: hello world!
PS> C:\Users\sjfke\hello-world.ps1
host: hello world!
output: hello world!
Adding a TimeStampServer should ensure that your code will not expire when the signing certificate expires.
ADM-PS> Set-AuthenticodeSignature -FilePath C:\Users\sjfke\hello-world.ps1 -Certificate $codeCertificate -TimeStampServer http://timestamp.digicert.com
# Freely available TimeStampServers
- http://timestamp.digicert.com
- http://timestamp.comodoca.com
- http://timestamp.globalsign.com
- http://tsa.starfieldtech.com
- http://timestamp.entrust.net/TSS/RFC3161sha2TS
- http://sha256timestamp.ws.symantec.com/sha256/timestamp
- http://tsa.swisssign.net
OpenSSL: Generating, Installing and Using a Self-Signed Certificate
In PowerShell Generating, Installing and Using a Self-Signed Certificate the sequence is:
Generate ata-authenticode (certificate, private key) in certificate store, LocalMachine\My
Import ata-authenticode into certificate store LocalMachine\Root for authentication;
Import ata-authenticode into certificate store LocalMachine\TrustedPublisher for authentication;
OpenSSL uses the CurrentUser Execution Policy Scope, with the same sequence and requires a few more steps
- Generate atb-authenticode (certificate, private key) in certificate store, CurrentUser\My
Generate atb-authenticode.key and atb-authenticode.csr
Generate self-signed atb-authenticode.crt
Merge atb-authenticode.crt and authenticode.key into authenticode.pfx
Import authenticode.pfx into certificate store CurrentUser\My
Import authenticode.pfx into certificate store CurrentUser\Root for authentication;
Import authenticode.pfx into certificate store CurrentUser\TrustedPublisher for authentication;
The following was done using Git Bash shell but the of atb-authenticode could
be built on any system with OpenSSL because all that is needed is the authenticode.pfx file.
An explicit OpenSSL configuration file, authenticode-selfsign-openssl.cnf is used to avoid issues resulting from
differences in the default configuration in the OpenSSL installation.
OpenSSL: Self-Signed Certificates Setup
Step 1a - generate atb-authenticode.key and atb-authenticode.csr
$ openssl req -new -newkey rsa:2048 -nodes -keyout authenticode.key -out authenticode.csr -config authenticode-selfsign-openssl.cnf
Generating a RSA private key
......................................................+++++
.....................................................+++++
writing new private key to 'authenticode.key'
-----
You are about to be asked to enter information that will be incorporated
into your certificate request.
What you are about to enter is what is called a Distinguished Name or a DN.
There are quite a few fields but you can leave some blank
For some fields there will be a default value,
If you enter '.', the field will be left blank.
-----
Country Name (2 letter code) [CH]:.
State or Province Name (full name) [Zurich]:.
Locality Name (eg, city) [Zurich]:.
Organization Name (eg, company) [Highly Dubious Inc]:.
Organizational Unit Name (eg, section) []:.
Common Name (eg, YOUR name) [HighlyDubious]:ATB Authenticode
Email Address []:.
Step 1b - generate self-signed atb-authenticode.crt
# Note options: -extensions v3_req -extfile authenticode-selfsign-openssl.cnf
$ openssl x509 -req -extensions v3_req -extfile authenticode-selfsign-openssl.cnf -days 366 -in authenticode.csr -signkey authenticode.key -out authenticode.crt
Signature ok .
subject=CN = ATB Authenticode
Getting Private key
# Check the certificate for the following section
$ openssl x509 -noout -text -in authenticode.crt | less
X509v3 extensions:
X509v3 Basic Constraints: critical
CA:FALSE
X509v3 Subject Key Identifier:
39:04:14:30:74:B8:00:51:2F:30:11:E6:D3:D5:FF:A9:3B:2A:21:53
X509v3 Extended Key Usage: critical
Code Signing, Microsoft Individual Code Signing
Step 1c - merge atb-authenticode.crt and authenticode.key -into- authenticode.pfx
Note: an empty password can be used
$ openssl pkcs12 -export -out authenticode.pfx -inkey authenticode.key -in authenticode.crt
Enter Export Password:
Verifying - Enter Export Password:
The next few steps involve importing the authenticode.pfx into the Windows certificate store, unlike
PowerShell Generating, Installing and Using a Self-Signed Certificate it uses CurrentUser\My, CurrentUser\Root and
CurrentUser\TrustedPublisher.
# Certificate Manager tools
PS1> C:\Windows\system32\certmgr.msc # Current User
PS1> or C:\Windows\system32\mmc.exe # MMC tool
Step 1d - import authenticode.pfx -into- CurrentUser\My
Step 2 - import authenticode.pfx -into- CurrentUser\Root - certificate trust/authentication;
Step 3 - import authenticode.pfx -into- CurrentUser\TrustedPublisher - certificate for trust/authentication;
OpenSSL: Using the Authenticode, Signing and Running
Requires using a PowerShell in Administrative mode to execute set-ExecutionPolicy commands, prompt ADM-PS1>
and a normal PowerShell, prompt PS1> for the rest.
# Enforce AllSigned, select '[A] Yes to All' option
ADM-PS> set-ExecutionPolicy -ExecutionPolicy AllSigned -Scope CurrentUser
ADM-PS> set-ExecutionPolicy -ExecutionPolicy AllSigned -Scope LocalMachine
ADM-PS> PS C:\Users\sjfke> Get-ExecutionPolicy -List
Scope ExecutionPolicy
----- ---------------
MachinePolicy Undefined
UserPolicy Undefined
Process Undefined
CurrentUser AllSigned
LocalMachine AllSigned
PS1> Get-Content -Path C:\Users\sjfke\hello-world.ps1
#requires -version 4
Set-StrictMode -Version 2
write-host 'host: hello world!'
write-output 'output: hello world!'
exit(0)
PS1> Get-ChildItem Cert:\CurrentUser\My | Where-Object {$_.Subject -eq "CN=ATB Authenticode"}
PSParentPath: Microsoft.PowerShell.Security\Certificate::CurrentUser\My
Thumbprint Subject
---------- -------
A6567CF9C6D5B0DCE4B7823B3DAF4CC4058DB396 CN=ATB Authenticode
PS1> $codeCertificate = Get-ChildItem Cert:\CurrentUser\My | Where-Object {$_.Subject -eq "CN=ATB Authenticode"}
PS1> Set-AuthenticodeSignature -FilePath C:\Users\sjfke\hello-world.ps1 -Certificate $codeCertificate
Directory: C:\Users\geoff
SignerCertificate Status Path
----------------- ------ ----
A6567CF9C6D5B0DCE4B7823B3DAF4CC4058DB396 Valid hello-world.ps1
PS1> C:\Users\sjfke\hello-world.ps1
host: hello world!
output: hello world!
PS1> Get-Content -Path C:\Users\sjfke\hello-world.ps1
#requires -version 4
Set-StrictMode -Version 2
write-host 'host: hello world!'
write-output 'output: hello world!'
exit(0)
# SIG # Begin signature block
# MIIFhQYJKoZIhvcNAQcCoIIFdjCCBXICAQExCzAJBgUrDgMCGgUAMGkGCisGAQQB
<-- text-removed -->
# dtUw8zNoZUTIq1eKdNJW+kxdDRPL56l3qQ==
# SIG # End signature block
OpenSSL file: authenticode-selfsign-openssl.cnf
This is the result of many iterations and consulting many references, most relevant being:
openssl-req, req - PKCS#10 certificate request and certificate generating utility
x509v3_config - X509 V3 certificate extension configuration format v3_req, v3_ca
####################################################################
# CA Definition
[ ca ]
default_ca = CA_default # The default ca section
[ CA_default ]
dir = . # Where everything is kept
certs = $dir/certsdb # Where the issued certs are kept
new_certs_dir = $certs # default place for new certs.
database = $dir/index.txt # database index file.
certificate = $dir/cacert.pem # The CA certificate
private_key = $dir/private/cakey.pem # The private key
serial = $dir/serial # The current serial number
RANDFILE = $dir/private/.rand # private random number file
default_days = 365 # how long to certify for
default_md = sha256 # which md to use.
preserve = no # keep passed DN ordering
email_in_dn = no
policy = policy_match
crldir = $dir/crl
crlnumber = $dir/crlnumber # the current crl number
crl = $crldir/crl.pem # The current CRL
#crl_extensions = crl_ext
default_crl_days= 30 # how long before next CRL
####################################################################
# The default policy for the CA when signing requests
[ policy_match ]
countryName = match # Must be the same as the CA
stateOrProvinceName = match # Must be the same as the CA
organizationName = match # Must be the same as the CA
organizationalUnitName = optional # not required
commonName = supplied # must be there, whatever it is
emailAddress = optional # not required
####################################################################
# This is where we define how to generate CSRs
[ req ]
default_bits = 2048
default_keyfile = privkey.pem
default_md = sha256 # which md to use.
# prompt = no
distinguished_name = req_distinguished_name # where to get DN for reqs
attributes = req_attributes # req attributes
string_mask = nombstr
# string_mask = utf8only
req_extensions = v3_req # The extensions to add to req's
x509_extensions = v3_ca # The extentions to add to self signed certs
[ req_distinguished_name ]
countryName = Country Name (2 letter code)
countryName_default = CH
countryName_min = 2
countryName_max = 2
stateOrProvinceName = State or Province Name (full name)
stateOrProvinceName_default = Zurich
localityName = Locality Name (eg, city)
localityName_default = Zurich
0.organizationName = Organization Name (eg, company)
0.organizationName_default = Highly Dubious Inc
organizationalUnitName = Organizational Unit Name (eg, section)
1.commonName = Common Name (eg, YOUR name)
1.commonName_default = HighlyDubious
1.commonName_max = 64
emailAddress = Email Address
emailAddress_max = 64
####################################################################
# We don't want these, but the section must exist
[ req_attributes ]
#challengePassword = A challenge password
#challengePassword_min = 4
#challengePassword_max = 20
#unstructuredName = An optional company name
####################################################################
# Extension for requests
[ v3_req ]
basicConstraints=critical,CA:FALSE
subjectKeyIdentifier = hash
#subjectAltName = @alternate_names
# * ATA Authenticate - Code Signing (1.3.6.1.5.5.7.3.3)
# * extendedKeyUsage=critical,codeSigning,1.3.6.1.5.5.7.3.3
extendedKeyUsage=critical,codeSigning,msCodeInd
####################################################################
# Convert a certificate request into a self signed certificate using extensions for a CA:
# https://www.openssl.org/docs/man1.1.1/man1/x509.html
[ v3_ca ]
#subjectAltName = @alternate_names
# * ATA Authenticate - Code Signing (1.3.6.1.5.5.7.3.3)
# * extendedKeyUsage=critical,codeSigning,1.3.6.1.5.5.7.3.3
extendedKeyUsage=critical,codeSigning,msCodeInd
subjectKeyIdentifier = hash
authorityKeyIdentifier = keyid:always,issuer
#[alternate_names]