Today I had an issue where getting an element at a certain index in a typed array failed. As I was troubleshooting this issue I found that it was simply because I had used a '.' (dot) instead of a ',' (comma) to separate the elements when I defined the typed array variable.
Okay, that is a clear mistake on my part. However, as I thought about it I would have loved to be alerted by the Powershell engine .

-- Experienced on v5.1.14409.1012 of PowerShell
-- Win7 x64 OS

TO REPRODUCE:

open vscode
create ps1 file
$array = "hej", "hej2", "hej3". "Hej4" $array --> prints = hej & hej2

EXTRA NOTES:

Also happens in Ps Shell.

If I do '$array = 1, 2. 4' --> I get an '...unexpexted token '4' in expression or statement.....'. So it seems to be an issue with [Strings]

There is also an github issue created on this. See more here >> https://github.com/PowerShell/vscode-powershell/issues/188

Drag and dropping any file to a script shortcut split this path before passing the splitted strings as multiple parameters. Here's my detailed Issue Report :

Steps to reproduce
------------------

1 - Get any script, for example :
[samplescript.psd1] :
```
param([string]$path)
Write-Host($path)
pause
```
or a simpler one-line sample: ```Write-Host($args);pause```

2 - Create a shortcut that runs the script :
[Shortcut Target ] : ```PowerShell "path\to\our\samplescript.psd1"```

3- Drag and drop a file containing spaces in its name to the shortcut
[sample file ] : ```C:\Some file with spaces.txt```

Expected behavior
-----------------

Drag and dropping the file should pass its path to the script as a parameter, (as it does correctly with batch scripts, executables, and other files that accept drag and drop [and shortcuts to these]),
and the script should then open a PowerShell window displaying "C:\Some file with spaces.txt"

Actual behavior
---------------

The file path is split into multiple parameters before being passed to the PowerShell script, which displays "C:\Some", instead of "C:\Some file with spaces.txt"

Workaround
---------------

Well, that isn't brilliant since I'm since starting to have a lot of these, but since this bug is only with powershell I simply create a .bat file which passes the argument to a PowerShell file. Instead of drag and dropping to the script shortcut; I drop it on the bat file [or a shortcut to it] to pass the path to the bat file, which then pass the path to the PowerShell script. Like this :
[dragdrop_for_samplescript.bat]
```start PowerShell -file "path\to\our\samplescript.psd1" %1```

Environment data
----------------

```
> $PSVersionTable

Name Value
---- -----
PSVersion 5.1.16257.1
PSEdition Desktop
PSCompatibleVersions {1.0, 2.0, 3.0, 4.0...}
BuildVersion 10.0.16257.1
CLRVersion 4.0.30319.42000
WSManStackVersion 3.0
PSRemotingProtocolVersion 2.3
SerializationVersion 1.1.0.1
```

OS: Windows Server 2016 1607 (14393.1593)
PSVersionTable:
Name Value
---- -----
PSVersion 5.1.14393.1532
PSEdition Desktop
PSCompatibleVersions {1.0, 2.0, 3.0, 4.0...}
BuildVersion 10.0.14393.1532
CLRVersion 4.0.30319.42000
WSManStackVersion 3.0
PSRemotingProtocolVersion 2.3
SerializationVersion 1.1.0.1

PSDscResource module: v2.8.0.0

Take a look at this configuration:
Configuration DependentConfig
{
Import-DscResource -ModuleName PSDscResources -Name Registry

Node MyNode01
{

File SetupScript
{
DestinationPath = 'C:\Scripts\hello.txt'
Contents = 'somecommand'
Type = 'File'
Ensure = 'Present'
DependsOn = '[Registry]regCheck'
}

Registry regCheck
{
Key = 'HKEY_LOCAL_MACHINE\SYSTEM\SomeKey'
ValueName = 'SomeValue'
ValueData = 0
ValueType = 'DWord'
Ensure = 'Present'
}
}
}

DependentConfig -OutputPath C:\DependentConfig

This requires PSDscResources module version 2.8.0.0 from the gallery and the file resource from the PSDesiredStateConfiguration in-box module. However, compiling this configuration will fail with a message that File resource cannot be loaded. If we remove -Name File from the first import command, you will see an error that Registry resource cannot be found. We have to use -Name with PSDscResources since PSDesiredStateConfiguration already exports the same resource.

So, the only work around at this time is to remove the PSDesiredStateConfiguration import completely from the configuration so that we can at least compile this. Here is something that works.

Configuration DependentConfig
{
Import-DscResource -ModuleName PSDesiredStateConfiguration
Import-DscResource -ModuleName PSDscResources -Name Registry

Node MyNode01
{

File SetupScript
{
DestinationPath = 'C:\Scripts\hello.txt'
Contents = 'somecommand'
Type = 'File'
Ensure = 'Present'
DependsOn = '[Registry]regCheck'
}

Registry regCheck
{
Key = 'HKEY_LOCAL_MACHINE\SYSTEM\SomeKey'
ValueName = 'SomeValue'
ValueData = 0
ValueType = 'DWord'
Ensure = 'Present'
}
}
}

DependentConfig -OutputPath C:\DependentConfig

When using a foreach loop to create a set of resources and trying to set the previous item key as the dependsOn resource, it adds it twice in the mof file.

Configuration example:
configuration Test-DependsOnWithLoop {
param([string[]]$Directories)
Import-DscResource -ModuleName PSDesiredStateConfiguration
node localhost {
$dependsOn = $null
$dirs.ForEach({
$res = $_ -replace '\\|:', '_'
File $res {
Ensure = 'Present'
DestinationPath = $_
DependsOn = $dependsOn
Type = 'Directory'
}
$dependsOn = "[File]$res"
})
}
}

$dirs = 1..10 | % { "C:\Temp\_$_" }
Test-DependsOnWithLoop -Directories $dirs

And if you look at the localhost.mof you'll see:
DependsOn = { "[File]C__Temp__1", "[File]C__Temp__1"};

I traced it back to the Test-DependsOn function in CimDSCParser.cs, though in this case, it's not a composite resource

I would have liked the default behaviour to be that an array of newline-split strings was returned from Get-ClipBoard. 97 % of the time I end up writing (Get-ClipBoard) -split "\r?\n" | ...... often filtering out empty elements - and that's just boring.

In my own version I used -Raw for a string and nothing for an array/list.

Of course you can't change that now in the 5.x version (or later, sigh, will this be our doom in some distant future), but can you maybe add at least -AsArray or similar? Make it so it can only be used with text, I guess. I see some other stuff in -Format I haven't tried yet.

A sort of implementation of the text part that I'd like is put up here on GitHub and demonstrates some things I'd want as well (I don't mean to start a storm with the little thought piece ;) https://github.com/EliteLoser/Get-Clip

I am trying to update Pester on Windows 2016 to 4.x because the preshipped Pester 3.4.0 emits a note about the depreciation of the -quiet switch when I use the quiet switch. This get's scooped into stdout when executing via AWS SSM remote commands. Version 3.x complains it does not understand "-Show None".

I want to update to 4.x - in which I ran into all the problems documented here: https://github.com/OneGet/oneget/issues/215

But ended up with "Install-module pester -force -SkipPublisherCheck.

So far so bad, but when I go to remove the 3.4.0 version (with admin rights) from: 'c:\program files\windowspowershell\modules\pester\3.4.0' I get errors because only TrustedInstaller has permissions.

If this scenario wasn't horrible enough, putting preinstalled modules under Windows Resource Protection really throws it over the top.

Could someone please make sure that preinstalled modules can be *replaced* cleanly? Or don't preinstall them?

A crashed powershell ISE session will re-open previously open files.

There is a bug, though, which manifested when a crash recovery happened to me recently:

The session/file which I had originally opened in standard user context (no elevation) recovered in High integrity level (aka elevated/admin),
The session/files which I had originally opened in elevated mode recovered in Medium integrity level (aka standard user mode).

The session should remember which context it was open in and re-open in the same context.

Need to consider this issue for normal ISE closure/re-opening, too, as in this UserVoice item:
"Windows PowerShell ISE: Add ability to save/load session state"
https://windowsserver.uservoice.com/forums/301869-powershell/suggestions/11087943-windows-powershell-ise-add-ability-to-save-load-s

I have system modules that are deployed in nodes according to the configuration data file (psd1). For example, one node can have three modules, or there may be three nodes each one with a module.
All modules have a common configuration, so I put this configuration in a separate composite resource. Then, if there are more than one module in a node, the resulting configuration will have duplicate resources.
The problem occur when one of these duplicate resources have some complex property. Here is a simplified version of my configuration:

Configuration BaseWebConfiguration
{
Import-DscResource -ModuleName xWebAdministration

xWebsite DefaultSite
{
Ensure = 'Present'
Name = 'Default Web Site'
PhysicalPath = 'C:\inetpub\wwwroot'
LogFlags = @('Date', 'Time')
}
}

Configuration ConfigurationTest
{
Node $AllNodes.Where{ $_.Modules -contains 'web_backend' }.NodeName
{
BaseWebConfiguration BaseWebConfigForWebBackend { }

# Web backend specific resources
}

Node $AllNodes.Where{ $_.Modules -contains 'mobile_backend' }.NodeName
{
BaseWebConfiguration BaseWebConfigForMobileBackend { }

# Mobile backend specific resources
}
}

$configData =
@{
AllNodes =
@(

@{
NodeName = 'localhost'
Modules = 'mobile_backend', 'web_backend'
}
);
}

ConfigurationTest -ConfigurationData $configData

When I run this configuration, the following error occur:

PSDesiredStateConfiguration\Configuration : A conflict was detected between resources '[xWebsite]DefaultSite::[BaseWebConfiguration]BaseWebConfigForWebBackend
(C:\Temp\Config\ConfigurationTest.ps1::5::5::xWebsite)' and '[xWebsite]DefaultSite::[BaseWebConfiguration]BaseWebConfigForMobileBackend
(C:\Temp\Config\ConfigurationTest.ps1::5::5::xWebsite)' in node 'localhost'. Resources have identical key properties but there are differences in the following non-key
properties: 'LogFlags'. Values 'System.Object[]' don't match values 'System.Object[]'. Please update these property values so that they are identical in both cases.

If the LogFlags property is not set, everything is ok. The error seems to be in the PSDesiredStateConfiguration.psm1 file, line 1489:

elseif ( $resource[$property] -ne $properties[$property] )

where a comparision ByValue is done even with complex properties.

Any way to workaround this for the time being?

When I run a simple function, like the following, I get an error, as expected that I've exceeded the number of values I can specify for the function parameter(-ComputerName). The wording says that I need to "Provide fewer than 3 arguments...". Because I can specify up to 3 arguments, I think the wording needs to change to say something like "Provide no more than 3 arguments,..." or "Provide 3 arguments, or less,...".

Function Get-OSInfo
{
[CmdletBinding()]
Param
(
[ValidateCount(1,3)]
[string[]] $ComputerName
)
Get-WmiObject -Class Win32_OperatingSystem -ComputerName $ComputerName
}
Get-OSInfo -ComputerName 'one','two','three','four'

Get-OSInfo : Cannot validate argument on parameter 'ComputerName'. The number of provided arguments, (4), exceeds the maximum number
of allowed arguments (3). Provide fewer than 3 arguments, and then try the command again.
At line:11 char:26
+ Get-OSInfo -ComputerName 'one','two','three','four'
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : InvalidData: (:) [Get-OSInfo], ParameterBindingValidationException
+ FullyQualifiedErrorId : ParameterArgumentValidationError,Get-OSInfo

Cannot declare [ordered] fields/properties in script classes

The following fails to work as expected, in multiple ways:

Class Foo { [ordered] $myfield }

When executing directly from an interactive shell, the output is:

An error occurred while creating the pipeline.
+ CategoryInfo : NotSpecified: (:) [], ParentContainsErrorRecordException
+ FullyQualifiedErrorId : RuntimeException

When executed from a script via ". .\myscript.ps1" the error is:

Value cannot be null.
Parameter name: type
At line:1 char:1
+ . .\myscript.ps1
+ ~~~~~~~~~~~~~~~~
+ CategoryInfo : OperationStopped: (:) [], ArgumentNullException
+ FullyQualifiedErrorId : System.ArgumentNullException

What I find especially interesting is that it 'line:1 char:1' regardless of where the class definition occurs in the script.

$PSVersionTable | clip
Name Value
---- -----
PSVersion 5.1.14393.1358
PSEdition Desktop
PSCompatibleVersions {1.0, 2.0, 3.0, 4.0...}
BuildVersion 10.0.14393.1358
CLRVersion 4.0.30319.42000
WSManStackVersion 3.0
PSRemotingProtocolVersion 2.3
SerializationVersion 1.1.0.1

I wrote a script that takes a MAC address in any common format as input and outputs it to all other standard MAC formats. This is all in a single row of data and is meant to be a convenient tool for working between systems that use different formats of MACs without having to constantly reformat them by hand. However, when copying the output from the gridview a trailing tab is placed at the end of the MAC even though there is only a single row in the view and no tabs in the original text. It's like it is setting up for the next row of data that isn't there.

I would expect a ctrl+c on a single row of data containing "64-00-6A-77-9A-50" to return "64-00-6A-77-9A-50" to the clipboard but instead it captures "64-00-6A-77-9A-50 " to the clipboard.

Once a node meta configuration is enacted, it is easy for an administrator or process (with malicious intent) to modify the MetaConfig.mof file in C:\Windows\System32\Configuration directory. The GetMetaConfiguration method in MSFT_DscMetaConfiguration class does not validate the property values against the allowed values of the CIM properties.

Steps to reproduce this behavior:
- Enact a simple meta configuration and enact it.
- Open the MetaConfig.MOF file in your favorite editor and change the value of ConfigurationMode to some random text.
- Save the file and close it.
- Run Get-DscLocalConfigurationManager.
- You will see the random value assigned to ConfigurationMode in the output although it is not a valid value for the ConfigurationMode property.

You can see this in action at https://pbs.twimg.com/media/DFervKvUIAA-7Na.jpg:large

Get- and Remove-DnsServerResourceRecord have a couple of deficiencies:

1) most cmdlets support piping from Get-* to Remove-*, this does not currently work properly with Remove-DnsServerResourceRecord...

"The input object cannot be bound because it did not contain the information required to bind all mandatory parameters: ZoneName"

So, the objects returned from Get-DnsServerResourceRecord should contain the ZoneName property.

2) Remove-DnsServerResourceRecord does not respect -Confirm:$false, even when it's specified, it asks for confirmation every time.

i.e. this will ask for confirmation 200 times if you have 200 records (and -ZoneName must be specified because of the aforementioned issue!), at least thankfully you can press & hold down enter:

Get-DnsServerResourceRecord -RRType aaaa -ZoneName ad.example.com | Remove-DnsServerResourceRecord -ZoneName ad.example.com -Confirm:$false