Adding a column to a datatable in powershell - powershell

Question

I am trying to add a column to data I have imported (and will export) as a CSV.
I am importing a CSV:
What I want to do add another column, perhaps "10/15/22" when the process runs, and then update the values under that date.
In effect, the document will grow to the right, adding a column each time it is run.
I have an object called $test. It will have values like:
$test.users = "7"
$test.SomeBSValue = "22"
$test.Computers = "52"
When all is said and done, my output should look like:
Adding to the list any values I have that are not part of the list already, but recording the values I have under the heading for the date.
So, if the script is run and collects 100 data points, those data point would all be in the CSV under the date.
I would have thought this would be easy, but now I am drawing a complete blank.
I've considered (but have not coded) even trying to put into a GUI grid view and then reading the data back and writing the CSV (but there should be an easier way, right?)

Answer 1

Since you don't actually use it as a CSV we can treat it like regular content.
Say we have a file in C:\test called test.csv that looks as follows:
"Settings","08/15/22","09/15/22"
"Users",0,0
"Computers",0,1
"SomeValue1",0,2
"SomeValue2",0,2
"SomeValue3",0,2
"Stat1",0,10
"Stat2",7,0
"Stat3",0,0
"SomeBSValue",1,2
We can import it, add the row from the object to each corresponding row and right the file to test2.csv.
$test = #{
Settings = "10/15/22"
users = "7"
Computers = "52"
SomeValue1 = "22"
SomeValue2 = "24"
SomeValue3 = "25"
Stat1 = "4"
Stat2 = "3"
Stat3 = "2"
SomeBSValue = "1"
}
$content = Get-Content "C:\test\test.csv"
$newContent = #()
foreach($row in $content){
foreach($key in $test.Keys){
if($row -like "*$key*"){
$row = $row + "," + $test."$key"
$newContent += $row
}
}
}
$newContent | Out-File "C:\test\test2.csv"
After running the script it will have added the values from the object:
"Settings","08/15/22","09/15/22",10/15/22
"Users",0,0,7
"Computers",0,1,52
"SomeValue1",0,2,22
"SomeValue2",0,2,22
"SomeValue3",0,2,22
"Stat1",0,10,4
"Stat2",7,0,4
"Stat3",0,0,4
Edit:
If you want the date between quotes, replace $row = $row + "," + $test."$key" with this:
if($key -eq "Settings"){
$row = $row + "," + '"' + $test."$key" + '"'
}else{
$row = $row + "," + $test."$key"
}

Answer 2

This idea is pretty terrible idea, as you stated, "grow to the right" is definitely not a good approach and you should consider a better way of doing it, data should always expand vertically.
As for the solution, you can create new columns easily with Select-Object and dynamically generated calculated properties.
Note, this should definitely not be considered an efficient approach. This will be slow because Select-Object is slow.
function Add-Column {
param(
[Parameter(ValueFromPipeline, DontShow, Mandatory)]
[object] $InputObject,
[Parameter(Mandatory)]
[string] $ColumnName,
[Parameter(Mandatory)]
[string] $ReferenceProperty,
[Parameter(Mandatory)]
[hashtable] $Values
)
begin {
$calculatedProp = #{ N = $ColumnName }
}
process {
$calculatedProp['E'] = { 0 }
if($value = $InputObject.$ReferenceProperty) {
if($Values.ContainsKey($value)) {
$calculatedProp['E'] = { $Values[$value] }
}
}
$InputObject | Select-Object *, $calculatedProp
}
}
Usage
Import-Csv path\to\csv | Add-Column -ColumnName '09/15/22' -ReferenceProperty Settings -Values #{
users = "7"
SomeBSValue = "22"
Computers = "52"
}
Result
Settings 08/15/22 09/15/22
-------- -------- --------
Users 0 7
Computers 0 52
SomeValue1 0 0
SomeValue2 0 0
SomeValue3 0 0
Stat1 0 0
Stat2 7 0
Stat3 0 0
SomeBSValue 1 22
This function allows then pipe into Export-Csv at ease:
Import-Csv path\to\csv | Add-Column ... | Export-Csv path\to\newCsv