ConvertFrom-StringData clear hash-table - powershell

Question

I am attempting to parse multiple mails in a specific mail folder, to eventually create a csv file with the information.
If I run the script on one user, the output I get is perfect;
Name Value
---- -----
Kontor. HR
Prøvetid Ja
Blomster på dagen Nej
Telefonnr. 85789845
Adresse Hovedgade 13
Ansættes i stedet for Ninette Kock
Navn Marianne Pedersen
Etage 1. sal
Fri telefon Ja
Benyttet navn Marianne Pedersen
Medarbejdertype Fastansat
Overtagelse af mobilnr. Ja
Placering Sydøst
Ansættelsesdato 01-01-2020
Stilling Fuldmægtig
Lokalnr. [phone]
Besked til IT-centret
Antal timer 37
Journalistpraktikant Nej
Tidl. ansat Nej
Initialer MAPE
Blomster anden dag 02-01-2020
Postnr/By 2630 Taastrup
Cpr. nr. XXXX
The problem is when I try to parse a second mail, the first mail will still get parsed correctly, but the second one will throw an error;
ConvertFrom-StringData : Data item 'Medarbejdertype' in line 'Medarbejdertype=Fastansat' is
already defined.
If I try to parse the second mail as the first mail, it will parse perfectly.
Problem happens when I try to parse more then one.
I have tried to clear the hash-table like so;
$data.clear()
clear-variable -name data
$data = #{}
But it somehow seems that the hash-table is not emptied?
The code I am using:
Clear-Host
$navne = "MAJE", "ASHO"
# create an instance of Outlook
Add-Type -AssemblyName "Microsoft.Office.Interop.Outlook"
$outlook = New-Object -ComObject Outlook.Application
$nameSpace = $outlook.GetNameSpace("MAPI")
$html = New-Object -ComObject "HTMlFile"
# Opens and save all mails in Nye Til Oprettelse in $mails.
$mails = $nameSpace.Folders.Item('hotline').Folders.Item('Indbakke').Folders.Item('#HOVSA''er').Folders.Item('#01Nye til oprettelse').Items | Select-Object -Property subject, htmlbody
# Loop over $mails, check if $navn = subject
foreach ($navn in $navne) {
foreach ($mail in $mails) {
if($mail | Where-Object {$_.subject -match "$Navn"}) {
#$mail.HTMLBody
# write data to $html, then find all <table> tags and parse that text.
$html.IHTMLDOCUMENT2_write($mail.HTMLBody)
$data = $html.all.tags("table") | % innertext
$data = $data -split '\r?\n' -match '^[^:]+:[^:]+$' -replace ':\s*', '=' -join "`n" | ConvertFrom-StringData
$navn
$data
$data = #[]
}
}
}

Answer 1

You're appending each email to the $html document using $html.IHTMLDOCUMENT2_write($mail.HTMLBody), but you only initialise it once at the top of your script, so each time your foreach ($mail in $mails) loops you add another email to the document.
That means the second time the loop iterates there are two emails in the $html document and you're getting an error about duplicate data items as a result.
Try moving $html = New-Object -ComObject "HTMlFile" inside your foreach loop so it reads:
foreach ($mail in $mails) {
...
$html = New-Object -ComObject "HTMlFile"
$html.IHTMLDOCUMENT2_write($mail.HTMLBody)
...
}
P.S.: You might also want to look at using a non-COM library like the HTML Agility Pack to do your html processing - it'll potentially be a bit faster and more reliable (and portable) in an automation script than COM is.