I am trying to write a script to pull all e-mail addresses but I am finding that the output of my first get command is adding alot of white space to the result text file
Get-AdUser -Filter * -Properties * | Select EmailAddress | Out-File "C:\T2\EMailAddresses.txt"
Can anyone tell me what I am doing wrong here?
You are seeing whitespace because not every AD account has a value assigned to that property in your environment so it appears as a blank line, I get the same thing upon a quick test.
This should help.
$Emails = Get-ADUser -Filter * -Properties EmailAddress
$Emails | select EmailAddress | Where {$_.EmailAddress -ne $null} | Out-File "C:\T2\EMailAddresses.txt"
$null on the LHS of equality / inequality operations, because it makes a difference with collections; compare $null -eq (1, 2) to (1, 2) -eq $null, for instance.$null; if they have "no value", their value is the empty string; e.g.: [string] $str = $null; $null -eq $str yields $false. String-typed object properties, however, may contain a true $null, and it sounds like that is the case here with the objects output by Get-ADUser, right? (I can't personally verify).
.EmailAddress property contains an actual $null when not set. In my answer, by casting to an array of[string]s first, $null is converted to '', which is why using -ne '' to filter out unset properties works.td;dr
The following writes all email addresses to the target file, ignoring AD users that don't have one:
([string[]] (Get-AdUser -Filter * -Properties EmailAddress).EmailAddress) -ne '' |
Set-Content C:\T2\EMailAddresses.txt
By writing just the - non-empty - .EmailAddress property values to the file, you're avoiding the problems that stem from saving for-display formatted object representations, which is what your attempt does (see below).
Note that -ne '' acts as a filter here, because its LHS operand is an array; that is, the result of the operation is the sub-array of those LHS elements that aren't the empty string ('').
As for what you tried:
By using Out-File in combination with objects subject to PowerShell's output formatting system, you're saving a for-display representation of your objects to a file, which, in the case at hand includes a table header, a leading and a trailing blank line and - in Windows PowerShell (but no longer in PowerShell (Core) 7+) - right-space-padding to the full console-line width of each line.
Even though you're only asking for one property - EmailAddress - Select-Object outputs not just that property's value for each input object, but a [pscustomobject] instance with an .EmailAddress property, and the resulting objects are implicitly formatted with Format-Table.
To get just the EmailAddress property values, use Select-Object -ExpandProperty EmailAddress. The resulting string values are not subject to formatting, so your command would work as intended except that it would still include $null values from those AD users who happen not to have a value stored in their .EmailAddress property.
While it often won't matter, for string input it's slightly faster to use Set-Content than Out-File / >; note that in Windows PowerShell you'll end up with different character encodings by default (ANSI vs. UTF-16 LE a.k.a "Unicode") - use the -Encoding parameter as needed; PowerShell Core 7+ fortunately now consistently defaults to BOM-less UTF-8.
The - faster, but more potentially memory-intensive - alternative to using Select-Object -ExpandProperty EmailAddress for extracting the EmailAddress property values is to use member-access enumeration ((...).EmailAddress, as shown above).
.Trim()removes leading and trailing whitespace (all forms), so it wouldn't solve the problem of interior blanks.