Powershell - need to parse a text file and past into email - email

Question

I have a very basic text file containing usernames. I need to be able to search for every section that begins with "User bsmith" (for example), copy all the text that follows and end at the line previous to the next "User XXXX" (User tford, for example below). I will then be pasting the copied text in the body of an email. I have already written the Powershell to create the email and send it. I just need to write the portion to grab the text from the text file listed. Still new to programming, so having a hard time determining what way to search and copy from the .txt file and paste into email body. The svnlog.txt is the file being parsed. I can get all content, but need to be able to search for sections for each "User XXXX" to the next "User XXX".
$olComObject = new-object -comobject outlook.application
$svn = (Get-Content C:\Dev\Powershell\svnlog.txt) -join "`n"
$NewMail = $olComObject.CreateItem(0)
$NewMail.Subject = "Testing Voting Options"
$NewMail.Body = "Please use the attached voting buttons in this email to acknowledge or reject if the following user permissions are correct!" + "`n" + $svn
$NewMail.To = "bsmith#company.com;tford#company.com"
$NewMail.VotingOptions = "Accept;Reject"
$NewMail.Send()
I was thinking of possibly searching for the individual "User XXX" group using the following, but again I am not sure if this is the right route to go or will even work.
$fullsvn = Get-Content "C:\Dev\Powershell\svnlog.txt"
Out-File () "C:\Dev\Powershell\svnlogBSmith.txt"
Text file content format is as follows.
User BSmith
repos
[functionalSpecs:/]
#admins = rw
#dba = r
#hca = rw
#businessAnalysts = rw
#qa = rw
#portal = rw
#intercept = rw
#pmo = rw
[restrictedDocs:/softtek]
#restrictedDocs = rw
Groups
BusinessAnalysts = ajones, pjohnson, ssmith, rjackson,
#pmo = lferguson
User tford
Ok. So I tried TessellatingHeckler's route. However it doesn't seem to be Tagging the lines with the username (in this instance it's TKelems instead of BSmith). Here is my current code and an example of the output I get in the console.
Get-Content C:\Dev\Powershell\svnlog.txt | foreach {
if ($_ -match '^User (.*)') {
$tag = $matches[1]
}
[Management.Automation.PSObject]#{
Tag=$tag
Item=$_
}
}
$TKelems | Where Tag -EQ 'tkelems'
Out-File "C:\Dev\Powershell\svnlogTKelems.txt" -InputObject $TKelems
Output:
Name Value
---- -----
Tag
Item --------------------------------------------------------------------------------
Tag
Item User
Tag
Item tkelems:
Tag
Item
Tag
Item Repos
Tag
Item --------------------
Tag
Item [functionalSpecs:/]
Tag
Item #admins = rw
Tag
Item #eclas = rw
Tag
Item #cfis = rw
Tag
Item #transMgmt = rw
Tag
Item #dba = r
Tag
Item #hca = rw
Tag
Item #businessAnalysts = rw
Tag
Item #qa = rw
Tag
Item #portal = rw
Tag
Item #intercept = rw
Tag
I also forgot to add the content of the Ou-File created in the last line. Not sure what to make of this either. Basically, no tags showing in the console, and the output file shows something totally different (See below):
IsReadOnly : False
IsFixedSize : False
IsSynchronized : False
Keys : {Tag, Item}
Values : {$null, --------------------------------------------------------------------------------}
SyncRoot : System.Object
Count : 2
IsReadOnly : False
IsFixedSize : False
IsSynchronized : False
Keys : {Tag, Item}
Values : {$null, User}
SyncRoot : System.Object
Count : 2

Answer 1

One option:
$textfile = 'c:\testfiles\test.txt'
$Search = 'BSmith'
$Found = (Get-Content $textfile -Raw) -split '(?ms)^(?=User)' -like "User $Search*"
That will read the text file in as a single string, then split it at the lines starting with "User", and filter out the ones that don't match user name in $Search.
You'll need to add error trapping to check for $found being empty, or returning multiple results and handle to suit.

Answer 2

something like this could be a starting point
$usernameregex = 'bsmith'
$found = $false
Get-Content "C:\Dev\Powershell\svnlog.txt" | % {
if (!$found -and $_ -match "^user $username$") {
$found = $true
Write-Output ''
Write-Output $_
} elseif ($found -and $_ -match '^user ') {
if (!($_ -match "^user $username$")) {
$found = $false
} elseif ($_ -match "^user $username$") {
Write-Output ''
Write-Output $_
} else {
Write-Output ''
}
} elseif ($found) {
Write-Output $_
}
} | select -skip 1 | Set-Content "C:\Dev\Powershell\svnlogBSmith.txt"
to check multiple names, change the $usernameregex to this
$usernameregex = (bsmith|tford)

Answer 3

I've seen this "I want lines in a file from [section header] to the next [section header]" before, and mjolinor's answer with the multiline regex is short and effective, but I don't think it's very clear what it's doing and why. And it won't work as nicely on a general array of strings without -joining them first.
As an alternative I quite like the concept of tagging each line, then you can group or filter by the tag. e.g. pull out the username and tag each line with the last-seen username:
gc .\log.txt | foreach {
if ($_ -match '^User (.*)') {
$tag = $matches[1]
}
[pscustomobject]#{
Tag=$tag
Item=$_
}
}
What it does with your file is this:
Tag Item
---- ----
BSmith User BSmith
BSmith repos
BSmith [functionalSpecs:/]
BSmith #admins = rw
BSmith #dba = r
BSmith #hca = rw
BSmith #businessAnalysts = rw
BSmith #qa = rw
BSmith #portal = rw
BSmith #intercept = rw
BSmith #pmo = rw
BSmith [restrictedDocs:/softtek]
BSmith #restrictedDocs = rw
BSmith Groups
BSmith BusinessAnalysts = ajones, pjohnson, ssmith, rjackson,
BSmith #pmo = lferguson
tford User tford
tford
Then you can easily:
$that | Where Tag -eq 'BSmith'
or
$that | group Tag
It could tag lines with any regex match for section delimiters.
[Edit: Added ^ to the regex, suggested by #mjolinor ]