Extract the end of every lines from multiple files - powershell

Question

I have several txt files distributed in several sub-folders.
This is what a file looks like:
Data file. version 01.10.
1
8
*
DAT\Trep\Typ10
DAT\Trep\Typ12
DAT\Trep\Typ13
what I would like to do is to extract only the part after the last "\" in order to get something like this:
Typ10 FileName.txt Path
Typ12 FileName.txt Path
Typ13 FileName.txt Path
...
I tried the following
Get-ChildItem -Path 'D:\MyData\*.txt' -Recurse | ForEach-Object {Write-Output $_; $data=Get-Content $_}
$data = $data -replace '.*\\'
$data
it works well for a single file but not with several (-recurse).
Being a powershell beginner I can't figure out how to improve my script.
I also tried to add this to get the result shown above in my post, but that doesn't work either.
Select-Object -Property #{Name = 'Result list'; Expression = { $data }}, Filename, Path
Thanks in advance for your kind help

Answer 1

Use Select-String:
Get-ChildItem -Path D:\MyData\*.txt -Recurse -File |
Select-String '^.+\\(.+)' |
ForEach-Object {
[pscustomobect] #{
Result = $_.Matches.Groups[1].Value
FileName = $_.FileName
Path = $_.Path
}
}
As for your desire to exclude certain folders during recursive traversal:
Unfortunately, Get-ChildItem -Exclude only excludes the matching folders themselves, not also their content. There are two relevant feature requests to potentially overcome this limitation in the future:
GitHub issue #4126 asks for path patterns to be supported too in the future.
GitHub issue #15159 proposes a new subtree-exclusion parameter, such as
-ExcludeRecursive.
For now, a different approach with post-filtering based on Where-Object is required, using folder names Folder1 and Folder2 as examples:
Get-ChildItem -Path D:\MyData\*.txt -Recurse |
Where-Object FullName -NotLike *\Folder1\* |
Where-Object FullName -NotLike *\Folder2\* |
Select-String '^.+\\(.+)' |
ForEach-Object {
[pscustomobect] #{
Result = $_.Matches.Groups[1].Value
FileName = $_.FileName
Path = $_.Path
}
}
For a more flexible, cross-platform approach based on regex matching (which is invariably more complex), see the bottom section of this answer.