Output multiple .csv files appending the .csv file name as the source folder name with powershell-CodePudding

I have 30 folders. Each folder contains 22 .text files. I am trying to get the filenames and row count of each .text files and output it in a .csv file, appending the name of the .csv file with the name of each subfolder.

The script I made works but it will pull all the .text files from all subfolders and output it in a single .csv file.

Any idea how I can create one .csv file per subfolder ?


$results = Get-ChildItem "C:\Users\testserver\Documents\logfiles\*.txt" -Recurse | % { $_ | select name, @{n="lines";e={get-content $_ | measure-object -line | select -expa lines } } } | Select-Object name, lines

$results | Export-Csv "C:\Users\testserver\Documents\results.csv" -notype

CodePudding user response：

Use the Group-Object cmdlet to process the files grouped by the directory they reside in:

$inDir = 'C:\Users\testserver\Documents\logfiles'
$outDir = 'C:\Users\testserver\Documents'

Get-ChildItem -File -Recurse -Filter *.txt $inDir |
  Group-Object DirectoryName | 
    ForEach-Object {
      $outFile = Join-Path $outDir "results-$(Split-Path -Leaf $_.Name).csv"
      $_.Group |
        Select-Object Name, @{ n="Lines"; e={ (Get-Content $_.FullName | Measure-Object -Line).lines } } |
          Export-Csv -NoTypeInformation $outFile
    }

The above creates output files such as results-foo.csv in $outDir, where foo is the name of a subdirectory containing *.txt files.

Note that the assumption is that no two subdirectories in the target $inDir directory tree have the same name; more work is needed if you need to handle such collisions, such as reflecting the relative paths in the file name, with \ replaced with a substitute char.