How do you make sure that all your source files are using the same encoding? With PowerShell of course!

Usually, the encoding you are using on your source files are not important. But if you are compiling on different platforms, you could run into unexpected problems if the source files are using “ANSI” and you have some special characters in the code. I have had unit test that run perfectly fine on Windows, but when they were running on Linux, they failed because of this.

Setting UTF-8 on multiple files

The little script below finds all cs-files in a directory (except these in the obj-folders), and checks if they are using UTF-8 by reading the byte order mark from each file. And if not, the files are read and saved again but with UTF-8.

PowerShell 
$files = Get-ChildItem -Filter *.cs -Recurse | ? { $_.FullName -notmatch '\\obj\\' }

foreach($file in $files) {
  $bytes = [byte[]] (Get-Content ($file.FullName) -Encoding byte -ReadCount 3 -TotalCount 3)
  if( $bytes.Length -lt 3 -or
      $bytes[0].ToString("x2") -ne "ef" -or
      $bytes[1].ToString("x2") -ne "bb" -or
      $bytes[2].ToString("x2") -ne "bf" ) {

    Write-Output "Updating: $($file.FullName)"
    $content = Get-Content "$($file.FullName)"
    Set-Content "$($file.FullName)" -Value $content -Encoding UTF8
  }
}

You might consider doing this for also for cstml, razor and other files extensions that you find suitable.

Summary

This little script is a simple solution to change encoding on multiple files at the same time. Maybe a bit too simple you may thin if you do not like byte order marks. But in this scenario, I do not worry about that.