<#
  Purges illegal XML characters from XML files in provided folder (and its subfolders).
  Encodes out file as UTF8 without BOM.
  Peter Tyrrell, Andornot
#>

param (
  [string]$indir
)

$files = ls $indir -filter "*.xml" -recurse
$utf8nobom = new-object System.Text.UTF8Encoding($false)  # write UTF8 output without BOM as Java xml parsers choke on it


foreach ($file in $files) {
  $f = (get-content $file.FullName) -replace '[^\u0009\u000a\u000d\u0020-\uD7FF\uE000-\uFFFD]', 'XXX'
  [System.IO.File]::WriteAllLines($file.FullName, $f, $utf8nobom)
}