To modify keys of a hash, it is necessary to copy it. The keys can be modified in the process of the copy by using 'map' like this:

my %HashCopy = map {tr/ /_/? $_:$_ => $OrigHash{$_}} keys %OrigHash; 
  #  Above line copies %OrigHash and Replaces Spaces with Underscore  
+in hash Key
[download]

Since the 'tr' function returns the number of translations, and we do not need that number, it is essentially discarded, using the ternary "?" operator. The KEY we need is contained in "$_", with is returned under both TRUE and FALSE conditions of the "?" operator. "$_" contains the translated KEY we need.

The VALUE is obtained simply by indexing %OrigHash.

If anyone has an insight into the inner workings of "map" - specificaly, how does it know whether to use the translated or untranslated version of "$_" when interpreting "=> $OrigHash{$_}", please post that info here. The bottom line is that it DOES work using the original $_, not the translated one, but WHY ? Is the right-side of the "=>" evaluated first ?

Originally posted as a Categorized Answer.