Yes, that looks like a good (simple) solution.  Interestingly though

  s/(^|[^[:alpha:]])$from($|[^[:alpha:]])/$1$to$2/i;
[download]

only works for me when I use locale (which I may not necessarily want to do in this case), while

  s/(^|[^\p{IsAlpha}])$from($|[^\p{IsAlpha}])/$1$to$2/i;
[download]

does work without...

My solution will also make "MÄR" work

...presuming other changes will be made as well — i.e. adding another list of month abbreviations to the definition of $$d{"month_abb"}=...

Sounds like you forgot to use utf8::upgrade($from);.

only works when I use locale

No, using unicode semantics is enough.

presuming other changes will be made as well

No, using unicode semantics is enough.

use HTML::Entities qw( decode_entities );

use locale qw();

my $lc = decode_entities('ä');
my $uc = decode_entities('Ä');

utf8::downgrade($uc);

for (0..2) {
   if ($_ == 0) {
      utf8::downgrade($lc);
      locale->unimport();
      print("Byte Semantics\n");
      print("--------------\n");
   }
   elsif ($_ == 1) {
      utf8::downgrade($lc);
      locale->import();
      print("Locale Semantics\n");
      print("----------------\n");
   }
   elsif ($_ == 2) {
      utf8::upgrade($lc);
      locale->unimport();
      print("Unicode Semantics\n");
      print("-----------------\n");
   }

   if ($lc =~ /^\Q$uc\E\z/) {
      print("case sensitive match\n");
   }
   elsif ($lc =~ /^\Q$uc\E\z/i) {
      print("case insensitive match\n");
   }
   else {
      print("no match\n");
   }

   if ($lc =~ /^[[:alpha:]]\z/) {
      print("[:alpha:]\n");
   } else {
      print("Not [:alpha:]\n");
   }

   if ($lc =~ /^[\p{IsAlpha}]\z/) {
      print("\\p{IsAlpha}\n");
   } else {
      print("Not \\p{IsAlpha}\n");
   }

   print("\n");
}
[download]

Byte Semantics
--------------
no match
Not [:alpha:]
\p{IsAlpha}

Locale Semantics
----------------
no match
Not [:alpha:]
\p{IsAlpha}

Unicode Semantics
-----------------
case insensitive match
[:alpha:]
\p{IsAlpha}
[download]

Sounds like you forgot to use utf8::upgrade($from);

No, I tried this:

#!/usr/bin/perl

$Lang{$L}{"Repl"} = { "m" => "Monat" };  # mimic Date::Manip
$_ = "Mär";

print "before: $_\n";

foreach $from (keys %{ $Lang{$L}{"Repl"} }) {
  $to=$Lang{$L}{"Repl"}{$from};

  utf8::upgrade($from);  # Use Unicode semantics
  s/(^|[^[:alpha:]])$from($|[^[:alpha:]])/$1$to$2/i;
}    
print "after:  $_\n";
[download]

which prints

before: Mär
after:  Monatär
[download]

Using

  s/(^|[^\p{IsAlpha}])$from($|[^\p{IsAlpha}])/$1$to$2/i;
[download]

in place of the above substitution does work fine, though (as does adding use locale to the [:alpha:] version):

before: Mär
after:  Mär
[download]




presuming other changes will be made as well

No, using unicode semantics is enough.

I was referring to "MÄR" also working (in addition to "Mär") in the context of Date::Manip...  which it won't unless that abbreviation is also being set up in the respective $$d{"month_abb"}=...