3 INSERT INTO config.upgrade_log (version) VALUES ('0446'); -- gmc
5 CREATE OR REPLACE FUNCTION public.naco_normalize( TEXT, TEXT ) RETURNS TEXT AS $func$
6 use Unicode::Normalize;
9 # When working with Unicode data, the first step is to decode it to
10 # a byte string; after that, lowercasing is safe
11 my $txt = lc(decode_utf8(shift));
15 $txt =~ s/\pM+//go; # Remove diacritics
17 # remove non-combining diacritics
18 # this list of characters follows the NACO normalization spec,
19 # but a looser but more comprehensive version might be
20 # $txt =~ s/\pLm+//go;
21 $txt =~ tr/\x{02B9}\x{02BA}\x{02BB}\x{02BC}//d;
23 $txt =~ s/\xE6/AE/go; # Convert ae digraph
24 $txt =~ s/\x{153}/OE/go;# Convert oe digraph
25 $txt =~ s/\xFE/TH/go; # Convert Icelandic thorn
27 $txt =~ tr/\x{2070}\x{2071}\x{2072}\x{2073}\x{2074}\x{2075}\x{2076}\x{2077}\x{2078}\x{2079}\x{207A}\x{207B}/0123456789+-/;# Convert superscript numbers
28 $txt =~ tr/\x{2080}\x{2081}\x{2082}\x{2083}\x{2084}\x{2085}\x{2086}\x{2087}\x{2088}\x{2089}\x{208A}\x{208B}/0123456889+-/;# Convert subscript numbers
30 $txt =~ tr/\x{0251}\x{03B1}\x{03B2}\x{0262}\x{03B3}/AABGG/; # Convert Latin and Greek
31 $txt =~ tr/\x{2113}\xF0\x{111}\!\"\(\)\-\{\}\<\>\;\:\.\?\xA1\xBF\/\\\@\*\%\=\xB1\+\xAE\xA9\x{2117}\$\xA3\x{FFE1}\xB0\^\_\~\`/LDD /; # Convert Misc
32 $txt =~ tr/\'\[\]\|//d; # Remove Misc
34 if ($sf && $sf =~ /^a/o) {
35 my $commapos = index($txt,',');
37 if ($commapos != length($txt) - 1) {
38 my @list = split /,/, $txt;
39 my $first = shift @list;
40 $txt = $first . ',' . join(' ', @list);
49 $txt =~ s/\s+/ /go; # Compress multiple spaces
50 $txt =~ s/^\s+//o; # Remove leading space
51 $txt =~ s/\s+$//o; # Remove trailing space
53 # Encoding the outgoing string is good practice, but not strictly
54 # necessary in this case because we've stripped everything from it
55 return encode_utf8($txt);
56 $func$ LANGUAGE 'plperlu' STRICT IMMUTABLE;