> > use Text::Unicode::Normalize;
> >
> > $stringNFD = NFD($string); # Normalization Form D
> > $stringNFC = NFC($string); # Normalization Form C
> > $stringNFKD = NFKD($string); # Normalization Form KD
> > $stringNFKC = NFKC($string); # Normalization Form KC
>
> a normalize function instead of these 4 functions with a
> form parameter, e.g.
>
> my $normalized = normalize( form => 'C' );
How about the following interface?
| $normalized_string = normalize($raw_string)
|
| You can use this function only if the normalization form
| you require is specified in the C<use> statement:
|
| use Text::Unicode::Normalize 'C'; # Normalization Form C
|
| or for clarity, say:
|
| use Text::Unicode::Normalize form => 'C'; # Normalization Form C
|
| and you can use C<normalize> as an alias for C<NFC> or another:
|
| $normalized_string = normalize($raw_string);
|
| As the form name, one of the following names will be accepted.
|
| 'C' or 'NFC' for Normalization Form C
| 'D' or 'NFD' for Normalization Form D
| 'KC' or 'NFKC' for Normalization Form KC
| 'KD' or 'NFKD' for Normalization Form KD
> It should accept the short forms (C, D, KC, KD) as alias
> for the NF.. There is currently no Text::Unicode tree on
> CPAN but there is a Unicode:: tree and it fits quite well
> there. The normalize function increases readabilty and
> looks nicer.
>
> Is it expectable, that Perl will normalize everything it
> puts out by itself or will we have to use this module?
Regards, SADAHIRO Tomoyuki
