[p5sagit/p5-mst-13.2.git] / ext / Unicode / Normalize / Normalize.pm

package Unicode::Normalize;

BEGIN {
    if (ord("A") == 193) {
       die "Unicode::Normalize not ported to EBCDIC\n";
    }
}

use 5.006;
use strict;
use warnings;
use Carp;

our $VERSION = '0.15';
our $PACKAGE = __PACKAGE__;

require Exporter;
require DynaLoader;
require AutoLoader;

our @ISA = qw(Exporter DynaLoader);
our @EXPORT = qw( NFC NFD NFKC NFKD );
our @EXPORT_OK = qw(
    normalize decompose reorder compose
    checkNFD checkNFKD checkNFC checkNFKC check
    getCanon getCompat getComposite getCombinClass
    isExclusion isSingleton isNonStDecomp isComp2nd isComp_Ex
    isNFD_NO isNFC_NO isNFC_MAYBE isNFKD_NO isNFKC_NO isNFKC_MAYBE
);
our %EXPORT_TAGS = (
    all       => [ @EXPORT, @EXPORT_OK ],
    normalize => [ @EXPORT, qw/normalize decompose reorder compose/ ],
    check     => [ qw/checkNFD checkNFKD checkNFC checkNFKC check/ ],
);

bootstrap Unicode::Normalize $VERSION;

use constant COMPAT => 1;

sub NFD  ($) { reorder(decompose($_[0])) }
sub NFKD ($) { reorder(decompose($_[0], COMPAT)) }
sub NFC  ($) { compose(reorder(decompose($_[0]))) }
sub NFKC ($) { compose(reorder(decompose($_[0], COMPAT))) }

sub normalize($$)
{
    my $form = shift;
    $form =~ s/^NF//;
    return
	$form eq 'D'  ? NFD ($_[0]) :
	$form eq 'C'  ? NFC ($_[0]) :
	$form eq 'KD' ? NFKD($_[0]) :
	$form eq 'KC' ? NFKC($_[0]) :
      croak $PACKAGE."::normalize: invalid form name: $form";
}

sub check($$)
{
    my $form = shift;
    $form =~ s/^NF//;
    return
	$form eq 'D'  ? checkNFD ($_[0]) :
	$form eq 'C'  ? checkNFC ($_[0]) :
	$form eq 'KD' ? checkNFKD($_[0]) :
	$form eq 'KC' ? checkNFKC($_[0]) :
      croak $PACKAGE."::check: invalid form name: $form";
}

1;
__END__

=head1 NAME

Unicode::Normalize - normalized forms of Unicode text

=head1 SYNOPSIS

  use Unicode::Normalize;

  $NFD_string  = NFD($string);  # Normalization Form D
  $NFC_string  = NFC($string);  # Normalization Form C
  $NFKD_string = NFKD($string); # Normalization Form KD
  $NFKC_string = NFKC($string); # Normalization Form KC

   or

  use Unicode::Normalize 'normalize';

  $NFD_string  = normalize('D',  $string);  # Normalization Form D
  $NFC_string  = normalize('C',  $string);  # Normalization Form C
  $NFKD_string = normalize('KD', $string);  # Normalization Form KD
  $NFKC_string = normalize('KC', $string);  # Normalization Form KC

=head1 DESCRIPTION

=head2 Normalization Forms

=over 4

=item C<$NFD_string = NFD($string)>

returns the Normalization Form D (formed by canonical decomposition).

=item C<$NFC_string = NFC($string)>

returns the Normalization Form C (formed by canonical decomposition
followed by canonical composition).

=item C<$NFKD_string = NFKD($string)>

returns the Normalization Form KD (formed by compatibility decomposition).

=item C<$NFKC_string = NFKC($string)>

returns the Normalization Form KC (formed by compatibility decomposition
followed by B<canonical> composition).

=item C<$normalized_string = normalize($form_name, $string)>

As C<$form_name>, one of the following names must be given.

  '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

=back

=head2 Decomposition and Composition

=over 4

=item C<$decomposed_string = decompose($string)>

=item C<$decomposed_string = decompose($string, $useCompatMapping)>

Decompose the specified string and returns the result.

If the second parameter (a boolean) is omitted or false, decomposes it
using the Canonical Decomposition Mapping.
If true, decomposes it using the Compatibility Decomposition Mapping.

The string returned is not always in NFD/NFKD.
Reordering may be required.

    $NFD_string  = reorder(decompose($string));       # eq. to NFD()
    $NFKD_string = reorder(decompose($string, TRUE)); # eq. to NFKD()

=item C<$reordered_string  = reorder($string)>

Reorder the combining characters and the like in the canonical ordering
and returns the result.

E.g., when you have a list of NFD/NFKD strings,
you can get the concatenated NFD/NFKD string from them, saying

    $concat_NFD  = reorder(join '', @NFD_strings);
    $concat_NFKD = reorder(join '', @NFKD_strings);

=item C<$composed_string   = compose($string)>

Returns the string where composable pairs are composed.

E.g., when you have a NFD/NFKD string,
you can get its NFC/NFKC string, saying

    $NFC_string  = compose($NFD_string);
    $NFKC_string = compose($NFKD_string);

=back

=head2 Quick Check

(see Annex 8, UAX #15; F<DerivedNormalizationProperties.txt>)

The following functions check whether the string is in that normalization form.

The result returned will be:

    YES     The string is in that normalization form.
    NO      The string is not in that normalization form.
    MAYBE   Dubious. Maybe yes, maybe no.

=over 4

=item C<$result = checkNFD($string)>

returns YES (1) or NO (empty string).

=item C<$result = checkNFC($string)>

returns YES (1), NO (empty string), or MAYBE (undef).

=item C<$result = checkNFKD($string)>

returns YES (1) or NO (empty string).

=item C<$result = checkNFKC($string)>

returns YES (1), NO (empty string), or MAYBE (undef).

=item C<$result = check($form_name, $string)>

returns YES (1), NO (empty string), or MAYBE (undef).

C<$form_name> is alike to that for C<normalize()>.

=back

B<Note>

In the cases of NFD and NFKD, the answer must be either C<YES> or C<NO>.
The answer C<MAYBE> may be returned in the cases of NFC and NFKC.

A MAYBE-NFC/NFKC string should contain at least
one combining character or the like.
For example, C<COMBINING ACUTE ACCENT> has
the MAYBE_NFC/MAYBE_NFKC property.
Both C<checkNFC("A\N{COMBINING ACUTE ACCENT}")>
and C<checkNFC("B\N{COMBINING ACUTE ACCENT}")> will return C<MAYBE>.
Though, C<"A\N{COMBINING ACUTE ACCENT}"> is not in NFC 
(its NFC is C<"\N{LATIN CAPITAL LETTER A WITH ACUTE}">),
while C<"B\N{COMBINING ACUTE ACCENT}"> is in NFC.

If you want to check exactly, compare the string with its NFC/NFKC; i.e.,

    $string eq NFC($string)    # more thorough than checkNFC($string)
    $string eq NFKC($string)   # more thorough than checkNFKC($string)

=head2 Character Data

These functions are interface of character data used internally.
If you want only to get Unicode normalization forms, you don't need
call them yourself.

=over 4

=item C<$canonical_decomposed = getCanon($codepoint)>

If the character of the specified codepoint is canonically
decomposable (including Hangul Syllables),
returns the B<completely decomposed> string canonically equivalent to it.

If it is not decomposable, returns undef.

=item C<$compatibility_decomposed = getCompat($codepoint)>

If the character of the specified codepoint is compatibility
decomposable (including Hangul Syllables),
returns the B<completely decomposed> string compatibility equivalent to it.

If it is not decomposable, returns undef.

=item C<$codepoint_composite = getComposite($codepoint_here, $codepoint_next)>

If two characters here and next (as codepoints) are composable
(including Hangul Jamo/Syllables and Composition Exclusions),
returns the codepoint of the composite.

If they are not composable, returns undef.

=item C<$combining_class = getCombinClass($codepoint)>

Returns the combining class of the character as an integer.

=item C<$is_exclusion = isExclusion($codepoint)>

Returns a boolean whether the character of the specified codepoint
is a composition exclusion.

=item C<$is_singleton = isSingleton($codepoint)>

Returns a boolean whether the character of the specified codepoint is
a singleton.

=item C<$is_non_startar_decomposition = isNonStDecomp($codepoint)>

Returns a boolean whether the canonical decomposition
of the character of the specified codepoint
is a Non-Starter Decomposition.

=item C<$may_be_composed_with_prev_char = isComp2nd($codepoint)>

Returns a boolean whether the character of the specified codepoint
may be composed with the previous one in a certain composition
(including Hangul Compositions, but excluding
Composition Exclusions and Non-Starter Decompositions).

=back

=head2 EXPORT

C<NFC>, C<NFD>, C<NFKC>, C<NFKD>: by default.

C<normalize> and other some functions: on request.

=head1 AUTHOR

SADAHIRO Tomoyuki, E<lt>SADAHIRO@cpan.orgE<gt>

  http://homepage1.nifty.com/nomenclator/perl/

  Copyright(C) 2001-2002, SADAHIRO Tomoyuki. Japan. All rights reserved.

  This program is free software; you can redistribute it and/or 
  modify it under the same terms as Perl itself.

=head1 SEE ALSO

=over 4

=item http://www.unicode.org/unicode/reports/tr15/

Unicode Normalization Forms - UAX #15

=item http://www.unicode.org/Public/UNIDATA/DerivedNormalizationProperties.txt

Derived Normalization Properties

=back

=cut
Commit	Line	Data
ac5ea531	1	package Unicode::Normalize;
ac5ea531	2
4a2e806c	3	BEGIN {
4a2e806c	4	if (ord("A") == 193) {
8f118dcd	5	die "Unicode::Normalize not ported to EBCDIC\n";
4a2e806c	6	}
	7	}
	8
ac5ea531	9	use 5.006;
	10	use strict;
	11	use warnings;
	12	use Carp;
	13
8f118dcd	14	our $VERSION = '0.15';
ac5ea531	15	our $PACKAGE = __PACKAGE__;
	16
	17	require Exporter;
	18	require DynaLoader;
	19	require AutoLoader;
	20
	21	our @ISA = qw(Exporter DynaLoader);
	22	our @EXPORT = qw( NFC NFD NFKC NFKD );
2a204b45	23	our @EXPORT_OK = qw(
2a204b45	24	normalize decompose reorder compose
8f118dcd	25	checkNFD checkNFKD checkNFC checkNFKC check
	26	getCanon getCompat getComposite getCombinClass
	27	isExclusion isSingleton isNonStDecomp isComp2nd isComp_Ex
	28	isNFD_NO isNFC_NO isNFC_MAYBE isNFKD_NO isNFKC_NO isNFKC_MAYBE
	29	);
	30	our %EXPORT_TAGS = (
	31	all => [ @EXPORT, @EXPORT_OK ],
	32	normalize => [ @EXPORT, qw/normalize decompose reorder compose/ ],
	33	check => [ qw/checkNFD checkNFKD checkNFC checkNFKC check/ ],
2a204b45	34	);
ac5ea531	35
	36	bootstrap Unicode::Normalize $VERSION;
	37
ac5ea531	38	use constant COMPAT => 1;
ac5ea531	39
d85850a7	40	sub NFD ($) { reorder(decompose($_[0])) }
ac5ea531	41	sub NFKD ($) { reorder(decompose($_[0], COMPAT)) }
d85850a7	42	sub NFC ($) { compose(reorder(decompose($_[0]))) }
ac5ea531	43	sub NFKC ($) { compose(reorder(decompose($_[0], COMPAT))) }
	44
	45	sub normalize($$)
	46	{
d85850a7	47	my $form = shift;
	48	$form =~ s/^NF//;
	49	return
	50	$form eq 'D' ? NFD ($_[0]) :
	51	$form eq 'C' ? NFC ($_[0]) :
	52	$form eq 'KD' ? NFKD($_[0]) :
	53	$form eq 'KC' ? NFKC($_[0]) :
	54	croak $PACKAGE."::normalize: invalid form name: $form";
ac5ea531	55	}
ac5ea531	56
8f118dcd	57	sub check($$)
	58	{
	59	my $form = shift;
	60	$form =~ s/^NF//;
	61	return
	62	$form eq 'D' ? checkNFD ($_[0]) :
	63	$form eq 'C' ? checkNFC ($_[0]) :
	64	$form eq 'KD' ? checkNFKD($_[0]) :
	65	$form eq 'KC' ? checkNFKC($_[0]) :
	66	croak $PACKAGE."::check: invalid form name: $form";
	67	}
	68
ac5ea531	69	1;
ac5ea531	70	__END__
2a204b45	71
	72	=head1 NAME
	73
	74	Unicode::Normalize - normalized forms of Unicode text
	75
	76	=head1 SYNOPSIS
	77
	78	use Unicode::Normalize;
	79
8f118dcd	80	$NFD_string = NFD($string); # Normalization Form D
	81	$NFC_string = NFC($string); # Normalization Form C
	82	$NFKD_string = NFKD($string); # Normalization Form KD
	83	$NFKC_string = NFKC($string); # Normalization Form KC
2a204b45	84
	85	or
	86
	87	use Unicode::Normalize 'normalize';
	88
8f118dcd	89	$NFD_string = normalize('D', $string); # Normalization Form D
	90	$NFC_string = normalize('C', $string); # Normalization Form C
	91	$NFKD_string = normalize('KD', $string); # Normalization Form KD
	92	$NFKC_string = normalize('KC', $string); # Normalization Form KC
2a204b45	93
	94	=head1 DESCRIPTION
	95
d85850a7	96	=head2 Normalization Forms
2a204b45	97
	98	=over 4
	99
8f118dcd	100	=item C<$NFD_string = NFD($string)>
2a204b45	101
	102	returns the Normalization Form D (formed by canonical decomposition).
	103
8f118dcd	104	=item C<$NFC_string = NFC($string)>
2a204b45	105
	106	returns the Normalization Form C (formed by canonical decomposition
	107	followed by canonical composition).
	108
8f118dcd	109	=item C<$NFKD_string = NFKD($string)>
2a204b45	110
	111	returns the Normalization Form KD (formed by compatibility decomposition).
	112
8f118dcd	113	=item C<$NFKC_string = NFKC($string)>
2a204b45	114
	115	returns the Normalization Form KC (formed by compatibility decomposition
	116	followed by B<canonical> composition).
	117
8f118dcd	118	=item C<$normalized_string = normalize($form_name, $string)>
2a204b45	119
	120	As C<$form_name>, one of the following names must be given.
	121
	122	'C' or 'NFC' for Normalization Form C
	123	'D' or 'NFD' for Normalization Form D
	124	'KC' or 'NFKC' for Normalization Form KC
	125	'KD' or 'NFKD' for Normalization Form KD
	126
	127	=back
	128
8f118dcd	129	=head2 Decomposition and Composition
	130
	131	=over 4
	132
	133	=item C<$decomposed_string = decompose($string)>
	134
	135	=item C<$decomposed_string = decompose($string, $useCompatMapping)>
	136
	137	Decompose the specified string and returns the result.
	138
	139	If the second parameter (a boolean) is omitted or false, decomposes it
	140	using the Canonical Decomposition Mapping.
	141	If true, decomposes it using the Compatibility Decomposition Mapping.
	142
	143	The string returned is not always in NFD/NFKD.
	144	Reordering may be required.
	145
	146	$NFD_string = reorder(decompose($string)); # eq. to NFD()
	147	$NFKD_string = reorder(decompose($string, TRUE)); # eq. to NFKD()
	148
	149	=item C<$reordered_string = reorder($string)>
	150
	151	Reorder the combining characters and the like in the canonical ordering
	152	and returns the result.
	153
	154	E.g., when you have a list of NFD/NFKD strings,
	155	you can get the concatenated NFD/NFKD string from them, saying
	156
	157	$concat_NFD = reorder(join '', @NFD_strings);
	158	$concat_NFKD = reorder(join '', @NFKD_strings);
	159
	160	=item C<$composed_string = compose($string)>
	161
	162	Returns the string where composable pairs are composed.
	163
	164	E.g., when you have a NFD/NFKD string,
	165	you can get its NFC/NFKC string, saying
	166
	167	$NFC_string = compose($NFD_string);
	168	$NFKC_string = compose($NFKD_string);
	169
	170	=back
	171
	172	=head2 Quick Check
	173
	174	(see Annex 8, UAX #15; F<DerivedNormalizationProperties.txt>)
	175
	176	The following functions check whether the string is in that normalization form.
	177
	178	The result returned will be:
	179
	180	YES The string is in that normalization form.
	181	NO The string is not in that normalization form.
	182	MAYBE Dubious. Maybe yes, maybe no.
	183
	184	=over 4
	185
	186	=item C<$result = checkNFD($string)>
	187
	188	returns YES (1) or NO (empty string).
	189
	190	=item C<$result = checkNFC($string)>
	191
	192	returns YES (1), NO (empty string), or MAYBE (undef).
193
194	=item C<$result = checkNFKD($string)>
195
196	returns YES (1) or NO (empty string).
197
198	=item C<$result = checkNFKC($string)>
199
200	returns YES (1), NO (empty string), or MAYBE (undef).
201
202	=item C<$result = check($form_name, $string)>
203
204	returns YES (1), NO (empty string), or MAYBE (undef).
205
206	C<$form_name> is alike to that for C<normalize()>.
207
208	=back
209
210	B<Note>
211
212	In the cases of NFD and NFKD, the answer must be either C<YES> or C<NO>.
213	The answer C<MAYBE> may be returned in the cases of NFC and NFKC.
214
215	A MAYBE-NFC/NFKC string should contain at least
216	one combining character or the like.
217	For example, C<COMBINING ACUTE ACCENT> has
218	the MAYBE_NFC/MAYBE_NFKC property.
219	Both C<checkNFC("A\N{COMBINING ACUTE ACCENT}")>
220	and C<checkNFC("B\N{COMBINING ACUTE ACCENT}")> will return C<MAYBE>.
221	Though, C<"A\N{COMBINING ACUTE ACCENT}"> is not in NFC
222	(its NFC is C<"\N{LATIN CAPITAL LETTER A WITH ACUTE}">),
223	while C<"B\N{COMBINING ACUTE ACCENT}"> is in NFC.
224
225	If you want to check exactly, compare the string with its NFC/NFKC; i.e.,
226
227	$string eq NFC($string) # more thorough than checkNFC($string)
228	$string eq NFKC($string) # more thorough than checkNFKC($string)
229
2a204b45	230	=head2 Character Data
	231
	232	These functions are interface of character data used internally.
d0ed0342	233	If you want only to get Unicode normalization forms, you don't need
d0ed0342	234	call them yourself.
2a204b45	235
	236	=over 4
	237
	238	=item C<$canonical_decomposed = getCanon($codepoint)>
	239
8f118dcd	240	If the character of the specified codepoint is canonically
	241	decomposable (including Hangul Syllables),
	242	returns the B<completely decomposed> string canonically equivalent to it.
	243
	244	If it is not decomposable, returns undef.
	245
2a204b45	246	=item C<$compatibility_decomposed = getCompat($codepoint)>
2a204b45	247
8f118dcd	248	If the character of the specified codepoint is compatibility
	249	decomposable (including Hangul Syllables),
	250	returns the B<completely decomposed> string compatibility equivalent to it.
2a204b45	251
	252	If it is not decomposable, returns undef.
	253
8f118dcd	254	=item C<$codepoint_composite = getComposite($codepoint_here, $codepoint_next)>
2a204b45	255
d85850a7	256	If two characters here and next (as codepoints) are composable
8f118dcd	257	(including Hangul Jamo/Syllables and Composition Exclusions),
2a204b45	258	returns the codepoint of the composite.
	259
	260	If they are not composable, returns undef.
	261
	262	=item C<$combining_class = getCombinClass($codepoint)>
	263
8f118dcd	264	Returns the combining class of the character as an integer.
2a204b45	265
	266	=item C<$is_exclusion = isExclusion($codepoint)>
	267
8f118dcd	268	Returns a boolean whether the character of the specified codepoint
	269	is a composition exclusion.
	270
	271	=item C<$is_singleton = isSingleton($codepoint)>
	272
2a204b45	273	Returns a boolean whether the character of the specified codepoint is
8f118dcd	274	a singleton.
	275
	276	=item C<$is_non_startar_decomposition = isNonStDecomp($codepoint)>
	277
	278	Returns a boolean whether the canonical decomposition
	279	of the character of the specified codepoint
	280	is a Non-Starter Decomposition.
	281
	282	=item C<$may_be_composed_with_prev_char = isComp2nd($codepoint)>
	283
	284	Returns a boolean whether the character of the specified codepoint
	285	may be composed with the previous one in a certain composition
	286	(including Hangul Compositions, but excluding
	287	Composition Exclusions and Non-Starter Decompositions).
2a204b45	288
	289	=back
	290
	291	=head2 EXPORT
	292
	293	C<NFC>, C<NFD>, C<NFKC>, C<NFKD>: by default.
	294
	295	C<normalize> and other some functions: on request.
	296
	297	=head1 AUTHOR
	298
	299	SADAHIRO Tomoyuki, E<lt>SADAHIRO@cpan.orgE<gt>
	300
	301	http://homepage1.nifty.com/nomenclator/perl/
	302
ab8fe378	303	Copyright(C) 2001-2002, SADAHIRO Tomoyuki. Japan. All rights reserved.
2a204b45	304
	305	This program is free software; you can redistribute it and/or
	306	modify it under the same terms as Perl itself.
	307
	308	=head1 SEE ALSO
	309
	310	=over 4
	311
	312	=item http://www.unicode.org/unicode/reports/tr15/
	313
	314	Unicode Normalization Forms - UAX #15
	315
8f118dcd	316	=item http://www.unicode.org/Public/UNIDATA/DerivedNormalizationProperties.txt
	317
	318	Derived Normalization Properties
	319
2a204b45	320	=back
	321
	322	=cut
	323