1
2BEGIN {
3    unless ('A' eq pack('U', 0x41)) {
4	print "1..0 # Unicode::Collate cannot pack a Unicode code point\n";
5	exit 0;
6    }
7    unless (0x41 == unpack('U', 'A')) {
8	print "1..0 # Unicode::Collate cannot get a Unicode code point\n";
9	exit 0;
10    }
11    if ($ENV{PERL_CORE}) {
12	chdir('t') if -d 't';
13	@INC = $^O eq 'MacOS' ? qw(::lib) : qw(../lib);
14    }
15}
16
17use strict;
18use warnings;
19BEGIN { $| = 1; print "1..41\n"; }
20my $count = 0;
21sub ok ($;$) {
22    my $p = my $r = shift;
23    if (@_) {
24	my $x = shift;
25	$p = !defined $x ? !defined $r : !defined $r ? 0 : $r eq $x;
26    }
27    print $p ? "ok" : "not ok", ' ', ++$count, "\n";
28}
29
30use Unicode::Collate;
31
32ok(1);
33
34#########################
35
36my $trad = Unicode::Collate->new(
37  table => 'keys.txt',
38  normalization => undef,
39  ignoreName => qr/HANGUL|HIRAGANA|KATAKANA|BOPOMOFO/,
40  level => 3,
41  entry => << 'ENTRIES',
42 0063 0068 ; [.0A3F.0020.0002.0063] % "ch" in traditional Spanish
43 0043 0068 ; [.0A3F.0020.0007.0043] # "Ch" in traditional Spanish
44 0043 0048 ; [.0A3F.0020.0008.0043] # "CH" in traditional Spanish
45ENTRIES
46);
47# 0063  ; [.0A3D.0020.0002.0063] # LATIN SMALL LETTER C
48# 0064  ; [.0A49.0020.0002.0064] # LATIN SMALL LETTER D
49
50##### 2..3
51
52ok(
53  join(':', $trad->sort( qw/ acha aca ada acia acka / ) ),
54  join(':',              qw/ aca acia acka acha ada / ),
55);
56
57ok(
58  join(':', $trad->sort( qw/ ACHA ACA ADA ACIA ACKA / ) ),
59  join(':',              qw/ ACA ACIA ACKA ACHA ADA / ),
60);
61
62##### 4..7
63
64ok($trad->gt("ocho", "oc\cAho")); # UCA v14
65ok($trad->gt("ocho", "oc\0\cA\0\cBho"));  # UCA v14
66ok($trad->eq("-", ""));
67ok($trad->gt("ocho", "oc-ho"));
68
69##### 8..11
70
71$trad->change(UCA_Version => 9);
72
73ok($trad->eq("ocho", "oc\cAho")); # UCA v9
74ok($trad->eq("ocho", "oc\0\cA\0\cBho")); # UCA v9
75ok($trad->eq("-", ""));
76ok($trad->gt("ocho", "oc-ho"));
77
78##### 12..15
79
80$trad->change(UCA_Version => 8);
81
82ok($trad->gt("ocho", "oc\cAho"));
83ok($trad->gt("ocho", "oc\0\cA\0\cBho"));
84ok($trad->eq("-", ""));
85ok($trad->gt("ocho", "oc-ho"));
86
87
88##### 16..19
89
90$trad->change(UCA_Version => 9);
91
92my $hiragana = "\x{3042}\x{3044}";
93my $katakana = "\x{30A2}\x{30A4}";
94
95# HIRAGANA and KATAKANA are ignorable via ignoreName
96ok($trad->eq($hiragana, ""));
97ok($trad->eq("", $katakana));
98ok($trad->eq($hiragana, $katakana));
99ok($trad->eq($katakana, $hiragana));
100
101
102##### 20..31
103
104# According to Conformance Test (UCA_Version == 9 or 11),
105# a L3-ignorable is treated as a completely ignorable.
106
107my $L3ignorable = Unicode::Collate->new(
108  alternate => 'Non-ignorable',
109  level => 3,
110  table => undef,
111  normalization => undef,
112  UCA_Version => 9,
113  entry => <<'ENTRIES',
1140000  ; [.0000.0000.0000.0000] # [0000] NULL (in 6429)
1150001  ; [.0000.0000.0000.0000] # [0001] START OF HEADING (in 6429)
1160591  ; [.0000.0000.0000.0591] # HEBREW ACCENT ETNAHTA
1171D165 ; [.0000.0000.0000.1D165] # MUSICAL SYMBOL COMBINING STEM
1180021  ; [*024B.0020.0002.0021] # EXCLAMATION MARK
11909BE  ; [.114E.0020.0002.09BE] # BENGALI VOWEL SIGN AA
12009C7  ; [.1157.0020.0002.09C7] # BENGALI VOWEL SIGN E
12109CB  ; [.1159.0020.0002.09CB] # BENGALI VOWEL SIGN O
12209C7 09BE ; [.1159.0020.0002.09CB] # BENGALI VOWEL SIGN O
1231D1B9 ; [*098A.0020.0002.1D1B9] # MUSICAL SYMBOL SEMIBREVIS WHITE
1241D1BA ; [*098B.0020.0002.1D1BA] # MUSICAL SYMBOL SEMIBREVIS BLACK
1251D1BB ; [*098A.0020.0002.1D1B9][.0000.0000.0000.1D165] # M.S. MINIMA
1261D1BC ; [*098B.0020.0002.1D1BA][.0000.0000.0000.1D165] # M.S. MINIMA BLACK
127ENTRIES
128);
129
130ok($L3ignorable->lt("\cA", "!"));
131ok($L3ignorable->lt("\x{591}", "!"));
132ok($L3ignorable->eq("\cA", "\x{591}"));
133ok($L3ignorable->eq("\x{9C7}\x{9BE}A", "\x{9C7}\cA\x{9BE}A"));
134ok($L3ignorable->eq("\x{9C7}\x{9BE}A", "\x{9C7}\x{591}\x{9BE}A"));
135ok($L3ignorable->eq("\x{9C7}\x{9BE}A", "\x{9C7}\x{1D165}\x{9BE}A"));
136ok($L3ignorable->eq("\x{9C7}\x{9BE}A", "\x{9CB}A"));
137ok($L3ignorable->lt("\x{1D1BB}", "\x{1D1BC}"));
138ok($L3ignorable->eq("\x{1D1BB}", "\x{1D1B9}"));
139ok($L3ignorable->eq("\x{1D1BC}", "\x{1D1BA}"));
140ok($L3ignorable->eq("\x{1D1BB}", "\x{1D1B9}\x{1D165}"));
141ok($L3ignorable->eq("\x{1D1BC}", "\x{1D1BA}\x{1D165}"));
142
143##### 32..41
144
145my $c = Unicode::Collate->new(
146  table => 'keys.txt',
147  normalization => undef,
148  level => 1,
149  UCA_Version => 14,
150  entry => << 'ENTRIES',
151034F  ; [.0000.0000.0000.034F] # COMBINING GRAPHEME JOINER
1520063 0068 ; [.0A3F.0020.0002.0063] % "ch" in traditional Spanish
1530043 0068 ; [.0A3F.0020.0007.0043] # "Ch" in traditional Spanish
1540043 0048 ; [.0A3F.0020.0008.0043] # "CH" in traditional Spanish
155ENTRIES
156);
157# 0063  ; [.0A3D.0020.0002.0063] # LATIN SMALL LETTER C
158# 0064  ; [.0A49.0020.0002.0064] # LATIN SMALL LETTER D
159
160ok($c->gt("ocho", "oc\x00\x00ho"));
161ok($c->gt("ocho", "oc\cAho"));
162ok($c->gt("ocho", "oc\x{34F}ho"));
163ok($c->gt("ocio", "oc\x{34F}ho"));
164ok($c->lt("ocgo", "oc\x{34F}ho"));
165ok($c->lt("oceo", "oc\x{34F}ho"));
166
167ok($c->viewSortKey("ocho"),         "[0B4B 0A3F 0B4B | | |]");
168ok($c->viewSortKey("oc\x00\x00ho"), "[0B4B 0A3D 0AB9 0B4B | | |]");
169ok($c->viewSortKey("oc\cAho"),      "[0B4B 0A3D 0AB9 0B4B | | |]");
170ok($c->viewSortKey("oc\x{34F}ho"),  "[0B4B 0A3D 0AB9 0B4B | | |]");
171
172
173