$ASCII->initialize([ 0..127 ]);
}
- # A number of the Perl synonyms have a restricted-range synonym whose name
- # begins with Posix. This hash gets filled in with them, so that they can
- # be populated in a small loop.
- my %posix_equivalent;
-
# Get the best available case definitions. Early Unicode versions didn't
# have Uppercase and Lowercase defined, so use the general category
# instead for them.
$Lower->set_equivalent_to($gc->table('Lowercase_Letter'),
Related => 1);
}
- $posix_equivalent{'Lower'} = $Lower;
+ $perl->add_match_table("PosixLower",
+ Description => "[a-z]",
+ Initialize => $Lower & $ASCII,
+ );
my $Upper = $perl->add_match_table('Upper');
my $Unicode_Upper = property_ref('Uppercase');
$Upper->set_equivalent_to($gc->table('Uppercase_Letter'),
Related => 1);
}
- $posix_equivalent{'Upper'} = $Upper;
+ $perl->add_match_table("PosixUpper",
+ Description => "[A-Z]",
+ Initialize => $Upper & $ASCII,
+ );
# Earliest releases didn't have title case. Initialize it to empty if not
# otherwise present
# one whose name generally begins with Posix that is posix-compliant, and
# one that matches Unicode characters beyond the Posix, ASCII range
- my $Alpha = $perl->add_match_table('Alpha',
- Description => '[[:Alpha:]] extended beyond ASCII');
+ my $Alpha = $perl->add_match_table('Alpha');
# Alphabetic was not present in early releases
my $Alphabetic = property_ref('Alphabetic');
+ $gc->table('Mn')
+ $gc->table('Mc'));
$Alpha += $gc->table('Nl') if defined $gc->table('Nl');
+ $Alpha->add_description('Alphabetic');
}
- $posix_equivalent{'Alpha'} = $Alpha;
+ $perl->add_match_table("PosixAlpha",
+ Description => "[A-Za-z]",
+ Initialize => $Alpha & $ASCII,
+ );
my $Alnum = $perl->add_match_table('Alnum',
- Description => "[[:Alnum:]] extended beyond ASCII",
+ Description => 'Alphabetic and (Decimal) Numeric',
Initialize => $Alpha + $gc->table('Decimal_Number'),
);
- $posix_equivalent{'Alnum'} = $Alnum;
+ $perl->add_match_table("PosixAlnum",
+ Description => "[A-Za-z0-9]",
+ Initialize => $Alnum & $ASCII,
+ );
my $Word = $perl->add_match_table('Word',
Description => '\w, including beyond ASCII',
- 0x200B, # ZWSP
);
$Blank->add_alias('HorizSpace'); # Another name for it.
- $posix_equivalent{'Blank'} = $Blank;
+ $perl->add_match_table("PosixBlank",
+ Description => "\\t and ' '",
+ Initialize => $Blank & $ASCII,
+ );
my $VertSpace = $perl->add_match_table('VertSpace',
Description => '\v',
# No Posix equivalent for vertical space
my $Space = $perl->add_match_table('Space',
- Description => '\s including beyond ASCII plus vertical tab = [[:Space:]]',
- Initialize => $Blank + $VertSpace,
+ Description => '\s including beyond ASCII plus vertical tab',
+ Initialize => $Blank + $VertSpace,
);
- $posix_equivalent{'Space'} = $Space;
+ $perl->add_match_table("PosixSpace",
+ Description => "\\t \\n, \\x0B, \\f, \\r, and ' '",
+ Initialize => $Space & $ASCII,
+ );
# Perl's traditional space doesn't include Vertical Tab
my $SpacePerl = $perl->add_match_table('SpacePerl',
);
my $Cntrl = $perl->add_match_table('Cntrl',
- Description => "[[:Cntrl:]] extended beyond ASCII");
+ Description => 'Control characters');
$Cntrl->set_equivalent_to($gc->table('Cc'), Related => 1);
- $posix_equivalent{'Cntrl'} = $Cntrl;
+ $perl->add_match_table("PosixCntrl",
+ Description => '[\x00-\x1F]',
+ Initialize => $Cntrl & $ASCII,
+ );
# $controls is a temporary used to construct Graph.
my $controls = Range_List->new(Initialize => $gc->table('Unassigned')
# Graph is ~space & ~(Cc|Cs|Cn) = ~(space + $controls)
my $Graph = $perl->add_match_table('Graph',
- Description => "[[:Graph:]] extended beyond ASCII",
+ Description => 'Characters that are graphical',
Initialize => ~ ($Space + $controls),
);
- $posix_equivalent{'Graph'} = $Graph;
+ $perl->add_match_table("PosixGraph",
+ Description => '[\x21-\x7E]',
+ Initialize => $Graph & $ASCII,
+ );
my $Print = $perl->add_match_table('Print',
- Description => "[[:Print:]] extended beyond ASCII",
+ Description => 'Characters that are graphical plus space characters (but no controls)',
Initialize => $Blank + $Graph - $gc->table('Control'),
);
- $posix_equivalent{'Print'} = $Print;
+ $perl->add_match_table("PosixPrint",
+ Description => '[\x20-\x7E]',
+ Initialize => $Print & $ASCII,
+ );
my $Punct = $perl->add_match_table('Punct');
$Punct->set_equivalent_to($gc->table('Punctuation'), Related => 1);
# \p{punct} doesn't include the symbols, which posix does
$perl->add_match_table('PosixPunct',
- Description => "[[:Punct:]]",
- Initialize => $ASCII & ($gc->table('Punctuation')
- + $gc->table('Symbol')),
- );
+ Description => 'Graphical characters that aren\'t Word characters = [\x21-\x2F\x3A-\x40\x5B-\x60\x7B-\x7E]',
+ Initialize => $ASCII & ($gc->table('Punctuation')
+ + $gc->table('Symbol')),
+ );
my $Digit = $perl->add_match_table('Digit',
Description => '\d, extended beyond just [0-9]');
$Digit->set_equivalent_to($gc->table('Decimal_Number'), Related => 1);
- $posix_equivalent{'Digit'} = $Digit;
+ my $PosixDigit = $perl->add_match_table("PosixDigit",
+ Description => '[0-9]',
+ Initialize => $Digit & $ASCII,
+ );
# AHex was not present in early releases
# XXX TUS recommends Hex_Digit, not ASCII_Hex_Digit.
$Xdigit->set_equivalent_to($AHex->table('Y'), Related => 1);
}
else {
- # (Have to use hex because could be running on an non-ASCII machine,
+ # (Have to use hex because could be running on a non-ASCII machine,
# and we want the Unicode (ASCII) values)
$Xdigit->initialize([ 0x30..0x39, 0x41..0x46, 0x61..0x66 ]);
}
- # Now, add the ASCII-restricted tables that get uniform treatment
- while (my ($name, $table) = each %posix_equivalent) {
- $perl->add_match_table("Posix$name",
- Description => "[[:$name:]]",
- Initialize => $table & $ASCII,
- );
- }
- $perl->table('PosixDigit')->add_description('\d, restricted to ASCII');
- $perl->table('PosixDigit')->add_description('[0-9]');
-
-
my $dt = property_ref('Decomposition_Type');
$dt->add_match_table('Non_Canon', Full_Name => 'Non_Canonical',
Initialize => ~ ($dt->table('None') + $dt->table('Canonical')),
projects / p5sagit/p5-mst-13.2.git / commitdiff