package Unicode::Collate; BEGIN { unless ("A" eq pack('U', 0x41)) { die "Unicode::Collate cannot stringify a Unicode code point\n"; } } use 5.006; use strict; use warnings; use Carp; use File::Spec; no warnings 'utf8'; our $VERSION = '0.89'; our $PACKAGE = __PACKAGE__; ### begin XS only ### require DynaLoader; our @ISA = qw(DynaLoader); bootstrap Unicode::Collate $VERSION; ### end XS only ### my @Path = qw(Unicode Collate); my $KeyFile = "allkeys.txt"; # Perl's boolean use constant TRUE => 1; use constant FALSE => ""; use constant NOMATCHPOS => -1; # A coderef to get combining class imported from Unicode::Normalize # (i.e. \&Unicode::Normalize::getCombinClass). # This is also used as a HAS_UNICODE_NORMALIZE flag. my $CVgetCombinClass; # Supported Levels use constant MinLevel => 1; use constant MaxLevel => 4; # Minimum weights at level 2 and 3, respectively use constant Min2Wt => 0x20; use constant Min3Wt => 0x02; # Shifted weight at 4th level use constant Shift4Wt => 0xFFFF; # A boolean for Variable and 16-bit weights at 4 levels of Collation Element # PROBLEM: The Default Unicode Collation Element Table # has weights over 0xFFFF at the 4th level. # The tie-breaking in the variable weights # other than "shift" (as well as "shift-trimmed") is unreliable. use constant VCE_TEMPLATE => 'Cn4'; # A sort key: 16-bit weights # See also the PROBLEM on VCE_TEMPLATE above. use constant KEY_TEMPLATE => 'n*'; # Level separator in a sort key: # i.e. pack(KEY_TEMPLATE, 0) use constant LEVEL_SEP => "\0\0"; # As Unicode code point separator for hash keys. # A joined code point string (denoted by JCPS below) # like "65;768" is used for internal processing # instead of Perl's Unicode string like "\x41\x{300}", # as the native code point is different from the Unicode code point # on EBCDIC platform. # This character must not be included in any stringified # representation of an integer. use constant CODE_SEP => ';'; # NOTE: in regex /;/ is used for $jcps! # boolean values of variable weights use constant NON_VAR => 0; # Non-Variable character use constant VAR => 1; # Variable character # specific code points use constant Hangul_SIni => 0xAC00; use constant Hangul_SFin => 0xD7A3; # Logical_Order_Exception in PropList.txt my $DefaultRearrange = [ 0x0E40..0x0E44, 0x0EC0..0x0EC4 ]; sub UCA_Version { "24" } sub Base_Unicode_Version { "6.1.0" } ###### sub pack_U { return pack('U*', @_); } ###### my (%VariableOK); @VariableOK{ qw/ blanked non-ignorable shifted shift-trimmed / } = (); # keys lowercased our @ChangeOK = qw/ alternate backwards level normalization rearrange katakana_before_hiragana upper_before_lower ignore_level2 overrideHangul overrideCJK preprocess UCA_Version hangul_terminator variable /; our @ChangeNG = qw/ entry mapping table maxlength contraction ignoreChar ignoreName undefChar undefName rewrite versionTable alternateTable backwardsTable forwardsTable rearrangeTable variableTable derivCode normCode rearrangeHash backwardsFlag suppress suppressHash __useXS /; ### XS only # The hash key 'ignored' is deleted at v 0.21. # The hash key 'isShift' is deleted at v 0.23. # The hash key 'combining' is deleted at v 0.24. # The hash key 'entries' is deleted at v 0.30. # The hash key 'L3_ignorable' is deleted at v 0.40. sub version { my $self = shift; return $self->{versionTable} || 'unknown'; } my (%ChangeOK, %ChangeNG); @ChangeOK{ @ChangeOK } = (); @ChangeNG{ @ChangeNG } = (); sub change { my $self = shift; my %hash = @_; my %old; if (exists $hash{variable} && exists $hash{alternate}) { delete $hash{alternate}; } elsif (!exists $hash{variable} && exists $hash{alternate}) { $hash{variable} = $hash{alternate}; } foreach my $k (keys %hash) { if (exists $ChangeOK{$k}) { $old{$k} = $self->{$k}; $self->{$k} = $hash{$k}; } elsif (exists $ChangeNG{$k}) { croak "change of $k via change() is not allowed!"; } # else => ignored } $self->checkCollator(); return wantarray ? %old : $self; } sub _checkLevel { my $level = shift; my $key = shift; # 'level' or 'backwards' MinLevel <= $level or croak sprintf "Illegal level %d (in value for key '%s') lower than %d.", $level, $key, MinLevel; $level <= MaxLevel or croak sprintf "Unsupported level %d (in value for key '%s') higher than %d.", $level, $key, MaxLevel; } my %DerivCode = ( 8 => \&_derivCE_8, 9 => \&_derivCE_9, 11 => \&_derivCE_9, # 11 == 9 14 => \&_derivCE_14, 16 => \&_derivCE_14, # 16 == 14 18 => \&_derivCE_18, 20 => \&_derivCE_20, 22 => \&_derivCE_22, 24 => \&_derivCE_24, ); sub checkCollator { my $self = shift; _checkLevel($self->{level}, "level"); $self->{derivCode} = $DerivCode{ $self->{UCA_Version} } or croak "Illegal UCA version (passed $self->{UCA_Version})."; $self->{variable} ||= $self->{alternate} || $self->{variableTable} || $self->{alternateTable} || 'shifted'; $self->{variable} = $self->{alternate} = lc($self->{variable}); exists $VariableOK{ $self->{variable} } or croak "$PACKAGE unknown variable parameter name: $self->{variable}"; if (! defined $self->{backwards}) { $self->{backwardsFlag} = 0; } elsif (! ref $self->{backwards}) { _checkLevel($self->{backwards}, "backwards"); $self->{backwardsFlag} = 1 << $self->{backwards}; } else { my %level; $self->{backwardsFlag} = 0; for my $b (@{ $self->{backwards} }) { _checkLevel($b, "backwards"); $level{$b} = 1; } for my $v (sort keys %level) { $self->{backwardsFlag} += 1 << $v; } } defined $self->{rearrange} or $self->{rearrange} = []; ref $self->{rearrange} or croak "$PACKAGE: list for rearrangement must be store in ARRAYREF"; # keys of $self->{rearrangeHash} are $self->{rearrange}. $self->{rearrangeHash} = undef; if (@{ $self->{rearrange} }) { @{ $self->{rearrangeHash} }{ @{ $self->{rearrange} } } = (); } $self->{normCode} = undef; if (defined $self->{normalization}) { eval { require Unicode::Normalize }; $@ and croak "Unicode::Normalize is required to normalize strings"; $CVgetCombinClass ||= \&Unicode::Normalize::getCombinClass; if ($self->{normalization} =~ /^(?:NF)D\z/) { # tweak for default $self->{normCode} = \&Unicode::Normalize::NFD; } elsif ($self->{normalization} ne 'prenormalized') { my $norm = $self->{normalization}; $self->{normCode} = sub { Unicode::Normalize::normalize($norm, shift); }; eval { $self->{normCode}->("") }; # try $@ and croak "$PACKAGE unknown normalization form name: $norm"; } } return; } sub new { my $class = shift; my $self = bless { @_ }, $class; ### begin XS only ### if (! exists $self->{table} && !defined $self->{rewrite} && !defined $self->{undefName} && !defined $self->{ignoreName} && !defined $self->{undefChar} && !defined $self->{ignoreChar}) { $self->{__useXS} = \&_fetch_simple; } else { $self->{__useXS} = undef; } ### end XS only ### # keys of $self->{suppressHash} are $self->{suppress}. if ($self->{suppress} && @{ $self->{suppress} }) { @{ $self->{suppressHash} }{ @{ $self->{suppress} } } = (); } # before read_table() # If undef is passed explicitly, no file is read. $self->{table} = $KeyFile if ! exists $self->{table}; $self->read_table() if defined $self->{table}; if ($self->{entry}) { while ($self->{entry} =~ /([^\n]+)/g) { $self->parseEntry($1, TRUE); } } $self->{level} ||= MaxLevel; $self->{UCA_Version} ||= UCA_Version(); $self->{overrideHangul} = FALSE if ! exists $self->{overrideHangul}; $self->{overrideCJK} = FALSE if ! exists $self->{overrideCJK}; $self->{normalization} = 'NFD' if ! exists $self->{normalization}; $self->{rearrange} = $self->{rearrangeTable} || ($self->{UCA_Version} <= 11 ? $DefaultRearrange : []) if ! exists $self->{rearrange}; $self->{backwards} = $self->{backwardsTable} if ! exists $self->{backwards}; $self->checkCollator(); return $self; } sub parseAtmark { my $self = shift; my $line = shift; # after s/^\s*\@// if ($line =~ /^version\s*(\S*)/) { $self->{versionTable} ||= $1; } elsif ($line =~ /^variable\s+(\S*)/) { # since UTS #10-9 $self->{variableTable} ||= $1; } elsif ($line =~ /^alternate\s+(\S*)/) { # till UTS #10-8 $self->{alternateTable} ||= $1; } elsif ($line =~ /^backwards\s+(\S*)/) { push @{ $self->{backwardsTable} }, $1; } elsif ($line =~ /^forwards\s+(\S*)/) { # parhaps no use push @{ $self->{forwardsTable} }, $1; } elsif ($line =~ /^rearrange\s+(.*)/) { # (\S*) is NG push @{ $self->{rearrangeTable} }, _getHexArray($1); } } sub read_table { my $self = shift; ### begin XS only ### if ($self->{__useXS}) { my @rest = _fetch_rest(); # complex matter need to parse for my $line (@rest) { next if $line =~ /^\s*#/; if ($line =~ s/^\s*\@//) { $self->parseAtmark($line); } else { $self->parseEntry($line); } } return; } ### end XS only ### my($f, $fh); foreach my $d (@INC) { $f = File::Spec->catfile($d, @Path, $self->{table}); last if open($fh, $f); $f = undef; } if (!defined $f) { $f = File::Spec->catfile(@Path, $self->{table}); croak("$PACKAGE: Can't locate $f in \@INC (\@INC contains: @INC)"); } while (my $line = <$fh>) { next if $line =~ /^\s*#/; if ($line =~ s/^\s*\@//) { $self->parseAtmark($line); } else { $self->parseEntry($line); } } close $fh; } ## ## get $line, parse it, and write an entry in $self ## sub parseEntry { my $self = shift; my $line = shift; my $tailoring = shift; my($name, $entry, @uv, @key); if (defined $self->{rewrite}) { $line = $self->{rewrite}->($line); } return if $line !~ /^\s*[0-9A-Fa-f]/; # removes comment and gets name $name = $1 if $line =~ s/[#%]\s*(.*)//; return if defined $self->{undefName} && $name =~ /$self->{undefName}/; # gets element my($e, $k) = split /;/, $line; croak "Wrong Entry: must be separated by ';' from " if ! $k; @uv = _getHexArray($e); return if !@uv; return if @uv > 1 && $self->{suppressHash} && !$tailoring && exists $self->{suppressHash}{$uv[0]}; $entry = join(CODE_SEP, @uv); # in JCPS if (defined $self->{undefChar} || defined $self->{ignoreChar}) { my $ele = pack_U(@uv); # regarded as if it were not entried in the table return if defined $self->{undefChar} && $ele =~ /$self->{undefChar}/; # replaced as completely ignorable $k = '[.0000.0000.0000.0000]' if defined $self->{ignoreChar} && $ele =~ /$self->{ignoreChar}/; } # replaced as completely ignorable $k = '[.0000.0000.0000.0000]' if defined $self->{ignoreName} && $name =~ /$self->{ignoreName}/; my $is_L3_ignorable = TRUE; foreach my $arr ($k =~ /\[([^\[\]]+)\]/g) { # SPACEs allowed my $var = $arr =~ /\*/; # exactly /^\*/ but be lenient. my @wt = _getHexArray($arr); push @key, pack(VCE_TEMPLATE, $var, @wt); $is_L3_ignorable = FALSE if $wt[0] || $wt[1] || $wt[2]; # Conformance Test for 3.1.1 and 4.0.0 shows Level 3 ignorable # is completely ignorable. # For expansion, an entry $is_L3_ignorable # if and only if "all" CEs are [.0000.0000.0000]. } $self->{mapping}{$entry} = $is_L3_ignorable ? [] : \@key; if (@uv > 1) { if (!$self->{maxlength}{$uv[0]} || $self->{maxlength}{$uv[0]} < @uv) { $self->{maxlength}{$uv[0]} = @uv; } } if (@uv > 2) { while (@uv) { pop @uv; my $fake_entry = join(CODE_SEP, @uv); # in JCPS $self->{contraction}{$fake_entry} = 1; } } } sub viewSortKey { my $self = shift; $self->visualizeSortKey($self->getSortKey(@_)); } ## ## arrayref of JCPS = splitEnt(string to be collated) ## arrayref of arrayref[JCPS, ini_pos, fin_pos] = splitEnt(string, true) ## sub splitEnt { my $self = shift; my $wLen = $_[1]; my $code = $self->{preprocess}; my $norm = $self->{normCode}; my $map = $self->{mapping}; my $max = $self->{maxlength}; my $reH = $self->{rearrangeHash}; my $vers = $self->{UCA_Version}; my $ver9 = $vers >= 9 && $vers <= 11; my $uXS = $self->{__useXS}; ### XS only my ($str, @buf); if ($wLen) { $code and croak "Preprocess breaks character positions. " . "Don't use with index(), match(), etc."; $norm and croak "Normalization breaks character positions. " . "Don't use with index(), match(), etc."; $str = $_[0]; } else { $str = $_[0]; $str = &$code($str) if ref $code; $str = &$norm($str) if ref $norm; } # get array of Unicode code point of string. my @src = unpack_U($str); # rearrangement: # Character positions are not kept if rearranged, # then neglected if $wLen is true. if ($reH && ! $wLen) { for (my $i = 0; $i < @src; $i++) { if (exists $reH->{ $src[$i] } && $i + 1 < @src) { ($src[$i], $src[$i+1]) = ($src[$i+1], $src[$i]); $i++; } } } # remove a code point marked as a completely ignorable. for (my $i = 0; $i < @src; $i++) { if (_isIllegal($src[$i]) || $vers <= 20 && _isNonchar($src[$i])) { $src[$i] = undef; } elsif ($ver9) { $src[$i] = undef if $map->{ $src[$i] } && @{ $map->{ $src[$i] } } == 0; ### begin XS only ### if ($uXS) { $src[$i] = undef if _ignorable_simple($src[$i]); } ### end XS only ### } } for (my $i = 0; $i < @src; $i++) { my $jcps = $src[$i]; # skip removed code point if (! defined $jcps) { if ($wLen && @buf) { $buf[-1][2] = $i + 1; } next; } my $i_orig = $i; # find contraction if ($max->{$jcps}) { my $temp_jcps = $jcps; my $jcpsLen = 1; my $maxLen = $max->{$jcps}; for (my $p = $i + 1; $jcpsLen < $maxLen && $p < @src; $p++) { next if ! defined $src[$p]; $temp_jcps .= CODE_SEP . $src[$p]; $jcpsLen++; if ($map->{$temp_jcps}) { $jcps = $temp_jcps; $i = $p; } } # discontiguous contraction with Combining Char (cf. UTS#10, S2.1). # This process requires Unicode::Normalize. # If "normalization" is undef, here should be skipped *always* # (in spite of bool value of $CVgetCombinClass), # since canonical ordering cannot be expected. # Blocked combining character should not be contracted. # $self->{normCode} is false in the case of "prenormalized". if ($self->{normalization}) { my $cont = $self->{contraction}; my $preCC = 0; my $preCC_uc = 0; my $jcps_uc = $jcps; my(@out, @out_uc); for (my $p = $i + 1; $p < @src; $p++) { next if ! defined $src[$p]; my $curCC = $CVgetCombinClass->($src[$p]); last unless $curCC; my $tail = CODE_SEP . $src[$p]; if ($preCC_uc != $curCC && ($map->{$jcps_uc.$tail} || $cont->{$jcps_uc.$tail})) { $jcps_uc .= $tail; push @out_uc, $p; } else { $preCC_uc = $curCC; } if ($preCC != $curCC && $map->{$jcps.$tail}) { $jcps .= $tail; push @out, $p; } else { $preCC = $curCC; } } if ($map->{$jcps_uc}) { $jcps = $jcps_uc; $src[$_] = undef for @out_uc; } else { $src[$_] = undef for @out; } } } # skip completely ignorable if ($uXS && $jcps !~ /;/ && _ignorable_simple($jcps) || ### XS only $map->{$jcps} && @{ $map->{$jcps} } == 0) { if ($wLen && @buf) { $buf[-1][2] = $i + 1; } next; } push @buf, $wLen ? [$jcps, $i_orig, $i + 1] : $jcps; } return \@buf; } ## ## VCE = _pack_override(input, codepoint, derivCode) ## sub _pack_override ($$$) { my $r = shift; my $u = shift; my $der = shift; if (ref $r) { return pack(VCE_TEMPLATE, NON_VAR, @$r); } elsif (defined $r) { return pack(VCE_TEMPLATE, NON_VAR, $r, Min2Wt, Min3Wt, $u); } else { return $der->($u); } } ## ## list of VCE = getWt(JCPS) ## sub getWt { my $self = shift; my $u = shift; my $map = $self->{mapping}; my $der = $self->{derivCode}; my $uXS = $self->{__useXS}; ### XS only return if !defined $u; return map($self->varCE($_), @{ $map->{$u} }) if $map->{$u}; ### begin XS only ### return map($self->varCE($_), _fetch_simple($u)) if $uXS && _exists_simple($u); ### end XS only ### # JCPS must not be a contraction, then it's a code point. if (Hangul_SIni <= $u && $u <= Hangul_SFin) { my $hang = $self->{overrideHangul}; my @hangulCE; if ($hang) { @hangulCE = map _pack_override($_, $u, $der), $hang->($u); } elsif (!defined $hang) { @hangulCE = $der->($u); } else { my $max = $self->{maxlength}; my @decH = _decompHangul($u); if (@decH == 2) { my $contract = join(CODE_SEP, @decH); @decH = ($contract) if $map->{$contract}; } else { # must be <@decH == 3> if ($max->{$decH[0]}) { my $contract = join(CODE_SEP, @decH); if ($map->{$contract}) { @decH = ($contract); } else { $contract = join(CODE_SEP, @decH[0,1]); $map->{$contract} and @decH = ($contract, $decH[2]); } # even if V's ignorable, LT contraction is not supported. # If such a situation were required, NFD should be used. } if (@decH == 3 && $max->{$decH[1]}) { my $contract = join(CODE_SEP, @decH[1,2]); $map->{$contract} and @decH = ($decH[0], $contract); } } @hangulCE = map({ $map->{$_} ? @{ $map->{$_} } : $uXS && _exists_simple($_) ? _fetch_simple($_) : ### XS only $der->($_); } @decH); } return map $self->varCE($_), @hangulCE; } else { my $cjk = $self->{overrideCJK}; my $vers = $self->{UCA_Version}; if ($cjk && _isUIdeo($u, $vers)) { my @cjkCE = map _pack_override($_, $u, $der), $cjk->($u); return map $self->varCE($_), @cjkCE; } if ($vers == 8 && defined $cjk && _isUIdeo($u, 0)) { return map $self->varCE($_), _uideoCE_8($u); } return map $self->varCE($_), $der->($u); } } ## ## string sortkey = getSortKey(string arg) ## sub getSortKey { my $self = shift; my $rEnt = $self->splitEnt(shift); # get an arrayref of JCPS my $vers = $self->{UCA_Version}; my $term = $self->{hangul_terminator}; my @buf; # weight arrays if ($term) { my $preHST = ''; my $termCE = $self->varCE(pack(VCE_TEMPLATE, NON_VAR, $term, 0,0,0)); foreach my $jcps (@$rEnt) { # weird things like VL, TL-contraction are not considered! my $curHST = join '', map getHST($_, $vers), split /;/, $jcps; if ($preHST && !$curHST || # hangul before non-hangul $preHST =~ /L\z/ && $curHST =~ /^T/ || $preHST =~ /V\z/ && $curHST =~ /^L/ || $preHST =~ /T\z/ && $curHST =~ /^[LV]/) { push @buf, $termCE; } $preHST = $curHST; push @buf, $self->getWt($jcps); } push @buf, $termCE if $preHST; # end at hangul } else { foreach my $jcps (@$rEnt) { push @buf, $self->getWt($jcps); } } return $self->mk_SortKey(\@buf); ### XS only } ## ## int compare = cmp(string a, string b) ## sub cmp { $_[0]->getSortKey($_[1]) cmp $_[0]->getSortKey($_[2]) } sub eq { $_[0]->getSortKey($_[1]) eq $_[0]->getSortKey($_[2]) } sub ne { $_[0]->getSortKey($_[1]) ne $_[0]->getSortKey($_[2]) } sub lt { $_[0]->getSortKey($_[1]) lt $_[0]->getSortKey($_[2]) } sub le { $_[0]->getSortKey($_[1]) le $_[0]->getSortKey($_[2]) } sub gt { $_[0]->getSortKey($_[1]) gt $_[0]->getSortKey($_[2]) } sub ge { $_[0]->getSortKey($_[1]) ge $_[0]->getSortKey($_[2]) } ## ## list[strings] sorted = sort(list[strings] arg) ## sub sort { my $obj = shift; return map { $_->[1] } sort{ $a->[0] cmp $b->[0] } map [ $obj->getSortKey($_), $_ ], @_; } ## ## bool _nonIgnorAtLevel(arrayref weights, int level) ## sub _nonIgnorAtLevel($$) { my $wt = shift; return if ! defined $wt; my $lv = shift; return grep($wt->[$_-1] != 0, MinLevel..$lv) ? TRUE : FALSE; } ## ## bool _eqArray( ## arrayref of arrayref[weights] source, ## arrayref of arrayref[weights] substr, ## int level) ## * comparison of graphemes vs graphemes. ## @$source >= @$substr must be true (check it before call this); ## sub _eqArray($$$) { my $source = shift; my $substr = shift; my $lev = shift; for my $g (0..@$substr-1){ # Do the $g'th graphemes have the same number of AV weights? return if @{ $source->[$g] } != @{ $substr->[$g] }; for my $w (0..@{ $substr->[$g] }-1) { for my $v (0..$lev-1) { return if $source->[$g][$w][$v] != $substr->[$g][$w][$v]; } } } return 1; } ## ## (int position, int length) ## int position = index(string, substring, position, [undoc'ed global]) ## ## With "global" (only for the list context), ## returns list of arrayref[position, length]. ## sub index { my $self = shift; my $str = shift; my $len = length($str); my $subE = $self->splitEnt(shift); my $pos = @_ ? shift : 0; $pos = 0 if $pos < 0; my $glob = shift; my $lev = $self->{level}; my $v2i = $self->{UCA_Version} >= 9 && $self->{variable} ne 'non-ignorable'; if (! @$subE) { my $temp = $pos <= 0 ? 0 : $len <= $pos ? $len : $pos; return $glob ? map([$_, 0], $temp..$len) : wantarray ? ($temp,0) : $temp; } $len < $pos and return wantarray ? () : NOMATCHPOS; my $strE = $self->splitEnt($pos ? substr($str, $pos) : $str, TRUE); @$strE or return wantarray ? () : NOMATCHPOS; my(@strWt, @iniPos, @finPos, @subWt, @g_ret); my $last_is_variable; for my $vwt (map $self->getWt($_), @$subE) { my($var, @wt) = unpack(VCE_TEMPLATE, $vwt); my $to_be_pushed = _nonIgnorAtLevel(\@wt,$lev); # "Ignorable (L1, L2) after Variable" since track. v. 9 if ($v2i) { if ($var) { $last_is_variable = TRUE; } elsif (!$wt[0]) { # ignorable $to_be_pushed = FALSE if $last_is_variable; } else { $last_is_variable = FALSE; } } if (@subWt && !$var && !$wt[0]) { push @{ $subWt[-1] }, \@wt if $to_be_pushed; } elsif ($to_be_pushed) { push @subWt, [ \@wt ]; } # else ===> skipped } my $count = 0; my $end = @$strE - 1; $last_is_variable = FALSE; # reuse for (my $i = 0; $i <= $end; ) { # no $i++ my $found_base = 0; # fetch a grapheme while ($i <= $end && $found_base == 0) { for my $vwt ($self->getWt($strE->[$i][0])) { my($var, @wt) = unpack(VCE_TEMPLATE, $vwt); my $to_be_pushed = _nonIgnorAtLevel(\@wt,$lev); # "Ignorable (L1, L2) after Variable" since track. v. 9 if ($v2i) { if ($var) { $last_is_variable = TRUE; } elsif (!$wt[0]) { # ignorable $to_be_pushed = FALSE if $last_is_variable; } else { $last_is_variable = FALSE; } } if (@strWt && !$var && !$wt[0]) { push @{ $strWt[-1] }, \@wt if $to_be_pushed; $finPos[-1] = $strE->[$i][2]; } elsif ($to_be_pushed) { push @strWt, [ \@wt ]; push @iniPos, $found_base ? NOMATCHPOS : $strE->[$i][1]; $finPos[-1] = NOMATCHPOS if $found_base; push @finPos, $strE->[$i][2]; $found_base++; } # else ===> no-op } $i++; } # try to match while ( @strWt > @subWt || (@strWt == @subWt && $i > $end) ) { if ($iniPos[0] != NOMATCHPOS && $finPos[$#subWt] != NOMATCHPOS && _eqArray(\@strWt, \@subWt, $lev)) { my $temp = $iniPos[0] + $pos; if ($glob) { push @g_ret, [$temp, $finPos[$#subWt] - $iniPos[0]]; splice @strWt, 0, $#subWt; splice @iniPos, 0, $#subWt; splice @finPos, 0, $#subWt; } else { return wantarray ? ($temp, $finPos[$#subWt] - $iniPos[0]) : $temp; } } shift @strWt; shift @iniPos; shift @finPos; } } return $glob ? @g_ret : wantarray ? () : NOMATCHPOS; } ## ## scalarref to matching part = match(string, substring) ## sub match { my $self = shift; if (my($pos,$len) = $self->index($_[0], $_[1])) { my $temp = substr($_[0], $pos, $len); return wantarray ? $temp : \$temp; # An lvalue ref \substr should be avoided, # since its value is affected by modification of its referent. } else { return; } } ## ## arrayref matching parts = gmatch(string, substring) ## sub gmatch { my $self = shift; my $str = shift; my $sub = shift; return map substr($str, $_->[0], $_->[1]), $self->index($str, $sub, 0, 'g'); } ## ## bool subst'ed = subst(string, substring, replace) ## sub subst { my $self = shift; my $code = ref $_[2] eq 'CODE' ? $_[2] : FALSE; if (my($pos,$len) = $self->index($_[0], $_[1])) { if ($code) { my $mat = substr($_[0], $pos, $len); substr($_[0], $pos, $len, $code->($mat)); } else { substr($_[0], $pos, $len, $_[2]); } return TRUE; } else { return FALSE; } } ## ## int count = gsubst(string, substring, replace) ## sub gsubst { my $self = shift; my $code = ref $_[2] eq 'CODE' ? $_[2] : FALSE; my $cnt = 0; # Replacement is carried out from the end, then use reverse. for my $pos_len (reverse $self->index($_[0], $_[1], 0, 'g')) { if ($code) { my $mat = substr($_[0], $pos_len->[0], $pos_len->[1]); substr($_[0], $pos_len->[0], $pos_len->[1], $code->($mat)); } else { substr($_[0], $pos_len->[0], $pos_len->[1], $_[2]); } $cnt++; } return $cnt; } 1; __END__ =head1 NAME Unicode::Collate - Unicode Collation Algorithm =head1 SYNOPSIS use Unicode::Collate; #construct $Collator = Unicode::Collate->new(%tailoring); #sort @sorted = $Collator->sort(@not_sorted); #compare $result = $Collator->cmp($a, $b); # returns 1, 0, or -1. B Strings in C<@not_sorted>, C<$a> and C<$b> are interpreted according to Perl's Unicode support. See L, L, L, L, L. Otherwise you can use C or should decode them before. =head1 DESCRIPTION This module is an implementation of Unicode Technical Standard #10 (a.k.a. UTS #10) - Unicode Collation Algorithm (a.k.a. UCA). =head2 Constructor and Tailoring The C method returns a collator object. If new() is called with no parameters, the collator should do the default collation. $Collator = Unicode::Collate->new( UCA_Version => $UCA_Version, alternate => $alternate, # alias for 'variable' backwards => $levelNumber, # or \@levelNumbers entry => $element, hangul_terminator => $term_primary_weight, ignoreName => qr/$ignoreName/, ignoreChar => qr/$ignoreChar/, ignore_level2 => $bool, katakana_before_hiragana => $bool, level => $collationLevel, normalization => $normalization_form, overrideCJK => \&overrideCJK, overrideHangul => \&overrideHangul, preprocess => \&preprocess, rearrange => \@charList, rewrite => \&rewrite, suppress => \@charList, table => $filename, undefName => qr/$undefName/, undefChar => qr/$undefChar/, upper_before_lower => $bool, variable => $variable, ); =over 4 =item UCA_Version If the revision (previously "tracking version") number of UCA is given, behavior of that revision is emulated on collating. If omitted, the return value of C is used. The following revisions are supported. The default is 24. UCA Unicode Standard DUCET (@version) ------------------------------------------------------- 8 3.1 3.0.1 (3.0.1d9) 9 3.1 with Corrigendum 3 3.1.1 (3.1.1) 11 4.0 4.0.0 (4.0.0) 14 4.1.0 4.1.0 (4.1.0) 16 5.0 5.0.0 (5.0.0) 18 5.1.0 5.1.0 (5.1.0) 20 5.2.0 5.2.0 (5.2.0) 22 6.0.0 6.0.0 (6.0.0) 24 6.1.0 6.1.0 (6.1.0) * Noncharacters (e.g. U+FFFF) are not ignored, and can be overridden since C 22. * Fully ignorable characters were ignored, and would not interrupt contractions with C 9 and 11. * Treatment of ignorables after variables and some behaviors were changed at C 9. * Characters regarded as CJK unified ideographs (cf. C) depend on C. * Many hangul jamo are assigned at C 20, that will affect C. =item alternate -- see 3.2.2 Alternate Weighting, version 8 of UTS #10 For backward compatibility, C (old name) can be used as an alias for C. =item backwards -- see 3.1.2 French Accents, UTS #10. backwards => $levelNumber or \@levelNumbers Weights in reverse order; ex. level 2 (diacritic ordering) in French. If omitted (or C<$levelNumber> is C or C<\@levelNumbers> is C<[]>), forwards at all the levels. =item entry -- see 3.1 Linguistic Features; 3.2.1 File Format, UTS #10. If the same character (or a sequence of characters) exists in the collation element table through C, mapping to collation elements is overridden. If it does not exist, the mapping is defined additionally. entry => <<'ENTRY', # for DUCET v4.0.0 (allkeys-4.0.0.txt) 0063 0068 ; [.0E6A.0020.0002.0063] # ch 0043 0068 ; [.0E6A.0020.0007.0043] # Ch 0043 0048 ; [.0E6A.0020.0008.0043] # CH 006C 006C ; [.0F4C.0020.0002.006C] # ll 004C 006C ; [.0F4C.0020.0007.004C] # Ll 004C 004C ; [.0F4C.0020.0008.004C] # LL 00F1 ; [.0F7B.0020.0002.00F1] # n-tilde 006E 0303 ; [.0F7B.0020.0002.00F1] # n-tilde 00D1 ; [.0F7B.0020.0008.00D1] # N-tilde 004E 0303 ; [.0F7B.0020.0008.00D1] # N-tilde ENTRY entry => <<'ENTRY', # for DUCET v4.0.0 (allkeys-4.0.0.txt) 00E6 ; [.0E33.0020.0002.00E6][.0E8B.0020.0002.00E6] # ae ligature as 00C6 ; [.0E33.0020.0008.00C6][.0E8B.0020.0008.00C6] # AE ligature as ENTRY B The code point in the UCA file format (before C<';'>) B be a Unicode code point (defined as hexadecimal), but not a native code point. So C<0063> must always denote C, but not a character of C<"\x63">. Weighting may vary depending on collation element table. So ensure the weights defined in C will be consistent with those in the collation element table loaded via C
. In DUCET v4.0.0, primary weight of C is C<0E60> and that of C is C<0E6D>. So setting primary weight of C to C<0E6A> (as a value between C<0E60> and C<0E6D>) makes ordering as C CH E D>. Exactly speaking DUCET already has some characters between C and C: C (C) with primary weight C<0E64>, C (C) with C<0E65>, and C (C) with C<0E69>. Then primary weight C<0E6A> for C makes C ordered between C and C. =item hangul_terminator -- see 7.1.4 Trailing Weights, UTS #10. If a true value is given (non-zero but should be positive), it will be added as a terminator primary weight to the end of every standard Hangul syllable. Secondary and any higher weights for terminator are set to zero. If the value is false or C key does not exist, insertion of terminator weights will not be performed. Boundaries of Hangul syllables are determined according to conjoining Jamo behavior in F and F. B (1) For expansion mapping (Unicode character mapped to a sequence of collation elements), a terminator will not be added between collation elements, even if Hangul syllable boundary exists there. Addition of terminator is restricted to the next position to the last collation element. (2) Non-conjoining Hangul letters (Compatibility Jamo, halfwidth Jamo, and enclosed letters) are not automatically terminated with a terminator primary weight. These characters may need terminator included in a collation element table beforehand. =item ignoreChar =item ignoreName -- see 3.2.2 Variable Weighting, UTS #10. Makes the entry in the table completely ignorable; i.e. as if the weights were zero at all level. Through C, any character matching C will be ignored. Through C, any character whose name (given in the C
file as a comment) matches C will be ignored. E.g. when 'a' and 'e' are ignorable, 'element' is equal to 'lament' (or 'lmnt'). =item ignore_level2 -- see 5.1 Parametric Tailoring, UTS #10. By default, case-sensitive comparison (that is level 3 difference) won't ignore accents (that is level 2 difference). If the parameter is made true, accents (and other primary ignorable characters) are ignored, even though cases are taken into account. B: C should be 3 or greater. =item katakana_before_hiragana -- see 7.3.1 Tertiary Weight Table, UTS #10. By default, hiragana is before katakana. If the parameter is made true, this is reversed. B: This parameter simplemindedly assumes that any hiragana/katakana distinctions must occur in level 3, and their weights at level 3 must be same as those mentioned in 7.3.1, UTS #10. If you define your collation elements which violate this requirement, this parameter does not work validly. =item level -- see 4.3 Form Sort Key, UTS #10. Set the maximum level. Any higher levels than the specified one are ignored. Level 1: alphabetic ordering Level 2: diacritic ordering Level 3: case ordering Level 4: tie-breaking (e.g. in the case when variable is 'shifted') ex.level => 2, If omitted, the maximum is the 4th. =item normalization -- see 4.1 Normalize, UTS #10. If specified, strings are normalized before preparation of sort keys (the normalization is executed after preprocess). A form name C accepts will be applied as C<$normalization_form>. Acceptable names include C<'NFD'>, C<'NFC'>, C<'NFKD'>, and C<'NFKC'>. See C for detail. If omitted, C<'NFD'> is used. C is performed after C (if defined). Furthermore, special values, C and C<"prenormalized">, can be used, though they are not concerned with C. If C (not a string C<"undef">) is passed explicitly as the value for this key, any normalization is not carried out (this may make tailoring easier if any normalization is not desired). Under C<(normalization =E undef)>, only contiguous contractions are resolved; e.g. even if C (and C) is ordered after C, C would be primary equal to C. In this point, C<(normalization =E undef, preprocess =E sub { NFD(shift) })> B equivalent to C<(normalization =E 'NFD')>. In the case of C<(normalization =E "prenormalized")>, any normalization is not performed, but discontiguous contractions with combining characters are performed. Therefore C<(normalization =E 'prenormalized', preprocess =E sub { NFD(shift) })> B equivalent to C<(normalization =E 'NFD')>. If source strings are finely prenormalized, C<(normalization =E 'prenormalized')> may save time for normalization. Except C<(normalization =E undef)>, B is required (see also B). =item overrideCJK -- see 7.1 Derived Collation Elements, UTS #10. By default, CJK unified ideographs are ordered in Unicode codepoint order, but those in the CJK Unified Ideographs block are lesser than those in the CJK Unified Ideographs Extension A etc. In the CJK Unified Ideographs block: U+4E00..U+9FA5 if UCA_Version is 8, 9 or 11. U+4E00..U+9FBB if UCA_Version is 14 or 16. U+4E00..U+9FC3 if UCA_Version is 18. U+4E00..U+9FCB if UCA_Version is 20 or 22. U+4E00..U+9FCC if UCA_Version is 24. In the CJK Unified Ideographs Extension blocks: Ext.A (U+3400..U+4DB5) and Ext.B (U+20000..U+2A6D6) in any UCA_Version. Ext.C (U+2A700..U+2B734) if UCA_Version is 20 or greater. Ext.D (U+2B740..U+2B81D) if UCA_Version is 22 or greater. Through C, ordering of CJK unified ideographs (including extensions) can be overridden. ex. CJK unified ideographs in the JIS code point order. overrideCJK => sub { my $u = shift; # get a Unicode codepoint my $b = pack('n', $u); # to UTF-16BE my $s = your_unicode_to_sjis_converter($b); # convert my $n = unpack('n', $s); # convert sjis to short [ $n, 0x20, 0x2, $u ]; # return the collation element }, The return value may be an arrayref of 1st to 4th weights as shown above. The return value may be an integer as the primary weight as shown below. If C is returned, the default derived collation element will be used. overrideCJK => sub { my $u = shift; # get a Unicode codepoint my $b = pack('n', $u); # to UTF-16BE my $s = your_unicode_to_sjis_converter($b); # convert my $n = unpack('n', $s); # convert sjis to short return $n; # return the primary weight }, The return value may be a list containing zero or more of an arrayref, an integer, or C. ex. ignores all CJK unified ideographs. overrideCJK => sub {()}, # CODEREF returning empty list # where ->eq("Pe\x{4E00}rl", "Perl") is true # as U+4E00 is a CJK unified ideograph and to be ignorable. If C is passed explicitly as the value for this key, weights for CJK unified ideographs are treated as undefined. But assignment of weight for CJK unified ideographs in C
or C is still valid. B In addition to them, 12 CJK compatibility ideographs (C, C, C, C, C, C, C, C, C, C, C, C) are also treated as CJK unified ideographs. But they can't be overridden via C when you use DUCET, as the table includes weights for them. C
or C has priority over C. =item overrideHangul -- see 7.1 Derived Collation Elements, UTS #10. By default, Hangul syllables are decomposed into Hangul Jamo, even if C<(normalization =E undef)>. But the mapping of Hangul syllables may be overridden. This parameter works like C, so see there for examples. If you want to override the mapping of Hangul syllables, NFD and NFKD are not appropriate, since NFD and NFKD will decompose Hangul syllables before overriding. FCD may decompose Hangul syllables as the case may be. If C is passed explicitly as the value for this key, weight for Hangul syllables is treated as undefined without decomposition into Hangul Jamo. But definition of weight for Hangul syllables in C
or C is still valid. =item preprocess -- see 5.1 Preprocessing, UTS #10. If specified, the coderef is used to preprocess each string before the formation of sort keys. ex. dropping English articles, such as "a" or "the". Then, "the pen" is before "a pencil". preprocess => sub { my $str = shift; $str =~ s/\b(?:an?|the)\s+//gi; return $str; }, C is performed before C (if defined). ex. decoding strings in a legacy encoding such as shift-jis: $sjis_collator = Unicode::Collate->new( preprocess => \&your_shiftjis_to_unicode_decoder, ); @result = $sjis_collator->sort(@shiftjis_strings); B Strings returned from the coderef will be interpreted according to Perl's Unicode support. See L, L, L, L, L. =item rearrange -- see 3.1.3 Rearrangement, UTS #10. Characters that are not coded in logical order and to be rearranged. If C is equal to or lesser than 11, default is: rearrange => [ 0x0E40..0x0E44, 0x0EC0..0x0EC4 ], If you want to disallow any rearrangement, pass C or C<[]> (a reference to empty list) as the value for this key. If C is equal to or greater than 14, default is C<[]> (i.e. no rearrangement). B =item rewrite If specified, the coderef is used to rewrite lines in C
or C. The coderef will get each line, and then should return a rewritten line according to the UCA file format. If the coderef returns an empty line, the line will be skipped. e.g. any primary ignorable characters into tertiary ignorable: rewrite => sub { my $line = shift; $line =~ s/\[\.0000\..{4}\..{4}\./[.0000.0000.0000./g; return $line; }, This example shows rewriting weights. C is allowed to affect code points, weights, and the name. B: C
is available to use another table file; preparing a modified table once would be more efficient than rewriting lines on reading an unmodified table every time. =item suppress -- see suppress contractions in 5.14.11 Special-Purpose Commands, UTS #35 (LDML). Contractions beginning with the specified characters are suppressed, even if those contractions are defined in C
. An example for Russian and some languages using the Cyrillic script: suppress => [0x0400..0x0417, 0x041A..0x0437, 0x043A..0x045F], where 0x0400 stands for C, CYRILLIC CAPITAL LETTER IE WITH GRAVE. B: Contractions via C are not be suppressed. =item table -- see 3.2 Default Unicode Collation Element Table, UTS #10. You can use another collation element table if desired. The table file should locate in the F directory on C<@INC>. Say, if the filename is F, the table file is searched as F in C<@INC>. By default, F (as the filename of DUCET) is used. If you will prepare your own table file, any name other than F may be better to avoid namespace conflict. B: When XSUB is used, the DUCET is compiled on building this module, and it may save time at the run time. Explicit saying C
'allkeys.txt'> (or using another table), or using C, C, C, C or C will prevent this module from using the compiled DUCET. If C is passed explicitly as the value for this key, no file is read (but you can define collation elements via C). A typical way to define a collation element table without any file of table: $onlyABC = Unicode::Collate->new( table => undef, entry => << 'ENTRIES', 0061 ; [.0101.0020.0002.0061] # LATIN SMALL LETTER A 0041 ; [.0101.0020.0008.0041] # LATIN CAPITAL LETTER A 0062 ; [.0102.0020.0002.0062] # LATIN SMALL LETTER B 0042 ; [.0102.0020.0008.0042] # LATIN CAPITAL LETTER B 0063 ; [.0103.0020.0002.0063] # LATIN SMALL LETTER C 0043 ; [.0103.0020.0008.0043] # LATIN CAPITAL LETTER C ENTRIES ); If C or C is used, character names should be specified as a comment (following C<#>) on each line. =item undefChar =item undefName -- see 6.3.4 Reducing the Repertoire, UTS #10. Undefines the collation element as if it were unassigned in the C
. This reduces the size of the table. If an unassigned character appears in the string to be collated, the sort key is made from its codepoint as a single-character collation element, as it is greater than any other assigned collation elements (in the codepoint order among the unassigned characters). But, it'd be better to ignore characters unfamiliar to you and maybe never used. Through C, any character matching C will be undefined. Through C, any character whose name (given in the C
file as a comment) matches C will be undefined. ex. Collation weights for beyond-BMP characters are not stored in object: undefChar => qr/[^\0-\x{fffd}]/, =item upper_before_lower -- see 6.6 Case Comparisons, UTS #10. By default, lowercase is before uppercase. If the parameter is made true, this is reversed. B: This parameter simplemindedly assumes that any lowercase/uppercase distinctions must occur in level 3, and their weights at level 3 must be same as those mentioned in 7.3.1, UTS #10. If you define your collation elements which differs from this requirement, this parameter doesn't work validly. =item variable -- see 3.2.2 Variable Weighting, UTS #10. This key allows for variable weighting of variable collation elements, which are marked with an ASTERISK in the table (NOTE: Many punctuation marks and symbols are variable in F). variable => 'blanked', 'non-ignorable', 'shifted', or 'shift-trimmed'. These names are case-insensitive. By default (if specification is omitted), 'shifted' is adopted. 'Blanked' Variable elements are made ignorable at levels 1 through 3; considered at the 4th level. 'Non-Ignorable' Variable elements are not reset to ignorable. 'Shifted' Variable elements are made ignorable at levels 1 through 3 their level 4 weight is replaced by the old level 1 weight. Level 4 weight for Non-Variable elements is 0xFFFF. 'Shift-Trimmed' Same as 'shifted', but all FFFF's at the 4th level are trimmed. =back =head2 Methods for Collation =over 4 =item C<@sorted = $Collator-Esort(@not_sorted)> Sorts a list of strings. =item C<$result = $Collator-Ecmp($a, $b)> Returns 1 (when C<$a> is greater than C<$b>) or 0 (when C<$a> is equal to C<$b>) or -1 (when C<$a> is lesser than C<$b>). =item C<$result = $Collator-Eeq($a, $b)> =item C<$result = $Collator-Ene($a, $b)> =item C<$result = $Collator-Elt($a, $b)> =item C<$result = $Collator-Ele($a, $b)> =item C<$result = $Collator-Egt($a, $b)> =item C<$result = $Collator-Ege($a, $b)> They works like the same name operators as theirs. eq : whether $a is equal to $b. ne : whether $a is not equal to $b. lt : whether $a is lesser than $b. le : whether $a is lesser than $b or equal to $b. gt : whether $a is greater than $b. ge : whether $a is greater than $b or equal to $b. =item C<$sortKey = $Collator-EgetSortKey($string)> -- see 4.3 Form Sort Key, UTS #10. Returns a sort key. You compare the sort keys using a binary comparison and get the result of the comparison of the strings using UCA. $Collator->getSortKey($a) cmp $Collator->getSortKey($b) is equivalent to $Collator->cmp($a, $b) =item C<$sortKeyForm = $Collator-EviewSortKey($string)> Converts a sorting key into its representation form. If C is 8, the output is slightly different. use Unicode::Collate; my $c = Unicode::Collate->new(); print $c->viewSortKey("Perl"),"\n"; # output: # [0B67 0A65 0B7F 0B03 | 0020 0020 0020 0020 | 0008 0002 0002 0002 | FFFF FFFF FFFF FFFF] # Level 1 Level 2 Level 3 Level 4 =back =head2 Methods for Searching The C, C, C, C methods work like C, C, C, C, respectively, but they are not aware of any pattern, but only a literal substring. B If C or C parameter is true for C<$Collator>, calling these methods (C, C, C, C, C) is croaked, as the position and the length might differ from those on the specified string. C and C parameters are neglected. C and C don't affect matching and searching, as it doesn't matter whether greater or lesser. =over 4 =item C<$position = $Collator-Eindex($string, $substring[, $position])> =item C<($position, $length) = $Collator-Eindex($string, $substring[, $position])> If C<$substring> matches a part of C<$string>, returns the position of the first occurrence of the matching part in scalar context; in list context, returns a two-element list of the position and the length of the matching part. If C<$substring> does not match any part of C<$string>, returns C<-1> in scalar context and an empty list in list context. e.g. you say my $Collator = Unicode::Collate->new( normalization => undef, level => 1 ); # (normalization => undef) is REQUIRED. my $str = "Ich muß studieren Perl."; my $sub = "MÜSS"; my $match; if (my($pos,$len) = $Collator->index($str, $sub)) { $match = substr($str, $pos, $len); } and get C<"muß"> in C<$match> since C<"muß"> is primary equal to C<"MÜSS">. =item C<$match_ref = $Collator-Ematch($string, $substring)> =item C<($match) = $Collator-Ematch($string, $substring)> If C<$substring> matches a part of C<$string>, in scalar context, returns B the first occurrence of the matching part (C<$match_ref> is always true if matches, since every reference is B); in list context, returns the first occurrence of the matching part. If C<$substring> does not match any part of C<$string>, returns C in scalar context and an empty list in list context. e.g. if ($match_ref = $Collator->match($str, $sub)) { # scalar context print "matches [$$match_ref].\n"; } else { print "doesn't match.\n"; } or if (($match) = $Collator->match($str, $sub)) { # list context print "matches [$match].\n"; } else { print "doesn't match.\n"; } =item C<@match = $Collator-Egmatch($string, $substring)> If C<$substring> matches a part of C<$string>, returns all the matching parts (or matching count in scalar context). If C<$substring> does not match any part of C<$string>, returns an empty list. =item C<$count = $Collator-Esubst($string, $substring, $replacement)> If C<$substring> matches a part of C<$string>, the first occurrence of the matching part is replaced by C<$replacement> (C<$string> is modified) and C<$count> (always equals to C<1>) is returned. C<$replacement> can be a C, taking the matching part as an argument, and returning a string to replace the matching part (a bit similar to C($1)/e>). =item C<$count = $Collator-Egsubst($string, $substring, $replacement)> If C<$substring> matches a part of C<$string>, all the occurrences of the matching part are replaced by C<$replacement> (C<$string> is modified) and C<$count> is returned. C<$replacement> can be a C, taking the matching part as an argument, and returning a string to replace the matching part (a bit similar to C($1)/eg>). e.g. my $Collator = Unicode::Collate->new( normalization => undef, level => 1 ); # (normalization => undef) is REQUIRED. my $str = "Camel donkey zebra came\x{301}l CAMEL horse cam\0e\0l..."; $Collator->gsubst($str, "camel", sub { "$_[0]" }); # now $str is "Camel donkey zebra came\x{301}l CAMEL horse cam\0e\0l..."; # i.e., all the camels are made bold-faced. Examples: levels and ignore_level2 - what does camel match? --------------------------------------------------------------------------- level ignore_level2 | camel Camel came\x{301}l c-a-m-e-l cam\0e\0l -----------------------|--------------------------------------------------- 1 false | yes yes yes yes yes 2 false | yes yes no yes yes 3 false | yes no no yes yes 4 false | yes no no no yes -----------------------|--------------------------------------------------- 1 true | yes yes yes yes yes 2 true | yes yes yes yes yes 3 true | yes no yes yes yes 4 true | yes no yes no yes --------------------------------------------------------------------------- note: if variable => non-ignorable, camel doesn't match c-a-m-e-l at any level. =back =head2 Other Methods =over 4 =item C<%old_tailoring = $Collator-Echange(%new_tailoring)> =item C<$modified_collator = $Collator-Echange(%new_tailoring)> Changes the value of specified keys and returns the changed part. $Collator = Unicode::Collate->new(level => 4); $Collator->eq("perl", "PERL"); # false %old = $Collator->change(level => 2); # returns (level => 4). $Collator->eq("perl", "PERL"); # true $Collator->change(%old); # returns (level => 2). $Collator->eq("perl", "PERL"); # false Not all C<(key,value)>s are allowed to be changed. See also C<@Unicode::Collate::ChangeOK> and C<@Unicode::Collate::ChangeNG>. In the scalar context, returns the modified collator (but it is B a clone from the original). $Collator->change(level => 2)->eq("perl", "PERL"); # true $Collator->eq("perl", "PERL"); # true; now max level is 2nd. $Collator->change(level => 4)->eq("perl", "PERL"); # false =item C<$version = $Collator-Eversion()> Returns the version number (a string) of the Unicode Standard which the C
file used by the collator object is based on. If the table does not include a version line (starting with C<@version>), returns C<"unknown">. =item C Returns the revision number of UTS #10 this module consults, that should correspond with the DUCET incorporated. =item C Returns the version number of UTS #10 this module consults, that should correspond with the DUCET incorporated. =back =head1 EXPORT No method will be exported. =head1 INSTALL Though this module can be used without any C
file, to use this module easily, it is recommended to install a table file in the UCA format, by copying it under the directory /Unicode/Collate. The most preferable one is "The Default Unicode Collation Element Table" (aka DUCET), available from the Unicode Consortium's website: http://www.unicode.org/Public/UCA/ http://www.unicode.org/Public/UCA/latest/allkeys.txt (latest version) If DUCET is not installed, it is recommended to copy the file from http://www.unicode.org/Public/UCA/latest/allkeys.txt to /Unicode/Collate/allkeys.txt manually. =head1 CAVEATS =over 4 =item Normalization Use of the C parameter requires the B module (see L). If you need not it (say, in the case when you need not handle any combining characters), assign C undef> explicitly. -- see 6.5 Avoiding Normalization, UTS #10. =item Conformance Test The Conformance Test for the UCA is available under L. For F, a collator via Cnew( )> should be used; for F, a collator via Cnew(variable =E "non-ignorable", level =E 3)>. B =back =head1 AUTHOR, COPYRIGHT AND LICENSE The Unicode::Collate module for perl was written by SADAHIRO Tomoyuki, . This module is Copyright(C) 2001-2012, SADAHIRO Tomoyuki. Japan. All rights reserved. This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself. The file Unicode/Collate/allkeys.txt was copied verbatim from L. For this file, Copyright (c) 2001-2011 Unicode, Inc. Distributed under the Terms of Use in L. =head1 SEE ALSO =over 4 =item Unicode Collation Algorithm - UTS #10 L =item The Default Unicode Collation Element Table (DUCET) L =item The conformance test for the UCA L L =item Hangul Syllable Type L =item Unicode Normalization Forms - UAX #15 L =item Unicode Locale Data Markup Language (LDML) - UTS #35 L =back =cut