package SQL::Translator::Schema::Table;
# ----------------------------------------------------------------------
-# $Id: Table.pm,v 1.15 2003-08-29 05:38:56 allenday Exp $
-# ----------------------------------------------------------------------
-# Copyright (C) 2003 Ken Y. Clark <kclark@cpan.org>
+# Copyright (C) 2002-2009 SQLFairy Authors
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License as
=cut
use strict;
-use Class::Base;
use SQL::Translator::Utils 'parse_list_arg';
use SQL::Translator::Schema::Constants;
use SQL::Translator::Schema::Constraint;
use SQL::Translator::Schema::Field;
use SQL::Translator::Schema::Index;
+use Data::Dumper;
+
+use base 'SQL::Translator::Schema::Object';
-use base 'Class::Base';
-use vars qw( $VERSION $FIELD_ORDER );
+use vars qw( $VERSION );
-$VERSION = sprintf "%d.%02d", q$Revision: 1.15 $ =~ /(\d+)\.(\d+)/;
+$VERSION = '1.59';
+
+# Stringify to our name, being careful not to pass any args through so we don't
+# accidentally set it to undef. We also have to tweak bool so the object is
+# still true when it doesn't have a name (which shouldn't happen!).
+use overload
+ '""' => sub { shift->name },
+ 'bool' => sub { $_[0]->name || $_[0] },
+ fallback => 1,
+;
# ----------------------------------------------------------------------
-sub init {
+
+__PACKAGE__->_attributes( qw/schema name comments options order/ );
=pod
=cut
- my ( $self, $config ) = @_;
-
- for my $arg ( qw[ schema name comments ] ) {
- next unless defined $config->{ $arg };
- defined $self->$arg( $config->{ $arg } ) or return;
- }
+sub new {
+ my $class = shift;
+ my $self = $class->SUPER::new (@_)
+ or return;
+
+ $self->{_order} = { map { $_ => 0 } qw/
+ field
+ /};
- return $self;
+ return $self;
}
+
+
# ----------------------------------------------------------------------
sub add_constraint {
Add a constraint to the table. Returns the newly created
C<SQL::Translator::Schema::Constraint> object.
- my $c1 = $table->add_constraint(
- name => 'pk',
- type => PRIMARY_KEY,
- fields => [ 'foo_id' ],
+ my $c1 = $table->add_constraint(
+ name => 'pk',
+ type => PRIMARY_KEY,
+ fields => [ 'foo_id' ],
);
my $c2 = SQL::Translator::Schema::Constraint->new( name => 'uniq' );
my %args = @_;
$args{'table'} = $self;
$constraint = $constraint_class->new( \%args ) or
- return $self->error( $constraint_class->error );
+ return $self->error( $constraint_class->error );
}
#
my $pk = $self->primary_key;
if ( $pk && $constraint->type eq PRIMARY_KEY ) {
$self->primary_key( $constraint->fields );
+ $pk->name($constraint->name) if $constraint->name;
+ my %extra = $constraint->extra;
+ $pk->extra(%extra) if keys %extra;
$constraint = $pk;
$ok = 0;
}
}
#
# See if another constraint of the same type
- # covers the same fields.
+ # covers the same fields. -- This doesn't work! ky
#
- elsif ( $constraint->type ne CHECK_C ) {
- my @field_names = $constraint->fields;
- for my $c (
- grep { $_->type eq $constraint->type }
- $self->get_constraints
- ) {
- my %fields = map { $_, 1 } $c->fields;
- for my $field_name ( @field_names ) {
- if ( $fields{ $field_name } ) {
- $constraint = $c;
- $ok = 0;
- last;
- }
- }
- last unless $ok;
- }
- }
+# elsif ( $constraint->type ne CHECK_C ) {
+# my @field_names = $constraint->fields;
+# for my $c (
+# grep { $_->type eq $constraint->type }
+# $self->get_constraints
+# ) {
+# my %fields = map { $_, 1 } $c->fields;
+# for my $field_name ( @field_names ) {
+# if ( $fields{ $field_name } ) {
+# $constraint = $c;
+# $ok = 0;
+# last;
+# }
+# }
+# last unless $ok;
+# }
+# }
if ( $ok ) {
push @{ $self->{'constraints'} }, $constraint;
}
# ----------------------------------------------------------------------
+sub drop_constraint {
+
+=pod
+
+=head2 drop_constraint
+
+Remove a constraint from the table. Returns the constraint object if the index
+was found and removed, an error otherwise. The single parameter can be either
+an index name or an C<SQL::Translator::Schema::Constraint> object.
+
+ $table->drop_constraint('myconstraint');
+
+=cut
+
+ my $self = shift;
+ my $constraint_class = 'SQL::Translator::Schema::Constraint';
+ my $constraint_name;
+
+ if ( UNIVERSAL::isa( $_[0], $constraint_class ) ) {
+ $constraint_name = shift->name;
+ }
+ else {
+ $constraint_name = shift;
+ }
+
+ if ( ! grep { $_->name eq $constraint_name } @ { $self->{'constraints'} } ) {
+ return $self->error(qq[Can't drop constraint: "$constraint_name" doesn't exist]);
+ }
+
+ my @cs = @{ $self->{'constraints'} };
+ my ($constraint_id) = grep { $cs[$_]->name eq $constraint_name } (0..$#cs);
+ my $constraint = splice(@{$self->{'constraints'}}, $constraint_id, 1);
+
+ return $constraint;
+}
+
+# ----------------------------------------------------------------------
sub add_index {
=pod
Add an index to the table. Returns the newly created
C<SQL::Translator::Schema::Index> object.
- my $i1 = $table->add_index(
+ my $i1 = $table->add_index(
name => 'name',
fields => [ 'name' ],
type => 'normal',
$index = $index_class->new( \%args ) or return
$self->error( $index_class->error );
}
-
+ foreach my $ex_index ($self->get_indices) {
+ return if ($ex_index->equals($index));
+ }
push @{ $self->{'indices'} }, $index;
return $index;
}
# ----------------------------------------------------------------------
+sub drop_index {
+
+=pod
+
+=head2 drop_index
+
+Remove an index from the table. Returns the index object if the index was
+found and removed, an error otherwise. The single parameter can be either
+an index name of an C<SQL::Translator::Schema::Index> object.
+
+ $table->drop_index('myindex');
+
+=cut
+
+ my $self = shift;
+ my $index_class = 'SQL::Translator::Schema::Index';
+ my $index_name;
+
+ if ( UNIVERSAL::isa( $_[0], $index_class ) ) {
+ $index_name = shift->name;
+ }
+ else {
+ $index_name = shift;
+ }
+
+ if ( ! grep { $_->name eq $index_name } @{ $self->{'indices'} }) {
+ return $self->error(qq[Can't drop index: "$index_name" doesn't exist]);
+ }
+
+ my @is = @{ $self->{'indices'} };
+ my ($index_id) = grep { $is[$_]->name eq $index_name } (0..$#is);
+ my $index = splice(@{$self->{'indices'}}, $index_id, 1);
+
+ return $index;
+}
+
+# ----------------------------------------------------------------------
sub add_field {
=pod
required. If you try to create a field with the same name as an
existing field, you will get an error and the field will not be created.
- my $f1 = $table->add_field(
+ my $f1 = $table->add_field(
name => 'foo_id',
data_type => 'integer',
size => 11,
);
- my $f2 = SQL::Translator::Schema::Field->new(
+ my $f2 = SQL::Translator::Schema::Field->new(
name => 'name',
table => $table,
);
- $f2 = $table->add_field( $field2 ) or die $table->error;
+ $f2 = $table->add_field( $field2 ) or die $table->error;
=cut
$self->error( $field_class->error );
}
- $field->order( ++$FIELD_ORDER );
- my $field_name = $field->name or return $self->error('No name');
+ $field->order( ++$self->{_order}{field} );
+ # We know we have a name as the Field->new above errors if none given.
+ my $field_name = $field->name;
if ( exists $self->{'fields'}{ $field_name } ) {
- return $self->error(qq[Can\'t create field: "$field_name" exists]);
+ return $self->error(qq[Can't create field: "$field_name" exists]);
}
else {
$self->{'fields'}{ $field_name } = $field;
return $field;
}
+# ----------------------------------------------------------------------
+sub drop_field {
+
+=pod
+
+=head2 drop_field
+
+Remove a field from the table. Returns the field object if the field was
+found and removed, an error otherwise. The single parameter can be either
+a field name or an C<SQL::Translator::Schema::Field> object.
+
+ $table->drop_field('myfield');
+
+=cut
+
+ my $self = shift;
+ my $field_class = 'SQL::Translator::Schema::Field';
+ my $field_name;
+
+ if ( UNIVERSAL::isa( $_[0], $field_class ) ) {
+ $field_name = shift->name;
+ }
+ else {
+ $field_name = shift;
+ }
+ my %args = @_;
+ my $cascade = $args{'cascade'};
+
+ if ( ! exists $self->{'fields'}{ $field_name } ) {
+ return $self->error(qq[Can't drop field: "$field_name" doesn't exists]);
+ }
+
+ my $field = delete $self->{'fields'}{ $field_name };
+
+ if ( $cascade ) {
+ # Remove this field from all indices using it
+ foreach my $i ($self->get_indices()) {
+ my @fs = $i->fields();
+ @fs = grep { $_ ne $field->name } @fs;
+ $i->fields(@fs);
+ }
+
+ # Remove this field from all constraints using it
+ foreach my $c ($self->get_constraints()) {
+ my @cs = $c->fields();
+ @cs = grep { $_ ne $field->name } @cs;
+ $c->fields(@cs);
+ }
+ }
+
+ return $field;
+}
# ----------------------------------------------------------------------
sub comments {
my $self = shift;
my $field_name = shift or return $self->error('No field name');
+ my $case_insensitive = shift;
+ if ( $case_insensitive ) {
+ $field_name = uc($field_name);
+ foreach my $field ( keys %{$self->{fields}} ) {
+ return $self->{fields}{$field} if $field_name eq uc($field);
+ }
+ return $self->error(qq[Field "$field_name" does not exist]);
+ }
return $self->error( qq[Field "$field_name" does not exist] ) unless
exists $self->{'fields'}{ $field_name };
return $self->{'fields'}{ $field_name };
return 1;
}
-sub is_data {
- my $self = shift;
- return $self->{'is_data'} if defined $self->{'is_data'};
+# ----------------------------------------------------------------------
+sub is_trivial_link {
- $self->{'is_data'} = 0;
+=pod
+
+=head2 is_trivial_link
+
+True if table has no data (non-key) fields and only uses single key joins.
+
+=cut
+
+ my $self = shift;
+ return 0 if $self->is_data;
+ return $self->{'is_trivial_link'} if defined $self->{'is_trivial_link'};
- foreach my $field ($self->get_fields){
- if(!$field->is_primary_key or !$field->is_foreign_key){
- $self->{'is_data'} = 1;
- return $self->{'is_data'}
+ $self->{'is_trivial_link'} = 1;
+
+ my %fk = ();
+
+ foreach my $field ( $self->get_fields ) {
+ next unless $field->is_foreign_key;
+ $fk{$field->foreign_key_reference->reference_table}++;
}
- }
- return $self->{'is_data'};
+ foreach my $referenced (keys %fk){
+ if($fk{$referenced} > 1){
+ $self->{'is_trivial_link'} = 0;
+ last;
+ }
+ }
+
+ return $self->{'is_trivial_link'};
+
}
+sub is_data {
+
+=pod
+
+=head2 is_data
+
+Returns true if the table has some non-key fields.
+
+=cut
+
+ my $self = shift;
+ return $self->{'is_data'} if defined $self->{'is_data'};
+
+ $self->{'is_data'} = 0;
+
+ foreach my $field ( $self->get_fields ) {
+ if ( !$field->is_primary_key and !$field->is_foreign_key ) {
+ $self->{'is_data'} = 1;
+ return $self->{'is_data'};
+ }
+ }
+
+ return $self->{'is_data'};
+}
+
+# ----------------------------------------------------------------------
sub can_link {
=pod
=cut
- my($self,$table1,$table2) = @_;
+ my ( $self, $table1, $table2 ) = @_;
- #get tables in abc order
- ($table1,$table2) = sort {$a->name cmp $b->name} ($table1,$table2);
+ return $self->{'can_link'}{ $table1->name }{ $table2->name }
+ if defined $self->{'can_link'}{ $table1->name }{ $table2->name };
- return $self->{'can_link'}{$table1->name}{$table2->name} if defined $self->{'can_link'}{$table1->name}{$table2->name};
+ if ( $self->is_data == 1 ) {
+ $self->{'can_link'}{ $table1->name }{ $table2->name } = [0];
+ $self->{'can_link'}{ $table2->name }{ $table1->name } = [0];
+ return $self->{'can_link'}{ $table1->name }{ $table2->name };
+ }
- if($self->is_data == 1){
- $self->{'can_link'}{$table1->name}{$table2->name} = 0;
- return $self->{'can_link'}{$table1->name}{$table2->name};
- }
+ my %fk = ();
+
+ foreach my $field ( $self->get_fields ) {
+ if ( $field->is_foreign_key ) {
+ push @{ $fk{ $field->foreign_key_reference->reference_table } },
+ $field->foreign_key_reference;
+ }
+ }
- my %fk = ();
+ if ( !defined( $fk{ $table1->name } ) or !defined( $fk{ $table2->name } ) )
+ {
+ $self->{'can_link'}{ $table1->name }{ $table2->name } = [0];
+ $self->{'can_link'}{ $table2->name }{ $table1->name } = [0];
+ return $self->{'can_link'}{ $table1->name }{ $table2->name };
+ }
- foreach my $field ($self->get_fields){
- #if the table has non-key fields, it can't be a link
- if(!$field->is_primary_key or !$field->is_foreign_key){
- $self->{'can_link'}{$table1->name}{$table2->name} = 0;
- return $self->{'can_link'}{$table1->name}{$table2->name};
+ # trivial traversal, only one way to link the two tables
+ if ( scalar( @{ $fk{ $table1->name } } == 1 )
+ and scalar( @{ $fk{ $table2->name } } == 1 ) )
+ {
+ $self->{'can_link'}{ $table1->name }{ $table2->name } =
+ [ 'one2one', $fk{ $table1->name }, $fk{ $table2->name } ];
+ $self->{'can_link'}{ $table1->name }{ $table2->name } =
+ [ 'one2one', $fk{ $table2->name }, $fk{ $table1->name } ];
+
+ # non-trivial traversal. one way to link table2,
+ # many ways to link table1
+ }
+ elsif ( scalar( @{ $fk{ $table1->name } } > 1 )
+ and scalar( @{ $fk{ $table2->name } } == 1 ) )
+ {
+ $self->{'can_link'}{ $table1->name }{ $table2->name } =
+ [ 'many2one', $fk{ $table1->name }, $fk{ $table2->name } ];
+ $self->{'can_link'}{ $table2->name }{ $table1->name } =
+ [ 'one2many', $fk{ $table2->name }, $fk{ $table1->name } ];
+
+ # non-trivial traversal. one way to link table1,
+ # many ways to link table2
+ }
+ elsif ( scalar( @{ $fk{ $table1->name } } == 1 )
+ and scalar( @{ $fk{ $table2->name } } > 1 ) )
+ {
+ $self->{'can_link'}{ $table1->name }{ $table2->name } =
+ [ 'one2many', $fk{ $table1->name }, $fk{ $table2->name } ];
+ $self->{'can_link'}{ $table2->name }{ $table1->name } =
+ [ 'many2one', $fk{ $table2->name }, $fk{ $table1->name } ];
+
+ # non-trivial traversal. many ways to link table1 and table2
+ }
+ elsif ( scalar( @{ $fk{ $table1->name } } > 1 )
+ and scalar( @{ $fk{ $table2->name } } > 1 ) )
+ {
+ $self->{'can_link'}{ $table1->name }{ $table2->name } =
+ [ 'many2many', $fk{ $table1->name }, $fk{ $table2->name } ];
+ $self->{'can_link'}{ $table2->name }{ $table1->name } =
+ [ 'many2many', $fk{ $table2->name }, $fk{ $table1->name } ];
+
+ # one of the tables didn't export a key
+ # to this table, no linking possible
+ }
+ else {
+ $self->{'can_link'}{ $table1->name }{ $table2->name } = [0];
+ $self->{'can_link'}{ $table2->name }{ $table1->name } = [0];
+ }
- #otherwise, count up how many fields refer to each FK table.field
- } elsif($field->is_foreign_key){
- $fk{$field->foreign_key_reference->reference_table->name}++;
- }
- }
-
- if($fk{ $table1->name } == 1
- and
- $fk{ $table2->name } == 1
- ){
- $self->{'can_link'}{$table1->name}{$table2->name} = 1;
- } else {
- $self->{'can_link'}{$table1->name}{$table2->name} = 0;
- }
-
- return $self->{'can_link'}{$table1->name}{$table2->name};
+ return $self->{'can_link'}{ $table1->name }{ $table2->name };
}
# ----------------------------------------------------------------------
=head2 name
-Get or set the table\'s name.
+Get or set the table's name.
+
+Errors ("No table name") if you try to set a blank name.
-If provided an argument, checks the schema object for a table of
-that name and disallows the change if one exists.
+If provided an argument, checks the schema object for a table of
+that name and disallows the change if one exists (setting the error to
+"Can't use table name "%s": table exists").
my $table_name = $table->name('foo');
my $self = shift;
- if ( my $arg = shift ) {
+ if ( @_ ) {
+ my $arg = shift || return $self->error( "No table name" );
if ( my $schema = $self->schema ) {
- return $self->error( qq[Can\'t use table name "$arg": table exists] )
+ return $self->error( qq[Can't use table name "$arg": table exists] )
if $schema->get_table( $arg );
}
$self->{'name'} = $arg;
=head2 schema
-Get or set the table\'s schema object.
+Get or set the table's schema object.
my $schema = $table->schema;
=pod
-=head2 options
+=head2 primary_key
-Gets or sets the table\'s primary key(s). Takes one or more field
+Gets or sets the table's primary key(s). Takes one or more field
names (as a string, list or array[ref]) as an argument. If the field
names are present, it will create a new PK if none exists, or it will
add to the fields of an existing PK (and will unique the field names).
=head2 options
-Get or set the table\'s options (e.g., table types for MySQL). Returns
+Get or set the table's options (e.g., table types for MySQL). Returns
an array or array reference.
my @options = $table->options;
push @{ $self->{'options'} }, @$options;
if ( ref $self->{'options'} ) {
- return wantarray ? @{ $self->{'options'} || [] } : $self->{'options'};
+ return wantarray ? @{ $self->{'options'} || [] } : ($self->{'options'} || '');
}
else {
return wantarray ? () : [];
=head2 order
-Get or set the table\'s order.
+Get or set the table's order.
my $order = $table->order(3);
}
# ----------------------------------------------------------------------
+sub field_names {
+
+=head2 field_names
+
+Read-only method to return a list or array ref of the field names. Returns undef
+or an empty list if the table has no fields set. Useful if you want to
+avoid the overload magic of the Field objects returned by the get_fields method.
+
+ my @names = $constraint->field_names;
+
+=cut
+
+ my $self = shift;
+ my @fields =
+ map { $_->name }
+ sort { $a->order <=> $b->order }
+ values %{ $self->{'fields'} || {} };
+
+ if ( @fields ) {
+ return wantarray ? @fields : \@fields;
+ }
+ else {
+ $self->error('No fields');
+ return wantarray ? () : undef;
+ }
+}
+
+# ----------------------------------------------------------------------
+sub equals {
+
+=pod
+
+=head2 equals
+
+Determines if this table is the same as another
+
+ my $isIdentical = $table1->equals( $table2 );
+
+=cut
+
+ my $self = shift;
+ my $other = shift;
+ my $case_insensitive = shift;
+
+ return 0 unless $self->SUPER::equals($other);
+ return 0 unless $case_insensitive ? uc($self->name) eq uc($other->name) : $self->name eq $other->name;
+ return 0 unless $self->_compare_objects(scalar $self->options, scalar $other->options);
+ return 0 unless $self->_compare_objects(scalar $self->extra, scalar $other->extra);
+
+ # Fields
+ # Go through our fields
+ my %checkedFields;
+ foreach my $field ( $self->get_fields ) {
+ my $otherField = $other->get_field($field->name, $case_insensitive);
+ return 0 unless $field->equals($otherField, $case_insensitive);
+ $checkedFields{$field->name} = 1;
+ }
+ # Go through the other table's fields
+ foreach my $otherField ( $other->get_fields ) {
+ next if $checkedFields{$otherField->name};
+ return 0;
+ }
+
+ # Constraints
+ # Go through our constraints
+ my %checkedConstraints;
+CONSTRAINT:
+ foreach my $constraint ( $self->get_constraints ) {
+ foreach my $otherConstraint ( $other->get_constraints ) {
+ if ( $constraint->equals($otherConstraint, $case_insensitive) ) {
+ $checkedConstraints{$otherConstraint} = 1;
+ next CONSTRAINT;
+ }
+ }
+ return 0;
+ }
+ # Go through the other table's constraints
+CONSTRAINT2:
+ foreach my $otherConstraint ( $other->get_constraints ) {
+ next if $checkedFields{$otherConstraint};
+ foreach my $constraint ( $self->get_constraints ) {
+ if ( $otherConstraint->equals($constraint, $case_insensitive) ) {
+ next CONSTRAINT2;
+ }
+ }
+ return 0;
+ }
+
+ # Indices
+ # Go through our indices
+ my %checkedIndices;
+INDEX:
+ foreach my $index ( $self->get_indices ) {
+ foreach my $otherIndex ( $other->get_indices ) {
+ if ( $index->equals($otherIndex, $case_insensitive) ) {
+ $checkedIndices{$otherIndex} = 1;
+ next INDEX;
+ }
+ }
+ return 0;
+ }
+ # Go through the other table's indices
+INDEX2:
+ foreach my $otherIndex ( $other->get_indices ) {
+ next if $checkedIndices{$otherIndex};
+ foreach my $index ( $self->get_indices ) {
+ if ( $otherIndex->equals($index, $case_insensitive) ) {
+ next INDEX2;
+ }
+ }
+ return 0;
+ }
+
+ return 1;
+}
+
+# ----------------------------------------------------------------------
+
+=head1 LOOKUP METHODS
+
+The following are a set of shortcut methods for getting commonly used lists of
+fields and constraints. They all return lists or array refs of Field or
+Constraint objects.
+
+=over 4
+
+=item pkey_fields
+
+The primary key fields.
+
+=item fkey_fields
+
+All foreign key fields.
+
+=item nonpkey_fields
+
+All the fields except the primary key.
+
+=item data_fields
+
+All non key fields.
+
+=item unique_fields
+
+All fields with unique constraints.
+
+=item unique_constraints
+
+All this tables unique constraints.
+
+=item fkey_constraints
+
+All this tables foreign key constraints. (See primary_key method to get the
+primary key constraint)
+
+=back
+
+=cut
+
+sub pkey_fields {
+ my $me = shift;
+ my @fields = grep { $_->is_primary_key } $me->get_fields;
+ return wantarray ? @fields : \@fields;
+}
+
+# ----------------------------------------------------------------------
+sub fkey_fields {
+ my $me = shift;
+ my @fields;
+ push @fields, $_->fields foreach $me->fkey_constraints;
+ return wantarray ? @fields : \@fields;
+}
+
+# ----------------------------------------------------------------------
+sub nonpkey_fields {
+ my $me = shift;
+ my @fields = grep { !$_->is_primary_key } $me->get_fields;
+ return wantarray ? @fields : \@fields;
+}
+
+# ----------------------------------------------------------------------
+sub data_fields {
+ my $me = shift;
+ my @fields =
+ grep { !$_->is_foreign_key and !$_->is_primary_key } $me->get_fields;
+ return wantarray ? @fields : \@fields;
+}
+
+# ----------------------------------------------------------------------
+sub unique_fields {
+ my $me = shift;
+ my @fields;
+ push @fields, $_->fields foreach $me->unique_constraints;
+ return wantarray ? @fields : \@fields;
+}
+
+# ----------------------------------------------------------------------
+sub unique_constraints {
+ my $me = shift;
+ my @cons = grep { $_->type eq UNIQUE } $me->get_constraints;
+ return wantarray ? @cons : \@cons;
+}
+
+# ----------------------------------------------------------------------
+sub fkey_constraints {
+ my $me = shift;
+ my @cons = grep { $_->type eq FOREIGN_KEY } $me->get_constraints;
+ return wantarray ? @cons : \@cons;
+}
+
+# ----------------------------------------------------------------------
sub DESTROY {
my $self = shift;
undef $self->{'schema'}; # destroy cyclical reference
=pod
-=head1 AUTHOR
+=head1 AUTHORS
-Ken Y. Clark E<lt>kclark@cpan.orgE<gt>
+Ken Youens-Clark E<lt>kclark@cpan.orgE<gt>,
+Allen Day E<lt>allenday@ucla.eduE<gt>.
=cut