Just in case anyone is interested, here are two implementations using XML::Rules. The first one prints the reactions as it parses through the XML, the other builds a more simplified data structure.
use strict;
use warnings;
use XML::Rules;
use Data::Dumper;
my $parser = XML::Rules->new(
rules => [
substrate => sub {return '@substrates' => $_[1]->{name}},
product => sub {return '@products' => $_[1]->{name}},
reaction => sub {
print "$_[1]->{name} ($_[1]->{type}):\n";
print "\tSubstrates: ", join(", ", @{$_[1]->{substrates}})
+, "\n";
print "\tProducts: ", join(", ", @{$_[1]->{products}}),
+"\n\n";
return;
},
],
);
$parser->parse(\*DATA);
__DATA__
<root>
<reaction name="rn:R00710" type="reversible">
<substrate name="cpd:C00084"/>
<product name="cpd:C00033"/>
</reaction>
<reaction name="rn:R00014" type="irreversible">
<substrate name="cpd:C00068"/>
<substrate name="cpd:C00022"/>
<product name="cpd:C05125"/>
</reaction>
</root>
and
use strict;
use warnings;
use XML::Rules;
use Data::Dumper;
my $parser = XML::Rules->new(
rules => [
substrate => sub {return '@substrates' => $_[1]->{name}},
product => sub {return '@products' => $_[1]->{name}},
reaction => sub {
my $name = delete($_[1]->{name});
delete($_[1]->{_content});
return $name => $_[1];
},
root => 'pass no content',
],
);
my $result = $parser->parse(\*DATA);
print Dumper($result);
while (my ($reaction, $data) = each(%$result)) {
print "$reaction ($data->{type})\n";
print "\tSubstrates: ", join(", ", @{$data->{substrates}}), "\n";
print "\tProducts: ", join(", ", @{$data->{products}}), "\n\n";
}
__DATA__
<root>
<reaction name="rn:R00710" type="reversible">
<substrate name="cpd:C00084"/>
<product name="cpd:C00033"/>
</reaction>
<reaction name="rn:R00014" type="irreversible">
<substrate name="cpd:C00068"/>
<substrate name="cpd:C00022"/>
<product name="cpd:C05125"/>
</reaction>
</root>
The substrate and product rules could be rewriten like this:
'substrate,product' => sub {return '@'.$_[0].'s' => $_[1]->{name}},
|