[Mulgara-general] empty results from SPARQL queries
Mark Stubbeman
MStubbeman at algebraixdata.com
Fri Feb 3 15:21:01 UTC 2012
I'm new to SPARQL and to triple-store technology in general. I have a couple of example DBs - Mulgara being one of them - against which I am trying to write some simple Perl code to submit queries and extract responses. The following works against one of my DB's but not against Mulgara:
sub runQuery
{
use RDF::Query::Client;
# $webURL scrubbed of actual hostname but otherwise true to my actual code
my $webURL = "http://$somehostname:8080/sparql";
# an array of lines read from a file containing a SPARQL query
my @queryIn = @_;
my $funcQuery = join('', @queryIn);
print "Query\n-----\n\n$funcQuery\n";
my $query = RDF::Query::Client->new($funcQuery);
my $iterator = $query->execute($webURL);
print "Response\n------\n\n";
print Dumper($iterator);
while (my $row = $iterator->next)
{
print "$row\n";
}
}
With $webURL directed at another DB we get the following:
Executing: C:\sandboxes\cdbms\Hawk\Tests\SP2Bench\queries\q1.sparql
Query
-----
SELECT DISTINCT ?s WHERE { ?s ?p ?o . }
Response
------
$VAR1 = bless( {
'_peek' => [],
'_row' => undef,
'_count' => 0,
'_type' => 'bindings',
'_finished' => 0,
'_names' => [],
'_args' => {},
'sorted_by' => [],
'_stream' => sub { "DUMMY" },
'_open' => 0
}, 'RDF::Trine::Iterator::Bindings' );
{ s=<http://localhost/publications/articles/Journal1/1944/Article4> }
(note that the output has been truncated to just one line). However, with $webURL set to our Mulgara installation the same code generates the following:
Query
-----
SELECT DISTINCT ?s WHERE { ?s ?p ?o . }
Response
------
$VAR1 = bless( {
'_peek' => [],
'_row' => undef,
'_count' => 0,
'_type' => 'bindings',
'_finished' => 0,
'_names' => [
's'
],
'_args' => {},
'sorted_by' => [],
'_stream' => sub { "DUMMY" },
'_open' => 0
}, 'RDF::Trine::Iterator::Bindings' );
We get back an iterator (with slightly different properties) but no results. I know via the Mulgara webui that the data is there. That's how I load the data. I'm not sure what to use for Graph URI (several settings seem to work) but for the above exercise I used "<hostname>:8080/sparql" where <hostname> is the same as $somehostname from the sample code.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mulgara.org/pipermail/mulgara-general/attachments/20120203/d45d752e/attachment.html>
More information about the Mulgara-general
mailing list