This is a command-line php script which creates UK Easting & Northing values from the lat long values. I plan to run it once a day on data.southampton.ac.uk — I know that the big local taxi firm uses Easting and Northing in their database so it might be interesting to see if I can generate a report for them which benefits our members by giving the taxi services better data.
The libraries it uses are my sparqllib (LGPL) and phpcoord-2.3 (GPL)
You’ll need the latest github version of sparqllib, if you want to include the “cgiParams” line, which tells 4store to give all the results — we’ve got more than 2000 lat/long pairs, and by default we’ve set 4store to set a limit of 2000 results on queries.
As this script uses a GPL library (not LGPL), I guess that makes it GPL, although it’s just a slip of script, I’m sure other UK open data services might want to consider something similar.
#!/usr/bin/php <?php require_once( "/var/wwwsites/tools/phpcoord/phpcoord-2.3.php" ); require_once( "/var/wwwsites/tools/PHP-SPARQL-Lib/sparqllib.php" ); $sparqlh = sparql_connect( "http://data-dev.ecs.soton.ac.uk:8002/sparql/" ); $sparqlh->cgiParams( "soft-limit=-1" ); # Limit lat and long to the same graph to try to minimise the # number of 'multiplying out' we get with multiple lat/long $result = $sparqlh->query( "SELECT DISTINCT * WHERE { GRAPH ?g { ?thing <http://www.w3.org/2003/01/geo/wgs84_pos#lat> ?lat . ?thing <http://www.w3.org/2003/01/geo/wgs84_pos#long> ?long . } }" ); if( !$result ) { print $db->errno() . ": " . $db->error(). "\n"; exit; } $fields = $result->field_array( $result ); while( $row = $result->fetch_array() ) { $ll2w = new LatLng($row["lat"],$row["long"]); $ll2w->WGS84ToOSGB36(); $os2w = $ll2w->toOSRef(); print "<".$row["thing"]."> <http://data.ordnancesurvey.co.uk/ontology/spatialrelations/easting> ""; print round($os2w->easting).""^^<http://www.w3.org/2001/XMLSchema#integer> .\n"; print "<".$row["thing"]."> <http://data.ordnancesurvey.co.uk/ontology/spatialrelations/northing> ""; print round($os2w->northing).""^^<http://www.w3.org/2001/XMLSchema#integer> .\n"; }
**UPDATE**
It’s been pointed out to me by Yang Yang (a PhD researcher in WAIS) that it’s a bit of an antipattern to publish data derived from other people’s datasets, and there’s no real value to me in providing Easting Northing data for bus-stops and a few places in wikipedia we import. I’ve made a rather elegant solution;
SELECT DISTINCT * WHERE { ?dataset <http://rdfs.org/ns/void#dataDump> ?graph . ?dataset <http://purl.org/dc/terms/publisher> <http://id.southampton.ac.uk/> . ?dataset a <http://purl.org/openorg/AuthoritativeDataset> . GRAPH ?graph { ?thing <http://www.w3.org/2003/01/geo/wgs84_pos#lat> ?lat . ?thing <http://www.w3.org/2003/01/geo/wgs84_pos#long> ?long . } }
This now only finds lat/long pairs in graphs which are listed as published by the University of Southampton AND Authoritative. That works pretty well.
One Response
Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.
Continuing the Discussion