Creating custom map regions

Note: In Looker 22.14, support for GeoJSON files was added to the map_layer parameter.

Looker includes several built-in map layers, but you can also define custom map regions to display geographical data. Suppose you have a table of average ages for three neighborhoods, shown below:

You might want to map Customer Neighborhood to a particular geographic region, so it can be visualized on a map. To accomplish this, follow the steps described on this page.

Prepare region data

To prepare the region data, you need a data file that contains the geographic shapes of each region and also contains metadata that associates the region with the data in the database.

Looker uses a format called TopoJSON to store this data in a compact way. TopoJSON files can be easily created from many common shapefile formats, or you can draw one yourself with online tools, as detailed in the Converting shapefiles to TopoJSON Community post.

For the purposes of this example, there would already be a TopoJSON file prepared that provides this mapping. Each region in the TopoJSON file has a property called neighborhood that matches the values of the Customer Neighborhood field, which is defined in the LookML project as a dimension called neighborhood.

Upload region data

To import the region data into Looker, you can drag and drop the TopoJSON file into the File Browser section of the project. This data will be committed alongside the LookML code and updated like any other code in a LookML project.

Create the map layer

Now that you have the region data in the project, you need to create a map layer in the LookML model.

You can add a definition in the model file using the map_layer parameter:

map_layer: my_neighborhood_layer {
  file: "neighborhoods.topojson"
  property_key: "neighborhood"
}

The file parameter references the name of the file in the project that contains the region data, and the property_key is the property of the dataset that you want to expose within Looker. If you're not sure what the key is, simply omit the property key and Looker will attempt to select the appropriate key.

Tip: You can also use a TopoJSON file that's hosted elsewhere online, by specifying url instead of file in the layer definition:

map_layer: my_neighborhood_layer {
  url: "https://raw.githubusercontent.com/cooluser/JSON_Stuff/master/Neighborhoods.topoJSON"
  property_key: "neighborhood"
}

Associate the map layer with data

The only thing left to do is to associate the dimension neighborhood in the view with the newly created map layer by applying the map_layer_name parameter.

dimension: neighborhood {
  sql: ${TABLE}.neighborhood ;;
  map_layer_name: my_neighborhood_layer
}
  

This tells Looker that the values of this dimension are associated with the property_key exposed by the map layer. It also lets Looker know that it's possible to display this data on a map.

Viewing the map

The queries that use the neighborhood dimension can now be displayed on a map from the Explore page: