We are developing a system to disseminate information about roadkills. The source data comes from three different databases. We have set up a FME project which pulls geojson from the source databases, joins them according too a set of business rules and then exports them to a front-end database.
We are currently exporting our data to a PostGIS database. FME does this very nicely. Now my colleague suggests we use Elasticsearch since this will make the presentation of our data in leaflet less complicated. Elasticsearch evidently indexes geojson on the fly.
We are using FME version 15.0. How can I write the data to Elasticsearch from FME? Which writer should I use?
Best Answer
GDAL/OGR supports Elasticsearch, which means that FME could support it reasonably easily.
I reckon you might like to hit up the FME people on Twitter and put in a request. Try Mark Ireland @FMEEvangelist or Dale Lutz @DaleAtSafe and Don Murray @DonAtSafe.
But anyway, you could use a FME script to prepare your data, then write it to a temporary file, and then use Python to write, either directly (tricky but solid), or using OGR (easy, less flexible), to Elasticsearch. I'd implement that, and then contact the FME guys and tell them to hurry up and implement an Elasticsearch writer!