Have Your Cake and Eat it Too: Considerations for Storing Layers
Regardless of whether you’re building a web map, RESTful service, desktop GUI, or ETL tool, your GIS application will most likely be storing and retrieving data from GIS layers. These layers will come in two flavors: layers you curated and created, and layers you didn’t create (possibly uploaded by your users).
After presenting an introductory demo on GDAL and PostGIS at the December 2015 GeoDevKC Meetup, a good deal of the discussion afterward centered around best-practices and considerations on how to store and retrieve your GIS data within your application. What if multiple users need to edit the geometries and attributes of their layers? What are the best options for performance if the layers are immutable? What type of validation should I perform on a user-submitted layer before importing it into my app?
At mySidewalk, we’ve built a layer-upload tool to allow clients to upload their GIS layers into our application in order to conduct visualization and analysis on them. In this talk, I will cover the key considerations when building an application that allows users to upload their arbitrary GIS layers into your application. We will focus on tools written for the C-family of programming languages, specifically GDAL/OGR, PostGIS, and postgres Foreign Data Wrappers.