Hackathon at University of Oklahoma 2018 Team: @brucezheng, @joeysapp
vquery is a web-based visualization of a dummy insurance data set. We provie a straightforward UX to query this data. The height of each line is determined by the amount of insurance policies that have been sold in the area.
Written and developed using primarily Javascript, vquery utilizes multiple frameworks such as node, express, p5.js and socket.io. These allowed us to develop for multiple platforms while having quick turn-over time.
A user visiting vquery is greeted by a projection of our globe with a panel of search options. The dataflow is as follows:
1. jquery: specify options (optional)
2. jquery: press go!
3. javascript: client-side socket emit
4. node: server-side on-receive send db response
5. javascript: client-side on-receive store db response
* jsonify lat/long
* lat/long -> x,y,z
* calculate normals, heights of line
6. p5.js: draws globe and current data
Upon querying your local browser instance, the options are sent from the client to our node.js instance running in the background using socket.io.
After this our query is handed off to a relational database implemented using an npm package fast-csv. The result is sent via socket.io back to the client and stored. This data is then translated to cartesian x, y, z coordinates.
We had planned to implement a fully-functional location system (i.e. search based on area) using geojson but did not finish this. The location form uses an autocomplete form autocmplete.