More on my Internet of Things project, GetFitY’all.
- I exposed 2 RESTful endpoints and just to prove that Azure “loves all, serves all”, these endpoints were implemented as a node.js app. In my previous “life” I was considering a few API proxy platforms, so I deployed an API proxy that hosts these endpoints. But then these endpoints could easily be deployed onto a simple and free Azure Web Site. Next candidate should be the Azure API management feature.
- These 2 endpoints perform these simple tasks:
- Mashup on demand – Let client apps consume a mash up of fitness activity data points from different target APIs based upon user ID and time period for a specific date. This has to be real-time and on demand although the mashup does take up quite a bit of computational time.
- Automated message pump – Generates a message pump that pulls fitness activity data points from different target APIs, and sends as AMQP messages asynchronously to Azure Event Hub. This is meant to decouple the processing of these messages from the message pump. The message pump endpoint could be fired up based upon a set schedule in Azure Scheduler.
- Publish these endpoints, it will return you valid JSON or XML depending on the HTTP Accept header.
- GET /v1/getfityall/mashup – no query strings to make things simpler based upon known parameters that return an activities mashup.
- POST /v1/getfityall/msgpump
Actual endpoint URLs to be updated later… this is the placeholder.. watch out…
- To further demonstrate that PowerQuery and PowerMap are good to consume data sources directly from API proxy endpoints, I’d created another spreadsheet, and this time it is for a mashup of how I got to Mobile Monday Sydney, a meetup I attended on the first Aug Monday evening, it was fun to get to know the community here. The skyscrapers are thinner now, and click the play button to see my routes.
