@hippo_2 - thank you so much for the detailed response (and apologies for the delay in mine)!
I was able to get the forecast data into openHAB and then have it pop the data into InfluxDB, but I was having a real rough time trying to deal with the data structures and manipulating timestamps within openHAB so i ended up employing the excellent Solcast API python library to do all the heavy lifting for our prediction models outside of the openHAB environment. I also bring some values directly to openHAB for plotting and some rules-based behaviour with a separate API call.
As an aside, it is possible to do a direct HTTP request using something along the lines of:
Number Solcast005 "Value: [%.1f W]" { http="<[solcast.url=https://api.solcast.com.au/pv_power/forecasts?longitude=yourlongitude&latitude=yourlatitude&capacity=14000&tilt=17&azimuth=-174&install_date=20160901&api_key=yourapikey&format=json
:60000:JSONPATH($.forecasts.[0].pv_estimate)]" }
Is there any value to doing the HTTP cache method? The openHAB HTTP binding documentation just says that it is possible to do a HTTP cache or a direct HTTP API call, and doesn’t make recommendations for one way or the other. My reasoning is doing a direct API call every 30 mins or so achieves the same ends as refreshing a HTTP cache every 30 mins. I am certainly no expert, so it would be great to know if you’ve selected the HTTP cache method for a specific reason that I’m missing (programming best-practices, flexibility, etc.).