Filter basend on condition

InfluxDB 2
1

Hi there,
I’m trying to filter data based on a value in a specific row:

whole (not correctly working) Query:

from(bucket: "OpenWeatherMap")
  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)
  |> filter(fn: (r) => r["_measurement"] == "weather")
  |> filter(fn: (r) => r["_field"] == "temperature")
  |> filter(fn: (r) => r["forecast"] == "6h")
  |> aggregateWindow(every: v.windowPeriod, fn: mean, createEmpty: false)

  |> filter(fn: (r) => 
    if r["condition_id"] == "800" then
      r["condition_id"] == "800"
    else if r["condition_id"] == "801" then
      r["condition_id"] == "801"
    else
      r["condition_id"] == "802"
      )

Note: The forecast = 6h should be forecast = *. Just using 6h because I have multiple datapoint to test.

Background:
I’m using a telegraf config to get the data from openweathermap api for the current location. Saved in bucket “OpenWeatherMap”
I want to get the current temperature to show in a grafana dashboard but sometimes openweathermap has multiple datapoint (condition_id) at the same time, that indicate cloud/rain status which beaks my dashboard.

What I’m trying:
Get the value that is available at the highest condition_id.
Let’s say condition_id is 800 (clear) and temperature value is 20, i want 20. But if there is 800 (clear) and 801 (clouds) i want to prioritize and only get the 801 value over the 800 value.

condition_id is predefined from 800 to 804. Value Priority should be 804 > 803 > .. > 800

Is this possible with a Query?

more on condition_id
https://openweathermap.org/weather-conditions

2

Hi @eventuallyLearnt and welcome to the InfluxData forum.

There is probably a somewhat messy way to do this in a Flux query (prioritize 804 > 803, etc.), but I think a much easier way to do this is either:

  1. use Node-RED instead of Telegraf to poll the openweathermap API, put your conditions with if/then, etc. (can be done visually or using javascript)
    OR
  2. Keep Telegraf in place and move to InfluxDB v3 with SQL and do it via a SQL function and Python script.