Dropping initial tables after pivot

Fluxlang
1

Hi all,

I am trying to get started with Flux (coming from influxQL), and having a hard time translating this simple influxQL query:

SELECT sum("east") - sum(eait) as "Soutirage / Injection", sum("pv") as "Prod PV" FROM $iota.autogen.energy WHERE  $timeFilter GROUP BY time($kWh_interval,$tz_offset) fill(null)

Going through the docs I cam up with the following: map() generates the table I need in Grafana, however the initial 3 tables are still in the stream… is there any way I can get rid of them?
Setting _measurement = pv_summary and calling filter at the end is the best I could think of, but that seems to have no effect (I am using Explore Data in the Influxdb Web UI, Table view):

from(bucket: "iota-autogen")
  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)
  |> filter(fn: (r) => r["_measurement"] == "energy")
  |> filter(fn: (r) => r["_field"] == "pv" or r["_field"] == "east" or r["_field"] == "eait")
  |> pivot(rowKey: ["_time"], columnKey: ["_field"], valueColumn: "_value")
  |> map(fn: (r) => ({_time: r._time, _measurement: "pv_summary", soutirage_injection: r.east - r.eait, pv: r.pv}))
  |> filter(fn: (r) => r["_measurement"] == "pv_summary")

As a side question, is there really no way to obtain the pivoted table directly without having to call pivot, knowing that those fields all come from a single InfluxDB data point?

2

Solved, silly UI mistake, when defining multiple queries in the Web UI “Submit” returns the results from all queries, not just the one being edited.

This is what I eventually came up with, I will need a second query for the “pv” since apparently a) aggregateWindow can only aggregate one column and b) the only “unpivot” I find is experimental

from(bucket: “iota-autogen”)
|> range(start: v.timeRangeStart, stop: v.timeRangeStop)
|> filter(fn: (r) => r[“_measurement”] == “energy”)
|> filter(fn: (r) => r[“_field”] == “pv” or r[“_field”] == “east” or r[“_field”] == “eait”)
|> drop(columns: [“device”])
|> schema.fieldsAsCols()
|> map(fn: (r) => ({r with soutirage_injection: r.east - r.eait}))
|> drop(columns: [“eait”, “east”])
|> aggregateWindow(every: ${kWh_interval}, fn: sum, column: “soutirage_injection” )

Comparing with InfluxQL I can’t say I completely sold on Flux yet…