Hi,
I have a recurring issue/request from my team where they need 15min averages of tabular data that is stored in InfluxDB. Currently, I am exporting the 15-min averages with a Flux query, and then pivoting the table in Excel to put it in proper tabular form, e.g.
| Timestamp | Field A | Field B | Field C |
|---|---|---|---|
| 2024-06-15 8:00:00 | 47.6 | true | 4 |
| 2024-06-15 8:15:00 | 49.5 | true | 5 |
| 2024-06-15 8:30:00 | 48.2 | true | 5 |
| 2024-06-15 8:45:00 | 49.2 | false | 6 |
| 2024-06-15 9:00:00 | 51.3 | true | 5 |
The query that should work is as follows:
from(bucket: "bucket")
|> range(start: v.timeRangeStart, stop: v.timeRangeStop)
|> filter(fn: (r) => r["_measurement"] == "measurement")
|> aggregateWindow(every: 15m, fn: mean, createEmpty: true)
|> pivot(rowKey: ["_time"], columnKey: ["_field"], valueColumn: "_value")
|> drop(columns: ["_measurement","_start","_stop"])
However, I am getting the following error:
unsupported input type for mean aggregate: boolean
It appears that, since some of the data is not in numerical form (float/int), the query breaks down completely when attempting to pipe the results from the aggregate into a pivot table.
Is there a way I can skip, or ignore, the errors (e.g. string and bool columns) in flux, so that I can export a CSV in the correct format, or am I doomed to always need Excel to finish the job for me?
Thanks,
Francis
