Start using mongo timeseries

mooving all database relevant function to influx.js resp. mongo.js

aktual data now also readable from mongo

sensorspecials/noise.js

@@ -2,10 +2,12 @@
 // rxf 2023-03-05
 
 import {returnOnError} from "../utilities/reporterror.js";
-import { getActData, getAvgData, getLongAvg, fetchFromInflux, calcRange} from "../actions/getsensorData.js"
+import { getActData, getAvgData, getLongAvg, calcRange} from "../actions/getsensorData.js"
 import checkParams from "../utilities/checkparams.js";
 import {DateTime} from 'luxon'
 import { translate as trans } from '../routes/api.js'
+import * as influx from "../databases/influx.js"
+import * as mongo from "../databases/mongo.js"
 
 
 const setoptionfromtable = (opt,tabval) => {
@@ -429,61 +431,8 @@ const getAPIprops = (opt) => {
 }
 
 const getNoiseAVGData = async (opts) => {
-    let ret = {err: null, values: []}
-    let emptyValues = {n_AVG: -1}
-    let small =  '|> keep(columns: ["_time", "peakcount", "n_AVG"])'
-    if (opts.long) {
-        small = ''
-        emptyValues.n_sum = -1
-    }
-    let queryAVG = `
-import "math"
-threshold = ${opts.peak}
+    let ret = await influx.fetchNoiseAVGData(opts)
 
-data = from(bucket: "sensor_data")
-  |> range(${opts.start}, ${opts.stop})  
-  |> filter(fn: (r) => r["sid"] == "${opts.sensorid}")
-e10 = data 
-  |> filter(fn: (r) => r._field == "E10tel_eq") 
-  |> aggregateWindow(every: 1h, fn: mean, createEmpty: false)
-  |> map(fn:  (r) => ({r with _value: (10.0 * math.log10(x: r._value))}))
-  |> keep(columns: ["_time","_field","_value"]) 
-  |> pivot(rowKey:["_time"], columnKey: ["_field"], valueColumn: "_value")
-  |> rename(columns: {"E10tel_eq" : "n_AVG"})
-ecnt = data
-  |> filter(fn: (r) => r._field == "E10tel_eq") 
-  |> aggregateWindow(every: 1h, fn: count, createEmpty: false)
-  |> keep(columns: ["_time","_field","_value"]) 
-  |> pivot(rowKey:["_time"], columnKey: ["_field"], valueColumn: "_value")
-  |> rename(columns: {"E10tel_eq" : "count"})
-esum = data
-  |> filter(fn: (r) => r._field == "E10tel_eq") 
-  |> aggregateWindow(every: 1h, fn: sum, createEmpty: false)
-  |> keep(columns: ["_time","_field","_value"]) 
-  |> pivot(rowKey:["_time"], columnKey: ["_field"], valueColumn: "_value")
-  |> rename(columns: {"E10tel_eq" : "n_sum"})
-peak = data
-  |> filter(fn: (r) => r._field == "noise_LA_max") 
-  |> aggregateWindow(
-        every: 1h,
-        fn: (column, tables=<-) => tables
-            |> reduce(
-                identity: {peakcount: 0.0},
-                fn: (r, accumulator) => ({
-                    peakcount: if r._value >= threshold then
-                        accumulator.peakcount + 1.0
-                    else
-                        accumulator.peakcount + 0.0,
-                }),
-            ),
-      )
-  |> keep(columns: ["_time","peakcount"]) 
-  part1 = join( tables: {e10: e10, ecnt: ecnt}, on: ["_time"])
-  part2 = join( tables: {esum: esum, peak: peak}, on: ["_time"])
-  join( tables: {P1: part1, P2: part2}, on: ["_time"])
-  ${small}
-`
-    ret = await fetchFromInflux(ret, queryAVG)
     if(ret.err) {
         return returnOnError(ret, ret.err, getNoiseAVGData.name)
     }
@@ -496,6 +445,10 @@ peak = data
     // hour in an element in docs becomes the index into the new array (for every new day this
     // index will be  incremented by 24). Missing values are marked by: {n_sum=-1, n_AVG=-1}.
     // For havg add the missed hours to the arry
+    let emptyValues = {n_AVG: -1}
+    if (opts.long) {
+        emptyValues.n_sum = -1
+    }
     const misshours = DateTime.fromISO(ret.values[0].datetime).get('hour')
     let hoursArr = new Array(opts.span * 24 + misshours);    // generate new array
     hoursArr.fill(emptyValues)                              // fill array with 'empty' values