Tuning the documentation of calculated field function

This commit is contained in:
Andrii Shvaika 2025-03-10 16:09:42 +02:00
parent fd0e7a543e
commit 07d81e58b1

View File

@ -1,287 +1,139 @@
#### Calculated field TBEL script function
## Calculated Field TBEL Script Function
The **calculate()** function is a user-defined script that allows you to perform custom calculations using [TBEL{:target="_blank"}](${siteBaseUrl}/docs${docPlatformPrefix}/user-guide/tbel/) on telemetry and attribute data.
It receives arguments configured in the calculated field setup and an additional `ctx` object, which provides access to all arguments.
The **calculate()** function is a user-defined script that enables custom calculations using [TBEL](\${siteBaseUrl}/docs\${docPlatformPrefix}/user-guide/tbel/) on telemetry and attribute data.
It receives arguments configured in the calculated field setup, along with an additional `ctx` object that provides access to all arguments.
##### Function signature
### Function Signature
```javascript
function calculate(ctx, arg1, arg2, ...): object | object[]
function calculate(ctx, arg1, arg2, ...): object | object[]
```
##### Argument representation in the script
### Supported Arguments
Before describing how arguments are passed to the function, let's define how different argument types are **represented** inside the script.
There are three types of arguments supported in the calculated field configuration:
There are two types of arguments that can be used in the function:
#### Attribute and Latest Telemetry Arguments
* single value arguments - represent the latest telemetry data or attribute.
```json
{
"altitude": {
"ts": 1740644636669,
"value": 1034
}
}
```
These arguments are single values and may be of type: boolean, int64 (long), double, string, or JSON.
* when accessed via `ctx.args`, they remain objects:
Attribute and Latest telemetry are single value arguments that may be one of: boolean, int64 (long), double, string and JSON.
```javascript
var altitudeTimestamp = ctx.args.altitude.ts;
var altitudeValue = ctx.args.altitude.value;
```
**Example: Convert Temperature from Fahrenheit to Celsius**
* when accessed as a **function parameter**, only the value is passed:
```javascript
var temperatureC = (temperatureF - 32) / 1.8;
return {
"temperatureC": toFixed(temperatureC, 2)
}
```
```javascript
function calculate(ctx, altitude/*(single value argument)*/, temperature/*(time series rolling argument)*/) {
// altitude = 1035
Alternatively, using `ctx` to access the argument as an object:
```json
{
"temperatureF": {
"ts": 1740644636669,
"value": 36.6
}
```
}
```
* time series rolling arguments - contain historical data within a defined time window.
```json
{
"temperature": {
"timeWindow": {
"startTs": 1740643762896,
"endTs": 1740644662896
},
"values": [
{ "ts": 1740644355935, "value": 72.32 },
{ "ts": 1740644365935, "value": 72.86 },
{ "ts": 1740644375935, "value": 73.58 },
{ "ts": 1740644385935, "value": "NaN" }
]
}
}
```
You may notice that the object includes both the `value` of an argument and its timestamp as `ts`.
Let's modify the function that converts Fahrenheit to Celsius to also return the timestamp information:
* when accessed via `ctx.args`, they remain rolling argument objects:
```javascript
var temperatureC = (temperatureF - 32) / 1.8;
return {
"ts": ctx.args.temperatureF.ts,
"values": { "temperatureC": toFixed(temperatureC, 2) }
};
```
```javascript
var startOfInterval = temperature.timeWindow.startTs;
var firstTimestamp = temperature.values[0].ts;
var firstValue = temperature.values[0].value;
```
#### Time Series Rolling Arguments
* when accessed as a **function parameter**, they are passed as rolling arguments, retaining their structure:
These contain time series data within a defined time window. Example format:
```javascript
function calculate(ctx, altitude/*(single value argument)*/, temperature/*(time series rolling argument)*/) {
var avgTemp = temperature.mean(); // Use rolling argument functions
}
```
**Built-in methods for rolling arguments**
Time series rolling arguments provide built-in functions for calculations.
These functions accept an optional `ignoreNaN` boolean parameter, which controls how NaN values are handled.
Each function has two function signatures:
* **Without parameters:** `method()` → called **without parameters** and defaults to `ignoreNaN = true`, meaning NaN values are ignored.
* **With an explicit parameter:** `method(boolean ignoreNaN)` → called with a boolean `ignoreNaN` parameter:
* `true` → ignores NaN values (default behavior).
* `false` → includes NaN values in calculations.
| Method | Default Behavior (`ignoreNaN = true`) | Alternative (`ignoreNaN = false`) |
|------------|--------------------------------------------------|---------------------------------------------|
| `max()` | Returns the highest value, ignoring NaN values. | Returns NaN if any NaN values exist. |
| `min()` | Returns the lowest value, ignoring NaN values. | Returns NaN if any NaN values exist. |
| `mean()` | Computes the average value, ignoring NaN values. | Returns NaN if any NaN values exist. |
| `std()` | Calculates the standard deviation, ignoring NaN. | Returns NaN if any NaN values exist. |
| `median()` | Returns the median value, ignoring NaN values. | Returns NaN if any NaN values exist. |
| `count()` | Counts only valid (non-NaN) values. | Counts all values, including NaN. |
| `last()` | Returns the most recent non-NaN value. | Returns the last value, even if it is NaN. |
| `first()` | Returns the oldest non-NaN value. | Returns the first value, even if it is NaN. |
| `sum()` | Computes the total sum, ignoring NaN values. | Returns NaN if any NaN values exist. |
The following calculations are executed over the provided above arguments:
**Usage: default (`ignoreNaN = true`)**
```javascript
var avgTemp = temperature.mean();
var tempMax = temperature.max();
var valueCount = temperature.count();
```
**Output:**
```json
{
"avgTemp": 72.92,
"tempMax": 73.58,
"valueCount": 3
}
```
**Usage: explicit (`ignoreNaN = false`)**
```javascript
var avgTemp = temperature.mean(false); // Returns NaN if any NaN values exist
var tempMax = temperature.max(false); // Returns NaN if any NaN values exist
var valueCount = temperature.count(false); // Counts all values, including NaN
```
**Output:**
```json
{
"avgTemp": "NaN",
"tempMax": "NaN",
"valueCount": 4
}
```
Time series rolling arguments can be **merged** to align timestamps across multiple datasets.
| Method | Description | Parameters | Returns |
|:-----------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------|
| `merge(other, settings)` | Merges the current rolling argument with another rolling argument by aligning timestamps and filling missing values with the previous available value. | <ul><li>`other` (another rolling argument)</li><li>`settings` (optional) - configuration object, supports:<ul><br/><li>`ignoreNaN` (boolean, default true) - controls whether NaN values should be ignored.</li><li>`timeWindow` (object, default {}) - defines a custom time window for filtering merged values.</li></ul></li></ul>| Merged object with timeWindow and aligned values. |
| `mergeAll(others, settings)` | Merges the current rolling argument with multiple rolling arguments by aligning timestamps and filling missing values with the previous available value. | <ul><li>`others` (array of rolling arguments)</li><li>`settings` (optional) - configuration object, supports:<ul><br/><li>`ignoreNaN` (boolean, default true) - controls whether NaN values should be ignored.</li><li>`timeWindow` (object, default {}) - defines a custom time window for filtering merged values.</li></ul></li></ul>| Merged object with timeWindow and aligned values.|
**Example arguments:**
```json
{
"humidity": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741356882759,
"value": 43
}, {
"ts": 1741356918779,
"value": 46
}]
```json
{
"temperature": {
"timeWindow": {
"startTs": 1740643762896,
"endTs": 1740644662896
},
"pressure": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741357047945,
"value": 1023
}, {
"ts": 1741357056144,
"value": 1026
}, {
"ts": 1741357147391,
"value": 1025
}]
},
"temperature": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741356874943,
"value": 76
}, {
"ts": 1741357063689,
"value": 77
}]
}
"values": [
{ "ts": 1740644350000, "value": 72.32 },
{ "ts": 1740644360000, "value": 72.86 },
{ "ts": 1740644370000, "value": 73.58 },
{ "ts": 1740644380000, "value": "NaN" }
]
}
```
}
```
**Usage:**
The values are always converted to type `double`, and `NaN` is used when conversion fails. One may use `isNaN(double): boolean` function to check that the value is a valid number.
```javascript
var mergedData = temperature.merge(humidity, { ignoreNaN: false });
```
**Example: Accessing time series rolling argument data**
**Output:**
```json
{
"mergedData": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741356874943,
"values": [76.0, "NaN"]
}, {
"ts": 1741356882759,
"values": [76.0, 43.0]
}, {
"ts": 1741356918779,
"values": [76.0, 46.0]
}, {
"ts": 1741357063689,
"values": [77.0, 46.0]
}]
}
```javascript
var startOfInterval = temperature.timeWindow.startTs;
var endOfInterval = temperature.timeWindow.endTs;
var firstItem = temperature.values[0];
var firstItemTs = firstItem.ts;
var firstItemValue = firstItem.value;
var sum = 0.0;
// iterate through all values and calculate the sum using foreach:
foreach(t: temperature) {
if(!isNaN(t.value)) { // check that the value is a valid number;
sum += t.value;
}
```
}
// iterate through all values and calculate the sum using for loop:
sum = 0.0;
for(var i = 0; i < temperature.values.size; i++) {
sum += temperature.values[i].value;
}
// use built-in function to calculate the sum
sum = t.sum();
```
**Usage:**
##### Built-in Methods for Rolling Arguments
```javascript
var mergedData = temperature.mergeAll([humidity, pressure], { ignoreNaN: true });
```
Time series rolling arguments support built-in functions for calculations. These functions accept an optional `ignoreNaN` boolean parameter.
**Output:**
| Method | Default Behavior (`ignoreNaN = true`) | Alternative (`ignoreNaN = false`) |
|------------|-----------------------------------------------------|---------------------------------------------|
| `max()` | Returns the highest value, ignoring NaN values. | Returns NaN if any NaN values exist. |
| `min()` | Returns the lowest value, ignoring NaN values. | Returns NaN if any NaN values exist. |
| `mean()` | Computes the average value, ignoring NaN values. | Returns NaN if any NaN values exist. |
| `std()` | Calculates the standard deviation, ignoring NaN. | Returns NaN if any NaN values exist. |
| `median()` | Returns the median value, ignoring NaN values. | Returns NaN if any NaN values exist. |
| `count()` | Counts values, ignoring NaN values. | Counts all values, including NaN. |
| `last()` | Returns the most recent value, skipping NaN values. | Returns the last value, even if it is NaN. |
| `first()` | Returns the oldest value, skipping NaN values. | Returns the first value, even if it is NaN. |
| `sum()` | Computes the total sum, ignoring NaN values. | Returns NaN if any NaN values exist. |
```json
{
"mergedData": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741357047945,
"values": [76.0, 46.0, 1023.0]
}, {
"ts": 1741357056144,
"values": [76.0, 46.0, 1026.0]
}, {
"ts": 1741357063689,
"values": [77.0, 46.0, 1026.0]
}, {
"ts": 1741357147391,
"values": [77.0, 46.0, 1025.0]
}]
}
}
```
##### Function arguments
Usage example:
* `ctx` - context object that contains all provided arguments, in the representations described above.
```javascript
var avgTemp = temperature.mean(); // Returns 72.92
var tempMax = temperature.max(); // Returns 73.58
var valueCount = temperature.count(); // Returns 3
Accessing arguments via `ctx`:
var avgTempNaN = temperature.mean(false); // Returns NaN
var tempMaxNaN = temperature.max(false); // Returns NaN
var valueCountNaN = temperature.count(false); // Returns 4
```
```javascript
var altitude = ctx.args.altitude; // single value argument
var temperature = ctx.args.temperature; // time series rolling argument
```
* `arg1, arg2, ...` - user-defined arguments configured in the calculated field setup.
How they are passed depends on their type:
* **single value arguments** are passed as raw values **(e.g., 22.5, "ON")**.
* **time series rolling arguments** are passed as objects (containing multiple values).
##### Example: Air Density Calculation
This function calculates air density using `altitude` (single value argument) and `temperature` (time series rolling argument).
This function calculates air density using `altitude` (single value) and `temperature` (time series rolling argument).
```javascript
function calculate(ctx, altitude, temperature) {
var avgTemperature = temperature.mean(); // Get average temperature
var temperatureK = (avgTemperature - 32) * (5 / 9) + 273.15; // Convert Fahrenheit to Kelvin
// Estimate air pressure based on altitude
var pressure = 101325 * Math.pow((1 - 2.25577e-5 * altitude), 5.25588);
@ -294,42 +146,255 @@ function calculate(ctx, altitude, temperature) {
}
```
* `altitude` is a single value, passed as number.
* `temperature` is a rolling argument, retaining its full structure.
##### Merging Time Series Arguments
##### Function return format
Time series rolling arguments can be **merged** to align timestamps across multiple datasets.
The script should return a JSON object formatted according to the [ThingsBoard Telemetry Upload API](${siteBaseUrl}/docs${docPlatformPrefix}/user-guide/telemetry/#time-series-data-upload-api/).
The return value must match one of the supported telemetry upload formats.
| Method | Description | Parameters | Returns |
|:-----------------------------|:--------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------|
| `merge(other, settings)` | Merges with another rolling argument. Aligns timestamps and filling missing values with the previous available value. | <ul><li>`other` (another rolling argument)</li><li>`settings` (optional) - configuration object, supports:<ul><br/><li>`ignoreNaN` (boolean, default true) - controls whether NaN values should be ignored.</li><li>`timeWindow` (object, default {}) - defines a custom time window for filtering merged values.</li></ul></li></ul>| Merged object with timeWindow and aligned values. |
| `mergeAll(others, settings)` | Merges with multiple rolling arguments. Aligns timestamps and filling missing values with the previous available value. | <ul><li>`others` (array of rolling arguments)</li><li>`settings` (optional) - configuration object, supports:<ul><br/><li>`ignoreNaN` (boolean, default true) - controls whether NaN values should be ignored.</li><li>`timeWindow` (object, default {}) - defines a custom time window for filtering merged values.</li></ul></li></ul>| Merged object with timeWindow and aligned values.|
The script must return data in a format compatible with ThingsBoards APIs.
The correct return format depends on the calculated field output configuration:
Assuming the following arguments and their values:
* if latest telemetry is used for output, the function must return data according to the [Telemetry Upload API](${siteBaseUrl}/docs${docPlatformPrefix}/reference/mqtt-api/#attributes-api/).
* without timestamps
```json
{
"airDensity": 1.06,
"someKey": "value"
```json
{
"humidity": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741356882759,
"value": 43
}, {
"ts": 1741356918779,
"value": 46
}]
},
"pressure": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741357047945,
"value": 1023
}, {
"ts": 1741357056144,
"value": 1026
}, {
"ts": 1741357147391,
"value": 1025
}]
},
"temperature": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741356874943,
"value": 76
}, {
"ts": 1741357063689,
"value": 77
}]
}
```
* with a timestamp:
```json
}
```
**Usage:**
```javascript
var mergedData = temperature.merge(humidity, { ignoreNaN: false });
```
**Output:**
```json
{
"mergedData": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741356874943,
"values": [76.0, "NaN"]
}, {
"ts": 1741356882759,
"values": [76.0, 43.0]
}, {
"ts": 1741356918779,
"values": [76.0, 46.0]
}, {
"ts": 1741357063689,
"values": [77.0, 46.0]
}]
}
}
```
**Usage:**
```javascript
var mergedData = temperature.mergeAll([humidity, pressure], { ignoreNaN: true });
```
**Output:**
```json
{
"mergedData": {
"timeWindow": {
"startTs": 1741356332086,
"endTs": 1741357232086
},
"values": [{
"ts": 1741357047945,
"values": [76.0, 46.0, 1023.0]
}, {
"ts": 1741357056144,
"values": [76.0, 46.0, 1026.0]
}, {
"ts": 1741357063689,
"values": [77.0, 46.0, 1026.0]
}, {
"ts": 1741357147391,
"values": [77.0, 46.0, 1025.0]
}]
}
}
```
**Example: Freezer temperature analysis**
This function merges `temperature` data with the fridge's `defrost` status. It then analyzes the merged data to identify instances where the fridge is not in defrost mode, yet the internal air temperature is too high ( > -5° C).
```javascript
function calculate(ctx, temperature, defrost) {
var merged = temperature.merge(defrost);
var result = [];
foreach(item: merged) {
if (item.v1 > -5.0 && item.v2 == 0) {
result.add({
ts: item.ts,
values: {
issue: {
temperature: item.v1,
defrostState: false
}
}
});
}
}
return result;
}
```
The result is a list of issues that may be used to configure alarm rules:
```json
[{
"ts": 1741613833843,
"values": {
"issue": {
"temperature": -3.12,
"defrostState": false
}
}
}, {
"ts": 1741613923848,
"values": {
"issue": {
"temperature": -4.16,
"defrostState": false
}
}
}]
```
### Function return format
The return format depends on the output type configured in the calculated field settings (default: **Time Series**).
##### Time Series Output
The function must return a JSON object or array with or without a timestamp.
Examples below return 5 data points: airDensity (double), humidity (integer), hvacEnabled (boolean), hvacState (string) and configuration (JSON):
Without timestamp:
```json
{
"airDensity": 1.06,
"humidity": 70,
"hvacEnabled": true,
"hvacState": "IDLE",
"configuration": {
"someNumber": 42,
"someArray": [1,2,3],
"someNestedObject": {"key": "value"}
}
}
```
With timestamp:
```json
{
"ts": 1740644636669,
"values": {
"airDensity": 1.06,
"humidity": 70,
"hvacEnabled": true,
"hvacState": "IDLE",
"configuration": {
"someNumber": 42,
"someArray": [1,2,3],
"someNestedObject": {"key": "value"}
}
}
}
```
Array containing multiple timestamps and different values of the `airDensity` :
```json
[
{
"ts": 1740644636669,
"values": {
"airDensity": 1.06,
"someKey": "value"
}
}
```
* if attributes are used for output, the function must return data according to the [Attributes API](${siteBaseUrl}/docs${docPlatformPrefix}/user-guide/telemetry/#time-series-data-upload-api/).
```json
"airDensity": 1.06
}
},
{
"airDensity": 1.06,
"someKey": "value"
"ts": 1740644636670,
"values": {
"airDensity": 1.07
}
}
]
```
##### Attribute Output
The function must return a JSON object **without timestamp** information.
Example below return 5 data points: airDensity (double), humidity (integer), hvacEnabled (boolean), hvacState (string) and configuration (JSON):
```json
{
"airDensity": 1.06,
"humidity": 70,
"hvacEnabled": true,
"hvacState": "IDLE",
"configuration": {
"someNumber": 42,
"someArray": [1,2,3],
"someNestedObject": {"key": "value"}
}
```
}
```