Hosted Extractors - maximum transformation code length (2000)

Question

Hi,From my MQTT topic I receive sensor data as a timestamp and 25 values, where each value should gets transformed as a datapoint for its own timeseries. However, there is a limit on 2000 char for the transformation code AND a constraint that does not allow you to set up duplicate streams on the same topic: meaning you can only do one transformation per topic, but this function is limited to 2000 chars. I run into a problem now where, for this sample data:“{"timestamp":1701093489556,"values":[-0.035999998450279236,-0.10400000214576721,-0.09099999815225601,-0.08299999684095383,-0.09399999678134918,-0.10999999940395355,-0.10700000077486038,-0.10499999672174454,-0.11299999803304672,-0.12099999934434891,-0.13899999856948853,-0.1340000033378601,-0.14399999380111694,-0.1340000033378601,-0.15000000596046448,-0.15000000596046448,-0.17000000178813934,-0.16699999570846558,-0.15399999916553497,-0.16500000655651093,-0.15000000596046448,-0.10199999809265137,-0.11100000143051147,-0.1469999998807907,-0.12600000202655792]}”I use this template, but cannot include the all 25 values as this is just over 4000 char. Is there a way to get around this? e.g using a loop function to generate the template? “[From 0…{  "externalId": concat("mysystem.ExtEnvironment.mysensor.input9.", context.topic),  "timestamp": input.timestamp,  "value": input.values[8],  "type": "datapoint"},{  "externalId": concat("mysystem.ExtEnvironment.mysensor.input10.", context.topic),  "timestamp": input.timestamp,  "value": input.values[9],  "type": "datapoint"},{  "externalId": concat("mysystem.ExtEnvironment.mysensor,input11.", context.topic),  "timestamp": input.timestamp,  "value": input.values[10],  "type": "datapoint"},{  "externalId": concat("mysystem.ExtEnvironment.mysensor.input12.", context.topic),  "timestamp": input.timestamp,  "value": input.values[11],  "type": "datapoint"},{  "externalId": concat("mysystem.ExtEnvironment.mysensor..input13.", context.topic),  "timestamp": input.timestamp,  "value": input.values[12],  "type": "datapoint"}...up to 25]”

Nicolai · Accepted Answer

Hey, I couldn’t find a way to generate the index-list (perhaps someone else here knows how to do so), but if the value-list is always static then perhaps this will work?

zip(
	input.values, 
	[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24], 
    (value, index) => {
		"externalId": concat("mysystem.ExtEnvironment.mysensor.input",index,".",context.topic),
		"value":value,
		"timestamp": input.timestamp,
		"type": "datapoint"
	}
)

Einar Omang · Answer

The zip approach will work fine, but we can add a function to get the indices in an array, as you here have a clear use case of such an approach, perhaps `input.enumerate((value, index) => ...)`?

Sign up

Welcome to the HUB

Scanning file for viruses.

This file cannot be downloaded