Skip to content

The json_emitter node

This node is used to produce data, that can be processed in a flow.

It can be used for debugging and testing/mocking purposes as well as produce default values for different kind of applications.

Starting with a given list of JSON strings (json structures in string format), this data can be emitted and/or modified on every emit in different ways:

  • determine, how data for output gets selected from the list of json inputs, see also select
  • modify the selected data with modifier lambda functions
  • output periodically with every, possibly adding a timing jitter, or have exactly 1 output with the one-shot parameter

experimental since 0.19.9: With the modify and modify_with parameters, certain fields in the json string(s) can be modified with every interval.

Examples

|json_emitter(
  ' {"condition": {"id": 0, "name": "OK"}, "condition_reason": ""} ',
  ' {"condition": {"id": 1, "name": "Warning"}, "condition_reason": "bad succer"} ',
  ' {"condition": {"id": 2, "name": "Error"}, "condition_reason": "something went really wrong!"} '
  )

Emit one of the three given json strings(selected randomly) every 5 seconds (default).

|json_emitter(
  ' {"condition": {"id": 0, "name": "OK"}, "condition_reason": ""} ',
  ' {"condition": {"id": 1, "name": "Warning"}, "condition_reason": "bad succer"} ',
  ' {"condition": {"id": 2, "name": "Error"}, "condition_reason": "something went really wrong!"} ' 
  )
.every(3s)
.select('seq')

Select one of the json strings in sequence every 3 seconds, starting with the first one. After the last json string has been selectect and outputted, start over with the first one again. Produces data-point items.

Using modifiers

|json_emitter(
  ' {"Current_Ph_A": 10.33, "Current_Ph_B": 10.53, "Current_Ph_C": 10.76} ' 
  )
.every(200ms)
.align(true)
.modify(
    'data.Current_Ph_A',
    'data.Current_Ph_B', 
    'data.Current_Ph_C'
     )
.modify_with(
    lambda: random_normal(), 
    lambda: "data.Current_Ph_A" + 0.45, 
    lambda: nth(random(4), [12.552, 44.2, 13.86, 16.22])
    )
.as('data')

Output the json data every 200ms, modifying it on every emit with the given lambda functions. Note the 'as' parameter here: It gives the output data a new root-object ('data') and the modifiers must target the 'rooted' fields.

Sample output:

{"ts":  1675147412200, "data":  
  {"Current_Ph_A": 1.6554763679823647, "Current_Ph_B": 2.105476367982365, "Current_Ph_C": 44.2}
}

Using __state in modifier

__state contains the data_point, that was emitted previously, we can implement a simple counter with that for example:

|json_emitter('{"field1": 1}')
.modify('field1')
.modify_with(lambda: "__state.field1" + 1)

Output is a data_point with an ever increasing value for field1.

Batch output and one_shot

def start_time = '2021-11-16T16:21:00.000Z'
|json_emitter(
  ' {"Current_Ph_A": 10.33, "Current_Ph_B": 10.53, "Current_Ph_C": 10.76} ',
  ' {"Current_Ph_A": 11.13, "Current_Ph_B": 10.74, "Current_Ph_C": 11.17} ',
  ' {"Current_Ph_A": 13.78, "Current_Ph_B": 11.02, "Current_Ph_C": 11.12} ',
  ' {"Current_Ph_A": 12.46, "Current_Ph_B": 10.98, "Current_Ph_C": 11.71} '
  )
.select('batch')
.one_shot(true)
.start_ts(start_time)

Output a data-batch item (using all the json items at once) exactly one time (one-shot), using start_time as timestamp. After the first output, the node will go to sleep for the rest of its life.

Parameters

Parameter Description Default
[node] json(string_list) list of json strings
every(duration) emit interval 5s
jitter(duration) max random value for time jitter added to every, adds time jitter to the values produced 0ms
align(bool) align the time to the every param, does not have an effect, if start_ts is given. false
select(string) entry select mode, possible values: rand or seq, batch 'rand'
start_ts(string or int) Timestamp to start from, instead of wall clock time (default). ISO8601 datetime string or millisecond timestamp allowed here. '2021-11-16T16:21:12.505Z'
modify(string_list) list of fields, that should be modified undefined
modify_with(lambda_list) list of lambda expressions, that perform the modification on the fields given with modify undefined
as(string) root object used for output items undefined
one_shot(bool) when set to true, the node gives exactly 1 output, before it goes to sleep false

Note : modify and modify_with must have the same number of parameters, if given.

select

value Description Resulting data-item
rand randomly selects one of the json structures data-point
seq selects the json entries in sequence starting from top data-point
batch selects all of the entries resulting in a batch of data-points data-batch