Instructions¶

Conditions¶

Conditionally execute instructions

Parameters :

No parameter

Children nodes :
Every conditions will be represented as parallel branches on your graph, each beginning with a blue box allowing you to edit the condition itself, followed by the instruction that you will add for that specific condition.

A default branch is automatically added to configure what instructions to execute when no other condition matches.

More details on Conditions syntax

Repeat¶

Execute a list of instructions N times, or for each record from an array or object variable

Parameters :

on : An array or object variable name to loop on. Each record will be available inside an item variable, and current index will be inside an $index variable.
until : A number of times to repeat this loop. Each iteration number will be stored inside an item variable

If on and until parameters are used together, only the first until items from on will be iterated.

Children nodes :
When adding a Repeat instruction, a new graph branch will appear, allowing you to create the instructions that will be repeated.

Break¶

Stops currently executing automation

Parameters :

scope : If set to all, breaks all parents automations

All¶

Execute the given instructions in parallel

Coming soon

Set¶

Set a new or existing variable

Parameters :

name : Variable name
value : Variable value (might be a JSON object, a string, a number, ...)
type : replace to replace target variable with given value, merge to try merging objects or arrays, or push to push to an array. replace by default

When setting object fields, parent objects are created on-the-fly :

- set:
    name: some.house.field
    value: ok

Here, some and some.house are automatically created :

  "some": {
    "house": {
      "field": "ok
    }
  }

It is also possible to create lists and automatically add items to their end with a variable name suffix [] :

- set:
    name: session.names
    type: push
    value: Mickael
# or :
- set:
    name: session.names[]
    value: Mickael

Each time this set is executed, Mickael is appended to the session.names list variable.

Same feature with type: merge option :

- set:
    name: session.names
    type: merge
    value: Mickael

When both previous value and value are arrays, type: merge concatenate value at the end of the previous value :

- set:
    name: myArray
    value:
      - one
- set:
    name: myArray
    type: merge
    value:
      - two
      - three
# {{myArray}} = ['one', 'two', 'three']

When both previous value and value are objects, type: merge merge them together :

- set:
    name: 'myObject'
    value:
      firstName: Martin
- set:
    name: 'myObject'
    type: merge
    value:
      age: 25 
# {{myObject}} = { "firstName": "Martin", "age": 25 }

Delete¶

Delete a variable

Parameters :

name : Variable name

Fetch¶

Sends an HTTP request to call external web services

Parameters :

URL : Target URL
method : Request method (get | post | put | patch | delete)
headers : Request headers
query : Query string (as an object)
body : Request body (might be a JSON object, a string, a number, ...)
multipart : List of field definitions for multipart/form-data requests
emitErrors : Boolean enabling or disabling error events upon 4xx or 5xx responses. Enabled by default.
stream : For streamed SSE responses, emit data chunks as they are received
outputMode : Choose between 'body' | 'detailed_response' | 'data_url'. Defaults to 'body'
output : Name of the variable that will store the response body

When receiving 4xx or 5xx HTTP errors, a native event runtime.fetch.failed is automatically emitted, including both request & response contents.

If Content-Type header is set to 'application/x-www-form-urlencoded', the body will be automatically transformed into an urlencoded body.

multipart

- fetch:
    url: ..
    multipart:
      - fieldname: file
        value: someBase64EncodedFile | someBufferArray
        filename: filename.png # Required if value is a file
        contentType: image/png # Optional
      - fieldname: metadata
        value: some random metadata

Handling SSE
By default, HTTP responses indicating a content-type: text/event-stream header will force their responseMode to detailed_response, and their {{output.body}} will be a stream that can be read in real time from the calling automation :

- fetch:
    url: "some SSE url (i.e openai /chat-completion endpoint)"
    output: output
- repeat:
    'on': '{{output.body}}'
    do:
      - set:
          name: test[]
          value: '{{item}}'

Here, the repeat instruction will block until all chunks have been processed & the HTTP socket is closed.
Each {{item}} will look like this :

{
  "chunk": {
    "data": [
      "data1",
      "data2"
    ]
  }
}

Here, data1 and data2 correspond to 2 individual chunks written by the remote server, which can be grouped together if they were received at the same time.
If these data strings are valid JSON, they will be automatically parsed into objects.

Alternatively, we can emit received chunks so they can be processed asynchronously & concurrently from other automations :

- fetch:
    url: ..
    stream:
      event: chunk
      payload:
        foo: bar

Each chunk will be emitted like this :

{
  "type": "chunk",
  "payload": {
    "chunk": {
      "data": [
        "data1",
        "data2"
      ]
    },
    "additionalPayload": {
      "foo": "bar"
    }
  }
}

stream option also accepts target and options fields from emit instruction.

Emit¶

Emit a new event.
Events size must not exceed 100Ko, larger events are blocked and an error event is emitted.

Parameters :

event : Event name
payload : JSON object that will be sent as the event payload
target : An optional object specifying this event target
autocomplete : An optional object helping to find all events that can be computed when you set variables in your event value
private : Optionnal Boolean. Exclude this event name from autocomplete values. Set it to false if you don't want your event is known from out of your application.
options : optional object

target parameters :

userId : target user id
sessionId : target session id
userTopic : target userTopic

Event targets are automatically granted read access to the event

options parameters :

persist : boolean enabling or disabling events persistence. Enabled by default.

autocomplete parameter :

When a field use an autocomplete widget and ask for events:emit or events:listen data source, they will be extracted from current workspace and all its installed apps instructions. If the event is a simple string, no problem, but if it contains variables, this is a way to help people to know what are the expected values.

from : should be config to target the current workspace configuration or appConfig to target the installed app from where the instruction comes configuration.
path : a JSON Path string starting from the from target. The syntax is described in this project.
template : sometimes the event name is not explicit because it comes from computed variables. You can help building values with a string containing static caracters and a ${value} placeholder.

Exemples :

With this in your workspace DSUL:

config:
  value:
    things:
      - foo
      - bar

- emit:
    event: 'event.{{var}}'
    autocomplete:
      var:
        from: config
        path: things~

or

- set:
    name: event
    value: 'event.{{var}}'
- emit:
    event: '{{event}}'
    autocomplete:
      var:
        from: config
        path: things~
        template: event.${value}

will generate automplete values : event.foo and event.bar.

Wait¶

Wait for an event. Pauses the current execution until the requested event is received.

Parameters :

oneOf : Pauses until one of these events is received
timeout : If waits timeouts after N seconds, resume execution & outputs an empty result. Defaults to 20
output : Name of the variable that will store the received event

oneOf parameter :
List of event candidates with the following parameters :

event : Event name, required
filters : Free form JSON object for filtering on any event fields. For example :

- wait:
    oneOf:
      event: someEvent
      filters:
        source.correlationId: '{{run.correlationId}}'

cancelTriggers : If true, cancels the execution of the usual triggers for this event. Other waits still receive this event.

Waiting for an event triggered automation to finish

In addition to every regular events which might be directly listened by automations, wait instruction can also waits for runtime.automations.executed event.
This native events tells when an automation finishes executing, with its source trigger & output.
Waiting for this allows you to emit an event, than hang until whichever automation that received it finishes processing :

AutomationA:
  name: AutomationA
  when:
    endpoint: true
  output: '{{output}}'
  do:
    - emit:
        event: 'someEvent'
        output: emittedIntentEvent
    - wait:
        oneOf:
          - event: runtime.automations.executed
            # Filters specifically for our emited event
            filters:
              payload.trigger.id: '{{emittedIntentEvent.id}}'
        output: event
    - set:
        name: output
        value: '{{event.payload.output}}'

AutomationB:
  name: AutomationB
  when:
    events:
      - someEvent
  output: 'someOutput'
  do: []

createUserTopic¶

Create a new userTopic.

User topics allow sending events to multiple users without knowing who they are in advance, automatically granting them read access to these events without requiring any API Key.

Parameters :

topic : Topic name
userIds : List of the first userIds to join this topic

joinUserTopic¶

Makes a given user join a userTopic.

Joining a userTopic automatically grants read access to any event sent within this topic.

Parameters :

topic : Topic name to join
userIds : List of the userIds to join this topic. If not defined, automatically pick current user's id

Instructions¶

Logic-related¶

Conditions¶

Repeat¶

Break¶

All¶

Variable-related¶

Set¶

Delete¶

Interaction-related¶

Fetch¶

Emit¶

Wait¶

createUserTopic¶

joinUserTopic¶