Below is a list of the current SEL functions you can use to populate building blocks and build models. The quick reference below includes a brief explanation of each function and its basic behavior. The functions also include links to our Technical Reference Guide with detailed descriptions and additional guidance about how each function can best be used to power your workflows.

If you have any questions or have a suggestion for a new SEL Function, please feel free to contact us at [email protected] or schedule a time to meet with a member of our team.

Requests & Response

Events to fetch, parse, and return data.

Type	Behavior	Input	Output	Example(s)
Requests	Make a `GET` or `POST` request for JSON, HTML, or CSV data.	`Object` to define `headers`, `params`, `json`, `cookies`, or `data`.	Web request response contents, HTML, CSV, JSON.	`=Request('get', 'https://api.example.com/x.json')`
Parser	Use a JSONPath expression to parse the output of a Request.	JSON, typically returned from a `Request`.	A list of matches.	`=Parser('$..coordinates.lat')`
Response	Compose the `data` field of your model's API response.	Numbers, strings, text, or JSON.	None	`=Response("foo.bar")`, `=Response("myDataList[]")`
Table	Create an in-memory table for querying.	Remote CSV data or a list of dictionaries.	A table reference for use by `Query` events.	`=Table("https://sheets.google.com/my.csv")` or `=Table("contacts")`
Query	Perform a SQL query on a `Table`	A `Table` reference.	A list of dictionaries representing the result set.	`=Query("SELECT * FROM $table WHERE name = 'John Doe'")` or `=Query("SELECT * FROM contacts WHERE age = 44")`

Objects & Arrays

Events to store and send fixed lists or object definitions.

Type	Behavior	Input	Output	Example(s)
Array	Defines a static list to relay to another event.	None, statically defined.	The entire array as a list.	`=Array(['peanut butter', 'jelly', 5])`
Object	Defines a literal object using JSON syntax.	None, statically defined. You may use liquid syntax within string components of the definition.	The object as a dictionary.	`=Object({"warm": true})`

Strings & Text

Events to store or create text.

Type	Behavior	Input	Output	Example(s)
String	Stores and relays a text string.	-	The string.	`=String("Jedi Council")`
Matches	Finds all values that match a provided regular expression.	A list, a string, or text.	A list of matches found within the input.	`=Matches("\d{1,}")`
Text	Write and create long strings of text using liquid syntax.	-	The text as a string.	`=Text(""" I am Ishmael. """)`

Iterators

When you need to cycle through a list.

Type	Behavior	Input	Output	Example(s)
Loop	Takes values in the provided order and sends them downstream.	A list of values.	The provided values in order.	`=Loop([1, 4, 3])`

Generators

Events that produce numeric or text values when executed.

Type	Behavior	Input	Output	Example(s)
AI	Sends prompt to an AI Model, returns a JSON object with the model response.	An `Object` defining a prompt	The content of the model response as JSON object.	`=Ai("gpt-4o", "{{ OPENAI_KEY}}")`
Search	Calls Tavily's Search API and serves up the results as an object.	Search term, phrase, or question.	Search result with up to 5 sources cited.	`=Search("When will Mariah Carey want us to start playing Christmas music this year?")`
Source	Generates a stream of numeric values.	Number	A stream of numeric values.	`=Source(100k)` `=Source(100k @ 3%)` `=Source(100k...110k @ 3%)`
Curve	Generates numeric values based on a specified curve shape.	-	A stream of numeric values.	`=Curve('s', [365, 1k])`
Trend	From a list of numeric values, generates a trended series of values.	-	A stream of numeric values.	`=Trend([12.1k,15.5k,13.6k,14.4k])` `=Trend([100,300,200,225])`
Extend	Extends a list of numeric values using the slope between the last two values of the provided input.	-	A stream of numeric values.	`=Extend([1,2,3,10])`

Containers

Events that store numeric value.

Type	Behavior	Example(s)
Pool	Collects numeric values with no limit.	`=Pool(1.4m)` `=Pool(1.4m @ 1%)`
Pipe	Relays numeric values downstream.	`=Pipe(0)` `=Pipe(0, 100)`
Bucket	Collects numeric values up to a specified limit; overflow passes downstream.	`=Bucket(0, 500)`
Tipping Bucket	Collects numeric values up to a limit; empties when limit is reached.	`=Tippingbucket(0, 500)`
Funnel	Releases a specified amount of numeric value from its store each time it is triggered.	`=Funnel(100k, 200)`
Gate	Opens on a schedule, releasing all of the numeric value it has collected. Any value not sent downstream is retained.	`=Gate(0)`
Silo	Stores numeric value in a container; container has a minimum and maximum size. Excess value and overdraws are passed downstream.	`=Silo(100, [-50, 1k])` `=Silo(100, [-100,])`
Wheel	Stores numeric value in a rotating list that follows a calendar.	`=Wheel(12)` `=Wheel(4, [250, 200, 175, 99])`
Base	Stores numeric values in a list; sends the sum of those values to downstream events. Second argument specifies retention by cohort.	`=Base([] @ [95, 65, 54, 32, 23]%)`

Timers

Events to manage time for simulation models.

Type	Behavior	Example(s)
Wait	Insert a delay.	`=Wait(7, 'days')` `=Wait(3, 'months')`
Block	Close off the flow of values for a certain amount of time.	`=Block(3, 'months')` `=Block(1, 'year')` `=Block(14, 'days')`
Halt	Stop the simulation.	`=Halt()`
Stopwatch	Stops the simulation after a specified amount of time.	`=Stopwatch(3600\<elapsed_time_allowed>)`

Transforms

Events to alter a value.

Type	Behavior	Example(s)
Transform	Modify the value (or list of values) received before relaying it downstream.	`=T('max')` `=T('numbers')` `=T('abs')` `=T('round')` `=T('liters', 'gallons')` `=T('yards', 'meters')` `=T('kilometers', 'miles')`

Conditionals

Events to test values.

Event	Behavior	Example(s)
If	Test a value and relay it if True.	`=If('>', 0)` `=If('gte', 5)`
Once	Test a value and relay it if True, but only once. False on all subsequent calls.	`=Once('>', 100k)`

Triggers

Events to control the simulated timing of other events.

Event	Behavior	Example(s)
Every	Cause downstream events to execute on a specific schedule within a simulation model.	`=Every("0 8,17 * * 1-5"<hours_of_operation>, 'America/Chicago'<timezone>)`

Tiering & Scale

Events to split values into arrays to perform matrix multiplication.

Event	Behavior	Example(s)
Tiering	Define a set of tiers and identify which tier an inputted number belongs to.	`=Tiering([5, 10, 20])`
Scale	Define breakpoints; given a value, split it into those breakpoints.	`=Scale([150, 250, 500])`

Keep reading: Event References in SEL