chunk-around

array →
chunk-around(
    position: integer
) → tuple

Separates the element at a zero-based position from the elements before and after it, returning the three parts as a tuple. Returns null when the position is invalid or the input cannot be evaluated.

Parameters

Name	Type	Required	Description
position	integer	Yes	The zero-based position of the element to separate.

Argument evaluation

• position: Evaluated once in the enclosing context.

Behavior

chunk-around materializes the input and returns T(before, selected, after). The selected value is preserved as a scalar, including when it is null or structured. A position is valid only when it identifies an existing element.

A common use is resuming an approval workflow where the item at the cursor has a distinct role. Given these steps:

{ValidateInvoice, ManagerApproval, FinanceApproval, ReleasePayment}

and a current position of 2, the required structure is completed steps | current step | future steps:

steps | chunk-around(2)
→ T(
    {ValidateInvoice, ManagerApproval},
    FinanceApproval,
    {ReleasePayment}
)

FinanceApproval remains a single workflow-step record, so it can be displayed, executed, or updated as the current step. By contrast, chunk-on(2) returns only the values before the cursor and the values from the cursor onward:

steps | chunk-on(2)
→ T(
    {ValidateInvoice, ManagerApproval},
    {FinanceApproval, ReleasePayment}
)

The same three-role structure—past | selected/current item | future—appears in workflow engines, carousel focus, undo/redo histories, breadcrumb navigation, and processing a specific failed event in a sequence. The equivalent result can be constructed by splitting the right chunk again, but chunk-on alone does not distinguish the current item from future items. chunk-around directly provides this array-zipper operation.

Examples

{10, 20, 30, 40} | chunk-around(2) → T({10, 20}, 30, {40})

Kind: Function
Scope: array/partitioning
Aliases: None