Values, Coercion, and Errors

Understand runtime value types, implicit conversions, and spreadsheet-style error propagation.

Formualizer operations work over a compact set of runtime values (number, text, boolean, blank, arrays, and errors).

Coercion rules in practice

Numeric functions coerce text when valid (for example, "42" -> 42).
Invalid coercions return spreadsheet errors instead of panicking.
Blank handling differs by operation (arithmetic vs logical vs text).

Error propagation

Errors usually short-circuit the current expression branch.
Functions may choose to absorb or expose errors based on semantics.
Consistent error values make cross-binding behavior predictable.

Short examples

="1" + 2        -> 3
="x" + 2        -> #VALUE!
=IFERROR(A1, 0)  -> 0 when A1 is an error

Coercion matrix

Source \ Target	Number	Text	Boolean
Number	identity	`"42"`	nonzero → TRUE
Text (numeric)	parse → number	identity	#VALUE!
Text (non-numeric)	#VALUE!	identity	#VALUE!
Boolean TRUE	1	`"TRUE"`	identity
Boolean FALSE	0	`"FALSE"`	identity
Empty	0	`""`	FALSE
Error	propagate	propagate	propagate

Function-specific error behavior

Most arithmetic functions propagate errors immediately. Aggregation functions like SUM and AVERAGE skip error values in ranges by default. IFERROR and IFNA explicitly trap errors. ISERROR, ISERR, and ISNA test for errors without propagating them.