» can Function

can evaluates the given expression and returns a boolean value indicating whether the expression produced a result without any errors.

This is a special function that is able to catch errors produced when evaluating its argument. For most situations where you could use can it's better to use try instead, because it allows for more concise definition of fallback values for failing expressions.

The primary purpose of can is to turn an error condition into a boolean validation result when writing custom variable validation rules. For example:

variable "timestamp" {
  type        = string

  validation { # NOTE: custom validation is currently an opt-in experiment (see link above)
    # formatdate fails if the second argument is not a valid timestamp
    condition     = can(formatdate("", var.timestamp))
    error_message = "The timestamp argument requires a valid RFC 3339 timestamp."
  }
}

The can function can only catch and handle dynamic errors resulting from access to data that isn't known until runtime. It will not catch errors relating to expressions that can be proven to be invalid for any input, such as a malformed resource reference.

» Examples

> local.foo
{
  "bar" = "baz"
}
> can(local.foo.bar)
true
> can(local.foo.boop)
false

The can function will not catch errors relating to constructs that are provably invalid even before dynamic expression evaluation, such as a malformed reference or a reference to a top-level object that has not been declared:

> can(local.nonexist)

Error: Reference to undeclared local value

A local value with the name "nonexist" has not been declared.
  • try, which tries evaluating a sequence of expressions and returns the result of the first one that succeeds.