Type conversion and reflection

Type conversion

Pipefish has a number of built-in functions for type conversion.

Usually they have the same name as the type they convert to:

float converts ints to float, or converts strings to floats if they have the right format.

int converts floats to integers by truncation, or converts strings to ints if they have the right format.

set will turn a tuple into a set.

string will convert anything to a string. This has been given some sensible defaults but can be overloaded for any type except tuple.

literal converts a value into a string which is the Pipefish literal of that value.

tuple will leave a tuple unchanged and turn anything else in to a tuple of length 1.

L ... will convert a list L into a tuple with the same elements.

To turn a tuple t into a list it is only necessary to wrap it in square brackets: [t].

type converts a value to its type.

Cast

cast <type>, <value> casts between types, specifically between clone types and their parent types and vice versa, between int and enum types, and vice-versa; and from lists of values to structs.

Unsafe

unsafe <type>, <value> casts from a parent type to its clone without running any type validation attached to the type. This obviously saves time, but is, as it says "unsafe", and should be used carefully while constructing the basic operations on a type in the library that declares it, not scattered willy-nilly in your code.

Reflection

In the list above we briefly noted that type converts a value to its type. That is, type "walrus" evaluates to string, type 42 evaluates to int, etc.

string and int are themselves members of the type type. That is, type string returns type.

type is a perfectly normal Pipefish type itself, and belongs, of course, to type type. Try not to let this bother you.

We can also use the in keyword to check for type membership: so for example "walrus" in string and string in type are true.