Reference: Data Types¶

This page gives reference information on Daml’s data types.

Built-in Types¶

Table of built-in primitive types¶

Type	For	Example	Notes
`Int`	integers	`1`, `1000000`, `1_000_000`	`Int` values are signed 64-bit integers which represent numbers between `-9,223,372,036,854,775,808` and `9,223,372,036,854,775,807` inclusive. Arithmetic operations raise an error on overflows and division by `0`. To make long numbers more readable you can optionally add underscores.
`Decimal`	short for `Numeric 10`	`1.0`	`Decimal` values are rational numbers with precision 38 and scale 10.
`Numeric n`	fixed point decimal numbers	`1.0`	`Numeric n` values are rational numbers with `38` total digits. The scale parameter `n` controls the number of digits after the decimal point, so for example, `Numeric 10` values have 28 digits before the decimal point and 10 digits after it, and `Numeric 20` values have 18 digits before the decimal point and 20 digits after it. The value of `n` must be between `0` and `37` inclusive.
`BigNumeric`	large fixed point decimal numbers	`1.0`	`BigNumeric` values are rational numbers with up to `2^16` decimal digits. They can have up to `2^15` digits before the decimal point, and up to `2^15` digits after the decimal point.
`Text`	strings	`"hello"`	`Text` values are strings of characters enclosed by double quotes.
`Bool`	boolean values	`True`, `False`
`Party`	unicode string representing a party	`alice <- getParty "Alice"`	Every party in a Daml system has a unique identifier of type `Party`. To create a value of type `Party`, use binding on the result of calling `getParty`. The party text can only contain alphanumeric characters, `-`, `_` and spaces.
`Date`	models dates	`date 2007 Apr 5`	Permissible dates range from `0001-01-01` to `9999-12-31` (using a year-month-day format). To create a value of type `Date`, use the function `date` (to get this function, import `DA.Date`).
`Time`	models absolute time (UTC)	`time (date 2007 Apr 5) 14 30 05`	`Time` values have microsecond precision with allowed range from `0001-01-01` to `9999-12-31` (using a year-month-day format). To create a value of type `Time`, use a `Date` and the function `time` (to get this function, import `DA.Time`).
`RelTime`	models differences between time values	`seconds 1`, `seconds (-2)`	`RelTime` values have microsecond precision with allowed range from -9,223,372,036,854,775,808ms to 9,223,372,036,854,775,807ms There are no literals for `RelTime`. Instead they are created using one of `days`, `hours`, `minutes`, `seconds`, `miliseconds` and `microseconds` (to get these functions, import `DA.Time`).

Escaping Characters¶

Text literals support backslash escapes to include their delimiter (\") and a backslash itself (\\).

Time¶

Definition of time on the ledger is a property of the execution environment. Daml assumes there is a shared understanding of what time is among the stakeholders of contracts.

Lists¶

[a] is the built-in data type for a list of elements of type a. The empty list is denoted by [] and [1, 3, 2] is an example of a list of type [Int].

You can also construct lists using [] (the empty list) and :: (which is an operator that appends an element to the front of a list). For example:

twoEquivalentListConstructions =
  script do
    assert ( [1, 2, 3] == 1 :: 2 :: 3 :: [] )

Sum a List¶

To sum a list, use a fold (because there are no loops in Daml). See Fold for details.

Records and Record Types¶

You declare a new record type using the data and with keyword:

data MyRecord = MyRecord
  with
    label1 : type1
    label2 : type2
    ...
    labelN : typeN
  deriving (Eq, Show)

where:

label1, label2, …, labelN are labels, which must be unique in the record type
type1, type2, …, typeN are the types of the fields

There’s an alternative way to write record types:

data MyRecord = MyRecord { label1 : type1; label2 : type2; ...; labelN : typeN }
  deriving (Eq, Show)

The format using with and the format using { } are exactly the same syntactically. The main difference is that when you use with, you can use newlines and proper indentation to avoid the delimiting semicolons.

The deriving (Eq, Show) ensures the data type can be compared (using ==) and displayed (using show). The line starting deriving is required for data types used in fields of a template.

In general, add the deriving unless the data type contains function types (e.g. Int -> Int), which cannot be compared or shown.

For example:

-- This is a record type with two fields, called first and second,
-- both of type `Int`
data MyRecord = MyRecord with first : Int; second : Int
  deriving (Eq, Show)

-- An example value of this type is:
newRecord = MyRecord with first = 1; second = 2

-- You can also write:
newRecord = MyRecord 1 2

Data Constructors¶

You can use data keyword to define a new data type, for example data Floor a = Floor a for some type a.

The first Floor in the expression is the type constructor. The second Floor is a data constructor that can be used to specify values of the Floor Int type: for example, Floor 0, Floor 1.

In Daml, data constructors may take at most one argument.

An example of a data constructor with zero arguments is data Empty = Empty {}. The only value of the Empty type is Empty.

Note

In data Confusing = Int, the Int is a data constructor with no arguments. It has nothing to do with the built-in Int type.

Access Record Fields¶

To access the fields of a record type, use dot notation. For example:

-- Access the value of the field `first`
val.first

-- Access the value of the field `second`
val.second

Update Record Fields¶

You can also use the with keyword to create a new record on the basis of an existing replacing select fields.

For example:

myRecord = MyRecord with first = 1; second = 2

myRecord2 = myRecord with second = 5

produces the new record value MyRecord with first = 1; second = 5.

If you have a variable with the same name as the label, Daml lets you use this without assigning it to make things look nicer:

-- if you have a variable called `second` equal to 5
second = 5

-- you could construct the same value as before with
myRecord2 = myRecord with second = second

-- or with
myRecord3 = MyRecord with first = 1; second = second

-- but Daml has a nicer way of putting this:
myRecord4 = MyRecord with first = 1; second

-- or even
myRecord5 = r with second

Note

The with keyword binds more strongly than function application. So for a function, say return, either write return IntegerCoordinate with first = 1; second = 5 or return (IntegerCoordinate {first = 1; second = 5}), where the latter expression is enclosed in parentheses.

Parameterized Data Types¶

Daml supports parameterized data types.

For example, to express a more general type for 2D coordinates:

-- Here, a and b are type parameters.
-- The Coordinate after the data keyword is a type constructor.
data Coordinate a b = Coordinate with first : a; second : b

An example of a type that can be constructed with Coordinate is Coordinate Int Int.

Type Synonyms¶

To declare a synonym for a type, use the type keyword.

For example:

type IntegerTuple = (Int, Int)

This makes IntegerTuple and (Int, Int) synonyms: they have the same type and can be used interchangeably.

You can use the type keyword for any type, including Built-in Types.

Function Types¶

A function’s type includes its parameter and result types. A function foo with two parameters has type ParamType1 -> ParamType2 -> ReturnType.

Note that this can be treated as any other type. You could for instance give it a synonym using type FooType = ParamType1 -> ParamType2 -> ReturnType.

Algebraic Data Types¶

An algebraic data type is a composite type: a type formed by a combination of other types. The enumeration data type is an example. This section introduces more powerful algebraic data types.

Product Types¶

The following data constructor is not valid in Daml: data AlternativeCoordinate a b = AlternativeCoordinate a b. This is because data constructors can only have one argument.

To get around this, wrap the values in a record: data Coordinate a b = Coordinate {first: a; second: b}.

These kinds of types are called product types.

A way of thinking about this is that the Coordinate Int Int type has a first and second dimension (that is, a 2D product space). By adding an extra type to the record, you get a third dimension, and so on.

Sum Types¶

Sum types capture the notion of being of one kind or another.

An example is the built-in data type Bool. This is defined by data Bool = True | False deriving (Eq,Show), where True and False are data constructors with zero arguments . This means that a Bool value is either True or False and cannot be instantiated with any other value.

Please note that all types which you intend to use as template or choice arguments need to derive at least from (Eq, Show).

A very useful sum type is data Optional a = None | Some a deriving (Eq,Show). It is part of the Daml standard library.

Optional captures the concept of a box, which can be empty or contain a value of type a.

Optional is a sum type constructor taking a type a as parameter. It produces the sum type defined by the data constructors None and Some.

The Some data constructor takes one argument, and it expects a value of type a as a parameter.

Pattern Matching¶

You can match a value to a specific pattern using the case keyword.

The pattern is expressed with data constructors. For example, the Optional Int sum type:

import Daml.Script
import DA.Assert

optionalIntegerToText (x : Optional Int) : Text =
  case x of
    None -> "Box is empty"
    Some val -> "The content of the box is " <> show val

optionalIntegerToTextTest =
  script do

In the optionalIntegerToText function, the case construct first tries to match the x argument against the None data constructor, and in case of a match, the "Box is empty" text is returned. In case of no match, a match is attempted for x against the next pattern in the list, i.e., with the Some data constructor. In case of a match, the content of the value attached to the Some label is bound to the val variable, which is then used in the corresponding output text string.

Note that all patterns in the case construct need to be complete, i.e., for each x there must be at least one pattern that matches. The patterns are tested from top to bottom, and the expression for the first pattern that matches will be executed. Note that _ can be used as a catch-all pattern.

You could also case distinguish a Bool variable using the True and False data constructors and achieve the same behavior as an if-then-else expression.

As an example, the following is an expression for a Text:

tmp =
  let
    l = [1, 2, 3]
  in case l of

Notice the use of nested pattern matching above.

Note

An underscore was used in place of a variable name. The reason for this is that Daml Studio produces a warning for all variables that are not being used. This is useful in detecting unused variables. You can suppress the warning by naming the variable with an initial underscore.