Expressions
An expression specifies the computation of a value by applying the operators and functions to operands.
Operands and primary expressions
Operands denote the elementary values in an expression.
Primary expressions are the operands for unary and binary expressions. A primary expressions may be a literal, an identifier denoting a variable, or a parenthesized expression.
PrimaryExpression = identifier | Literal | "(" Expression ")" .
Literals
Literals construct a value.
Literal = int_lit
| float_lit
| string_lit
| regex_lit
| duration_lit
| date_time_lit
| pipe_receive_lit
| RecordLiteral
| ArrayLiteral
| DictLiteral
| FunctionLiteral .
Record literals
Record literals construct a value with the record type.
RecordLiteral = "{" RecordBody "}" .
RecordBody = WithProperties | PropertyList .
WithProperties = identifier "with" PropertyList .
PropertyList = [ Property { "," Property } ] .
Property = identifier [ ":" Expression ]
| string_lit ":" Expression .
Examples
{a: 1, b: 2, c: 3}
{a, b, c}
{o with x: 5, y: 5}
{o with a, b}
Array literals
Array literals construct a value with the array type.
ArrayLiteral = "[" ExpressionList "]" .
ExpressionList = [ Expression { "," Expression } ] .
Dictionary literals
Dictionary literals construct a value with the dict type.
DictLiteral = EmptyDict | "[" AssociativeList "]" .
EmptyDict = "[" ":" "]" .
AssociativeList = Association { "," AssociativeList } .
Association = Expression ":" Expression .
Keys can be arbitrary expressions. The type system enforces that all keys are of the same type.
Examples
a = "a"
b = [:] // empty dictionary
c = [a: 1, "b": 2] // dictionary mapping string values to integers
d = [a: 1, 2: 3] // type error: cannot mix string and integer keys
Function literals
A function literal defines a new function with a body and parameters. The function body may be a block or a single expression. The function body must have a return statement if it is an explicit block, otherwise the expression is the return value.
FunctionLiteral = FunctionParameters "=>" FunctionBody .
FunctionParameters = "(" [ ParameterList [ "," ] ] ")" .
ParameterList = Parameter { "," Parameter } .
Parameter = identifier [ "=" Expression ] .
FunctionBody = Expression | Block .
Examples of function literals
() => 1 // function returns the value 1
(a, b) => a + b // function returns the sum of a and b
(x=1, y=1) => x * y // function with default values
(a, b, c) => { // function with a block body
d = a + b
return d / c
}
All function literals are anonymous. A function may be given a name using a variable assignment.
add = (a,b) => a + b
mul = (a,b) => a * b
Function literals are closures and may refer to variables defined in a surrounding block. Those variables are shared between the function literal and the surrounding block.
Function arguments are named. There are no positional arguments. Values implementing a function type must use the same argument names.
apply = (f, x) => f(x: x)
apply(f: (x) => x + 1, x: 2) // 3
apply(f: (a) => a + 1, x: 2) // error, function must use the same argument name `x`.
apply(f: (x, a=3) => a + x, x: 2) // 5, extra default arguments are allowed
Call expressions
A call expression invokes a function with the provided arguments. Arguments must be specified using the argument name. Positional arguments are not supported. Argument order does not matter. When an argument has a default value, it is not required to be specified.
CallExpression = "(" PropertyList ")" .
Examples of call expressions
f(a:1, b:9.6)
float(v:1)
Use short notation in a call expression when the name of every argument matches the name of every parameter.
Examples of short notation in call expressions
add(a: a, b: b) //long notation
add(a, b) // short notation equivalent
add = (a,b) => a + b
a = 1
b = 2
// Don't mix short and long notation.
add(a: a, b)
add(a, b: b)
Pipe expressions
A pipe expression is a call expression with an implicit piped argument. Pipe expressions simplify creating long nested call chains.
Pipe expressions pass the result of the left hand expression as the pipe argument to the right hand call expression. Function literals specify which if any argument is the pipe argument using the pipe literal as the argument’s default value. It is an error to use a pipe expression if the function does not declare a pipe argument.
pipe_receive_lit = "<-" .
Examples of pipe expressions
foo = () => // function body elided
bar = (x=<-) => // function body elided
baz = (y=<-) => // function body elided
foo() |> bar() |> baz() // equivalent to baz(x:bar(y:foo()))
Index expressions
Index expressions access a value from an array based on a numeric index.
IndexExpression = "[" Expression "]" .
Member expressions
Member expressions access a property of a record. They are specified using an expression in one of the following forms:
rec.k
// or
rec["k"]
The property being accessed must be either an identifier or a string literal. In either case the literal value is the name of the property being accessed, the identifier is not evaluated. It is not possible to access a record’s property using an arbitrary expression.
If rec contains an entry with property k, both rec.k and rec["k"] return the value associated with k.
If rec does not contain an entry with property k, both rec.k and rec["k"] return null.
MemberExpression = DotExpression | MemberBracketExpression .
DotExpression = "." identifier .
MemberBracketExpression = "[" string_lit "]" .
Conditional expressions
Conditional expressions evaluate a boolean-valued condition.
If the result is true, the expression that follows the then keyword is evaluated and returned.
If the result is false, the expression that follows the else keyword is evaluated and returned.
In either case, only the branch taken is evaluated and only side effects associated this branch will occur.
ConditionalExpression = "if" Expression "then" Expression "else" Expression .
Conditional expression example
color = if code == 0 then "green" else if code == 1 then "yellow" else "red"
According to the definition above, if a condition evaluates to a null or unknown value, the else branch is evaluated.
Operators
Operators combine operands into expressions. The precedence of the operators is given in the table below. Operators with a lower number have higher precedence.
| Precedence | Operator | Description |
|---|---|---|
| 1 | a() | Function call |
a[] | Member or index access | |
. | Member access | |
| 2 | |> | Pipe forward |
| 3 | () => 1 | FunctionLiteral |
| 4 | ^ | Exponentiation |
| 5 | * / % | Multiplication, division, and modulo |
| 6 | + - | Addition and subtraction |
| 7 | == != | Comparison operators |
< <= | ||
> >= | ||
=~ !~ | ||
| 8 | not | Unary logical operator |
exists | Null check operator | |
| 9 | and | Logical AND |
| 10 | or | Logical OR |
| 11 | if then else | Conditional |
The operator precedence is encoded directly into the grammar as the following.
Expression = ConditionalExpression .
ConditionalExpression = LogicalExpression
| "if" Expression "then" Expression "else" Expression .
LogicalExpression = UnaryLogicalExpression
| LogicalExpression LogicalOperator UnaryLogicalExpression .
LogicalOperator = "and" | "or" .
UnaryLogicalExpression = ComparisonExpression
| UnaryLogicalOperator UnaryLogicalExpression .
UnaryLogicalOperator = "not" | "exists" .
ComparisonExpression = MultiplicativeExpression
| ComparisonExpression ComparisonOperator MultiplicativeExpression .
ComparisonOperator = "==" | "!=" | "<" | "<=" | ">" | ">=" | "=~" | "!~" .
AdditiveExpression = MultiplicativeExpression
| AdditiveExpression AdditiveOperator MultiplicativeExpression .
AdditiveOperator = "+" | "-" .
MultiplicativeExpression = ExponentExpression
| ExponentExpression ExponentOperator MultiplicativeExpression .
| ExponentExpression MultiplicativeOperator MultiplicativeExpression .
MultiplicativeOperator = "*" | "/" | "%" .
ExponentExpression = PipeExpression
| ExponentExpression ExponentOperator PipeExpression .
ExponentOperator = "^" .
PipeExpression = PostfixExpression
| PipeExpression PipeOperator UnaryExpression .
PipeOperator = "|>" .
UnaryExpression = PostfixExpression
| PrefixOperator UnaryExpression .
PrefixOperator = "+" | "-" .
PostfixExpression = PrimaryExpression
| PostfixExpression PostfixOperator .
PostfixOperator = MemberExpression
| CallExpression
| IndexExpression .
Dividing by 0 or using the mod operator with a divisor of 0 will result in an error. Floating point divide by zero produces positive or negative infinity according to the IEEE-754 floating point specification.
Also see Flux Operators.
Was this page helpful?
Thank you for your feedback!
Support and feedback
Thank you for being part of our community! We welcome and encourage your feedback and bug reports for Flux and this documentation. To find support, use the following resources:
Customers with an annual or support contract can contact InfluxData Support.