sicher

CRAN status R-CMD-check

sicher (German for safe or certain and pronounced zeesher) is an R package that brings runtime type safety to R programming, inspired by TypeScript for JavaScript. Declare types for your variables and have them enforced automatically on every assignment, catching type errors early and making your code more robust and self-documenting. 🛡️

sicher implements runtime type safety in R by attaching types to variables via active bindings rather than traditional attributes or classes. When a variable is declared with a type (e.g., x %:% Numeric %<-% 5), it is replaced by an active binding that wraps the value inside a closure with a private environment storing both the value and its associated type.

Every read and write to the variable is intercepted: reads return the stored value, while writes trigger validation through the type’s checker function before updating. This design provides strong guarantees that values cannot violate their declared types, enabling expressive and composable type definitions (including unions, structured lists, and data frame schemas).

📦 Installation

You can install the sicher package from CRAN with:

install.packages("sicher")

or install the development version from GitHub:

# install.packages("devtools")
devtools::install_github("feddelegrand7/sicher")

🚀 Quick Start

library(sicher)

# Annotate a variable with a type, then assign a value
name %:% String  %<-% "Alice"
age  %:% Numeric %<-% 30

# The type is enforced on every subsequent assignment
age <- "thirty"   # Error: Type error
#> Error: Type error in 'age': Expected numeric, got string
#> Received: thirty

✨ Core Features

🧩 Built-in Types

sicher ships with a complete set of primitive and container types:

Type Checks
Integer is.integer()
Double is.double()
Numeric is.numeric()
String is.character()
Bool is.logical()
List is.list()
DataFrame is.data.frame()
Function is.function()
Any always passes
Null is.null() 
x    %:% Integer   %<-% 42L
y    %:% Double    %<-% 3.14
flag %:% Bool      %<-% TRUE
df   %:% DataFrame %<-% data.frame(a = 1:3)

⚙️ Operators

Operator Purpose
%:% Annotate a variable with a type
%<-% Assign a value (type-checked)

🔧 Type Modifiers

1️⃣ Scalar() — single-element values

single %:% Scalar(Numeric) %<-% 42
single <- c(1, 2, 3)   # Error: length > 1
#> Error: Type error in 'single': Expected scalar<numeric>, got double of length 3
#> Received: [1, 2, 3]

🔒 Readonly() — immutable variables

PI %:% Readonly(Double) %<-% 3.14159
PI <- 3.0   # Error: cannot reassign readonly variable
#> Error: Cannot reassign readonly variable 'PI'. Remove Readonly() from the type declaration if mutation is needed.

❓ Optional() — nullable values

middle_name %:% Optional(String) %<-% NULL   # OK
middle_name <- "Marie"                        # Also OK
middle_name <- 123                            # Error: not string or null
#> Error: Type error in 'middle_name': Expected string | null, got double
#> Received: 123

🚫 NonNA() — reject NA values

Prevents any NA from entering a typed variable — a common source of silent bugs in data pipelines. The underlying type is validated first; NonNA adds an extra NA-free guarantee on top.

salary %:% NonNA(Numeric) %<-% c(1800, 2300, 4000)
salary <- c(1800, NA, 4000)   # Error: value contains NA(s)
#> Error: Type error: expected non_na<numeric> (no NA values), but value contains NA(s)

Composes freely with other modifiers:

clean_tag %:% NonNA(Scalar(String)) %<-% "admin"
clean_tag <- NA_character_   # Error
#> Error: Type error: expected non_na<scalar<string>> (no NA values), but value contains NA(s)

📐 Between(min, max) — closed-interval range

Accepts numeric values within the closed interval [min, max]. Every element of a vector must satisfy the bounds. NA values are always rejected.

age   %:% Between(0, 150)   %<-% 30
score %:% Between(0.0, 1.0) %<-% 0.95
score <- 1.5    # Error: 1.5 is outside [0, 1]
#> Error: Type error: expected between[0, 1], but value [1.5] is outside [0, 1]
score <- NA     # Error: value contains NA(s)
#> Error: Type error: Expected between[0, 1], got bool
#> Received: NA

🔍 Matches(pattern) — regex-constrained strings

Validates every element of a character vector against a Perl-compatible regex. The pattern is compiled eagerly at construction time, so typos in the regex are caught immediately. An empty character(0) passes vacuously; pair with NonEmpty() to also require at least one element.

email %:% Matches("^[^@]+@[^@]+\\.[^@]+$") %<-% "user@example.com"
email <- "not-an-email"   # Error: does not match pattern
#> Error: Type error: expected matches("^[^@]+@[^@]+\.[^@]+$"), but value ["not-an-email"] does not match the pattern

hex_color %:% NonEmpty(Matches("^#[0-9A-Fa-f]{6}$")) %<-% c("#FF5733", "#1A2B3C")

📭 NonEmpty() — reject empty vectors and lists

Rejects zero-length vectors, strings, and lists. For data frames it checks nrow > 0. The base type is validated first, then the length is checked.

tags %:% NonEmpty(String) %<-% c("r", "types")
tags <- character(0)   # Error: length > 0
#> Error: Type error: expected non_empty<string> (length > 0), but got an empty value
tags <- "r"            # OK — length 1 is fine

🔀 Union Types

Accept more than one type with |:

id %:% (String | Numeric) %<-% "user123"
id <- 456   # Also OK
id <- TRUE  # Error: not string or numeric
#> Error: Type error in 'id': Expected string | numeric, got bool
#> Received: TRUE

🏷️ Enum Types

Restrict a value to an explicit set of allowed values with Enum():

status %:% Enum("draft", "published", "archived") %<-% "draft"
status <- "published"   # OK
status <- "deleted"     # Error: not one of the allowed enum values
#> Error: Type error in 'status': Expected enum["draft", "published", "archived"], got string
#> Received: deleted

priority %:% Enum(1, 2, 3) %<-% 2
priority <- 3            # OK
priority <- 5            # Error
#> Error: Type error in 'priority': Expected enum[1, 2, 3], got double
#> Received: 5

🔒 Literal Types

Use Literal() when the value itself is part of the type, similar to TypeScript literal types:

direction %:% Literal("left", "right") %<-% "left"
direction <- "right"                 # OK
direction <- c("left", "right")    # Error: Literal only accepts scalar values
#> Error: Type error in 'direction': Expected literal["left", "right"], got string of length 2
#> Received: [left, right]

status_code %:% Literal(200, 404) %<-% 200
status_code <- 404                    # OK
status_code <- 200L                   # Error: 200L is an integer, not the double literal 200
#> Error: Type error in 'status_code': Expected literal[200, 404], got integer
#> Received: 200

📏 Size-constrained Vectors

Append [n] to any type to require an exact vector length:

coords %:% Numeric[3] %<-% c(1, 2, 3)
coords <- c(4, 5, 6)   # OK — same length
coords <- c(1, 2)      # Error: wrong length
#> Error: Type error in 'coords': Expected numeric[3], got double of length 2
#> Received: [1, 2]

📋 Structured List Types

Define object-like schemas with create_list_type():

Person <- create_list_type(list(
  name  = String,
  age   = Numeric,
  email = Optional(String)   # nullable field
))

person %:% Person %<-% list(name = "Alice", age = 30, email = "alice@example.com")

person <- list(name = "Bob")   # Error: missing required field 'age'
#> Error: Type error: Expected {name: string, age: numeric, email?: string | null}, got list
#> Details: Missing required field(s): age (expected fields: name, age)
#> Received: list with fields: [name]

Use extend() when a new schema should reuse an existing list type instead of redefining every field:

Employee <- extend(Person, list(
  role       = String,
  department = Optional(String)
))

employee %:% Employee %<-% list(
  name = "Alice",
  age = 30,
  role = "Engineer"
)

employee <- list(
  name = "Alice",
  age = 30,
  role = "Engineer",
  department = "Platform"
)

If extra reuses a field name from the base type, the new field definition overrides the old one and extend() emits a warning so the change is explicit.

🗄️ Data Frame Schemas

Validate column names and types with create_dataframe_type():

UserTable <- create_dataframe_type(list(
  id       = Integer,
  username = String,
  active   = Bool
))

users %:% UserTable %<-% data.frame(
  id       = 1:2,
  username = c("alice", "bob"),
  active   = c(TRUE, FALSE)
)

# Wrong column type fails immediately
users <- data.frame(
  id       = c("1", "2"),   # Error: id must be integer
  username = c("alice", "bob"),
  active   = c(TRUE, FALSE)
)
#> Error: Type error in 'id': Expected integer, got string of length 2
#> Received: [1, 2]

📦 Homogeneous Lists with ListOf()

Validate every element of a list against the same type:

TodoItem <- create_list_type(list(
  id        = Numeric,
  title     = String,
  completed = Bool
))

TodoList <- ListOf(TodoItem)

todos %:% TodoList %<-% list(
  list(id = 1, title = "Buy milk",  completed = FALSE),
  list(id = 2, title = "Read book", completed = TRUE)
)

todos <- list(
  list(id = 1, title = "Buy milk", completed = FALSE),
  list(wrong = "shape")   # Error: element does not match TodoItem
)
#> Error: Type error in 'todos': Expected list<{id: numeric, title: string, completed: bool}>, got list of length 2
#> Received: list of length 2

🛠️ Custom Types

Use create_type() to define your own validator with any predicate function:

Positive <- create_type("positive", function(x) is.numeric(x) && all(x > 0))

value %:% Positive %<-% 5
value <- -1   # Error
#> Error: Type error in 'value': Expected positive, got double
#> Received: -1

Use Literal() for TypeScript-style exact scalar values, Enum() for membership in a finite set that may also allow vectors of allowed values, and create_type() when the rule is more general than membership in a predefined list.

🔤 Typed Functions

Use typed_function() to wrap any function with runtime type checks on its parameters and, optionally, its return value — a typed function signature for R:

# Basic typed function — checks params and return type
add <- typed_function(
  function(x, y) x + y,
  params  = list(x = Numeric, y = Numeric),
  .return = Numeric
)

add(1, 2)     # Returns 3
#> [1] 3
add("a", 2)   # Error: Type error in 'x'
#> Error: Type error in 'x': Expected numeric, got string
#> Received: a
Num_not_inf <- create_type(
  name = "Num_not_inf", 
  checker = function(x) {
    is.numeric(x) && !is.infinite(x)
  }
)


divide <- function(a, b) {
  return(a / b)
}

divide_safe <- typed_function(
  fn = divide, 
  params = list(
    a = Num_not_inf, 
    b = Num_not_inf
  ), 
  .return = Num_not_inf
)


divide_safe(10, 2) # works normally
#> [1] 5
divide_safe(10, 0) # fails as 10/0 returns Inf
#> Error: Type error in '<return value>': Expected Num_not_inf, got double
#> Received: Inf
# Optional parameter
greet <- typed_function(
  function(name, title = NULL) {
    if (is.null(title)) paste("Hello,", name)
    else paste("Hello,", title, name)
  },
  params = list(name = String, title = Optional(String))
)

greet("Alice")                   # "Hello, Alice"
#> [1] "Hello, Alice"
greet("Alice", title = "Dr.")    # "Hello, Dr. Alice"
#> [1] "Hello, Dr. Alice"
greet("Alice", title = 42)       # Error: Type error in 'title'
#> Error: Type error in 'title': Expected string | null, got double
#> Received: 42
# Union type in params
describe <- typed_function(
  function(id) paste("ID:", id),
  params  = list(id = String | Numeric),
  .return = String
)

describe("abc")   # "ID: abc"
#> [1] "ID: abc"
describe(123)     # "ID: 123"
#> [1] "ID: 123"
describe(TRUE)    # Error: Type error in 'id'
#> Error: Type error in 'id': Expected string | numeric, got bool
#> Received: TRUE

You can also define an object to expect as a return value:

Person <- create_list_type(
  type_spec = list(
    name = String, 
    age = Numeric
  )
)

get_person_info_as_list <- function(name, age) {
  return(list(
    name = name, 
    age = age
  ))
}

get_person_info_as_message <- function(name, age) {
  return(
    paste("Hi my name is ", name, " I'm ", age, " years old")
  )
}

get_person_info_as_list_safe <- typed_function(
  fn = get_person_info_as_list, 
  params = list(name = String, age = Numeric), 
  .return = Person
)

get_person_info_as_list_safe(name = "Omar", age = 30) # works fine
#> $name
#> [1] "Omar"
#> 
#> $age
#> [1] 30

get_person_info_as_message_safe <- typed_function(
  fn = get_person_info_as_message, 
  params = list(name = String, age = Numeric), 
  .return = Person
)

# Should fail as the function does not return a Person list anymore
get_person_info_as_message_safe(name = "Omar", age = 30) 
#> Error: Type error in '<return value>': Expected {name: string, age: numeric}, got string
#> Received: Hi my name is  Omar  I'm  30  years old

🌍 Function usage

# Catch bad payroll data early instead of getting silent NAs
calculate_mean_payroll <- function(salaries) {
  salaries %:% Numeric %<-% salaries
  mean(salaries)
}

calculate_mean_payroll(c(1800, 2300, 4000))   # Works fine
#> [1] 2700

calculate_mean_payroll(c(1800, "2300", 4000)) # Error: type mismatch
#> Error: Type error in 'salaries': Expected numeric, got string of length 3
#> Received: [1800, 2300, 4000]

🔍 Type inference

You can automatically infer the type for an R object using infer_type():

# Primitives
infer_type(42L)                # Integer
#> <type: integer >
infer_type(3.14)               # Double
#> <type: double >
infer_type("abc")              # String
#> <type: string >
infer_type(TRUE)               # Bool
#> <type: bool >
infer_type(NULL)               # Null
#> <type: null >
infer_type(function(x) x + 1)  # Function
#> <type: Function >

# Default mode infers types, not observed lengths
infer_type(c(1L, 2L, 3L))      # Integer
#> <type: integer >
infer_type(c(1, 2, 3))         # Double
#> <type: double >
infer_type(c("a", "b"))        # String
#> <type: string >
infer_type(c(TRUE, FALSE))     # Bool
#> <type: bool >

# Named and unnamed lists
infer_type(list(a = 1L, b = "x")) # create_list_type(list(a = Integer, b = String))
#> <type: {a: integer, b: string} >
infer_type(list(1L, 2L, 3L))   # ListOf(Integer)
#> <type: list<integer> >
infer_type(list(1L, "a"))      # List
#> <type: list >
infer_type(list(a = NULL, b = 1)) # create_list_type(list(a = Optional(Any), b = Double))
#> <type: {a?: any | null, b: double} >

# Data frames
infer_type(data.frame(x = 1:3, y = c("a", "b", "c"), stringsAsFactors = FALSE))
#> <type: data.frame{x: integer, y: string} >
# create_dataframe_type(list(x = Integer, y = String))

# Use strict = TRUE to also infer Scalar() and [n] size constraints
infer_type(42L, strict = TRUE)           # Scalar(Integer)
#> <type: scalar<integer> >
infer_type(c("a", "b"), strict = TRUE) # String[2]
#> <type: string[2] >
infer_type(data.frame(x = 1:3), strict = TRUE)
#> <type: data.frame{x: integer[3]} >

Global Options

The behavior of sicher can be controlled globally using R’s options() mechanism. These options allow you to switch between strict type enforcement and a fully disabled mode depending on your use case.

sicher.mode

Controls how typed assignments (%:% and %<-%) behave.

Values

How to set options

You can configure the mode globally, for example, disabling strict type checking with:

options(sicher.mode = "off")

Or enable strict typing with:

options(sicher.mode = "on") # this is the default

When to use each mode

Example

my_var_x %:% Numeric %<-% 10   # validated and bound with type enforcement

my_var_x <- "string" # trigger an error
#> Error: Type error in 'my_var_x': Expected numeric, got string
#> Received: string
options(sicher.mode = "off")
my_var_y %:% Numeric %<-% 10 
my_var_y <- "string" # nothing happens

It can also be temporarily scoped:

withr::with_options(
  list(sicher.mode = "off"),
  {
    my_var_y %:% Numeric %<-% 10
    my_var_y <- "string"
  }
)

📚 Learn More

Full documentation and worked examples are available at the package website.

Code of Conduct

Please note that the sicher project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.