Generics & Templates

Generics allow you to write type-safe, reusable code. Here's how different languages handle generic programming.

Basic Generics

JavaScript

// JavaScript has no built-in generics
// Use TypeScript for generic support

// TypeScript generics:
// function identity<T>(value: T): T {
//     return value;
// }
// identity<string>("hello");
// identity<number>(42);

// Generic interface (TypeScript)
// interface Container<T> {
//     value: T;
//     getValue(): T;
// }

// Generic class (TypeScript)
// class Box<T> {
//     constructor(private value: T) {}
//     get(): T { return this.value; }
// }
// const box = new Box<string>("hello");

PHP

<?php
// PHP has limited generics (PHPStan/Psalm annotations)

// Template annotation (PHPStan)
/**
* @template T
* @param T $value
* @return T
*/
function identity($value) {
  return $value;
}

/**
* @template T
*/
class Box {
  /** @var T */
  private $value;

  /** @param T $value */
  public function __construct($value) {
      $this->value = $value;
  }

  /** @return T */
  public function get() {
      return $this->value;
  }
}

// Usage analyzed by static analysis tools
$box = new Box("hello");  // Box<string>

Rust

// Generic function
fn identity<T>(value: T) -> T {
  value
}

let s = identity("hello");
let n = identity(42);

// Generic struct
struct Box<T> {
  value: T,
}

impl<T> Box<T> {
  fn new(value: T) -> Self {
      Box { value }
  }

  fn get(&self) -> &T {
      &self.value
  }
}

let box_str = Box::new("hello");
let box_num = Box::new(42);

// Multiple type parameters
struct Pair<A, B> {
  first: A,
  second: B,
}

let pair = Pair { first: "key", second: 42 };

Go

// Generic function (Go 1.18+)
func Identity[T any](value T) T {
  return value
}

s := Identity("hello")
n := Identity(42)

// Generic struct
type Box[T any] struct {
  Value T
}

func NewBox[T any](value T) Box[T] {
  return Box[T]{Value: value}
}

box := NewBox("hello")

// Multiple type parameters
type Pair[A any, B any] struct {
  First  A
  Second B
}

pair := Pair[string, int]{First: "key", Second: 42}

// Generic interface
type Container[T any] interface {
  Get() T
  Set(value T)
}

Python

# Type hints with generics (Python 3.12+)
def identity[T](value: T) -> T:
  return value

# Older syntax (Python 3.9+)
from typing import TypeVar, Generic

T = TypeVar('T')

def identity(value: T) -> T:
  return value

# Generic class
class Box(Generic[T]):
  def __init__(self, value: T) -> None:
      self.value = value

  def get(self) -> T:
      return self.value

box: Box[str] = Box("hello")
box_num: Box[int] = Box(42)

# Multiple type parameters
from typing import Tuple
K = TypeVar('K')
V = TypeVar('V')

class Pair(Generic[K, V]):
  def __init__(self, key: K, value: V):
      self.key = key
      self.value = value

Zig

// Comptime generics
fn identity(comptime T: type, value: T) T {
  return value;
}

const s = identity([]const u8, "hello");
const n = identity(i32, 42);

// Generic struct (function returning type)
fn Box(comptime T: type) type {
  return struct {
      value: T,

      const Self = @This();

      pub fn init(value: T) Self {
          return .{ .value = value };
      }

      pub fn get(self: Self) T {
          return self.value;
      }
  };
}

const StringBox = Box([]const u8);
const box = StringBox.init("hello");

// Generic with multiple types
fn Pair(comptime A: type, comptime B: type) type {
  return struct { first: A, second: B };
}

C#

// Generic method
T Identity<T>(T value) => value;

string s = Identity("hello");
int n = Identity(42);

// Generic class
class Box<T>
{
  public T Value { get; }
  public Box(T value) => Value = value;
  public T Get() => Value;
}

var box = new Box<string>("hello");

// Multiple type parameters
class Pair<TKey, TValue>
{
  public TKey Key { get; }
  public TValue Value { get; }
  public Pair(TKey key, TValue value)
  {
      Key = key;
      Value = value;
  }
}

// Generic interface
interface IContainer<T>
{
  T Get();
  void Set(T value);
}

C++

// Function template
template<typename T>
T identity(T value) {
  return value;
}

auto s = identity(std::string("hello"));
auto n = identity(42);

// Class template
template<typename T>
class Box {
  T value_;
public:
  Box(T value) : value_(value) {}
  T get() const { return value_; }
};

Box<std::string> box("hello");
Box<int> boxNum(42);

// Multiple type parameters
template<typename A, typename B>
struct Pair {
  A first;
  B second;
};

Pair<std::string, int> pair{"key", 42};

// Variadic templates
template<typename... Args>
void print(Args... args) {
  ((std::cout << args << " "), ...);
}

C

// C has no generics
// Use void pointers for generic-like behavior

// Generic swap using void*
void swap(void* a, void* b, size_t size) {
  void* temp = malloc(size);
  memcpy(temp, a, size);
  memcpy(a, b, size);
  memcpy(b, temp, size);
  free(temp);
}

int a = 1, b = 2;
swap(&a, &b, sizeof(int));

// _Generic (C11 type-generic expressions)
#define print_val(x) _Generic((x), \
  int: printf("%d\n", x),          \
  double: printf("%f\n", x),       \
  char*: printf("%s\n", x)         \
)

print_val(42);       // prints int
print_val(3.14);     // prints double
print_val("hello");  // prints string

Java

// Generic method
<T> T identity(T value) {
  return value;
}

String s = identity("hello");
Integer n = identity(42);

// Generic class
class Box<T> {
  private T value;

  public Box(T value) { this.value = value; }
  public T get() { return value; }
}

Box<String> box = new Box<>("hello");

// Multiple type parameters
class Pair<K, V> {
  private K key;
  private V value;
  public Pair(K key, V value) {
      this.key = key;
      this.value = value;
  }
}

// Wildcards
void printList(List<?> list) { }
void addNumbers(List<? super Integer> list) { }
void readNumbers(List<? extends Number> list) { }

Ruby

# Ruby is dynamically typed, no generics needed
# Duck typing serves a similar purpose

# Generic-like behavior via duck typing
class Box
attr_reader :value

def initialize(value)
  @value = value
end

def get
  @value
end
end

box_str = Box.new("hello")
box_num = Box.new(42)

# RBS type signatures (for type checking)
# class Box[T]
#   attr_reader value: T
#   def initialize: (T) -> void
#   def get: () -> T
# end

# Sorbet (static type checker)
# sig { type_parameters(:T).params(value: T).returns(T) }
# def identity(value); value; end

Swift

// Generic function
func identity<T>(_ value: T) -> T {
  return value
}

let s = identity("hello")
let n = identity(42)

// Generic struct
struct Box<T> {
  let value: T

  func get() -> T {
      return value
  }
}

let box = Box(value: "hello")

// Multiple type parameters
struct Pair<A, B> {
  let first: A
  let second: B
}

let pair = Pair(first: "key", second: 42)

// Generic enum
enum Result<Success, Failure: Error> {
  case success(Success)
  case failure(Failure)
}

Type Constraints

JavaScript

// TypeScript constraints
// function max<T extends number>(a: T, b: T): T {
//     return a > b ? a : b;
// }

// Interface constraint
// interface HasLength { length: number; }
// function logLength<T extends HasLength>(item: T): void {
//     console.log(item.length);
// }
// logLength("hello");     // OK
// logLength([1, 2, 3]);   // OK
// logLength(42);           // Error

// keyof constraint
// function getProperty<T, K extends keyof T>(
//     obj: T, key: K
// ): T[K] {
//     return obj[key];
// }

PHP

<?php
// PHPStan template bounds

/**
* @template T of Comparable
* @param T $a
* @param T $b
* @return T
*/
function max_val($a, $b) {
  return $a->compareTo($b) > 0 ? $a : $b;
}

/**
* @template T of \Countable
* @param T $item
* @return int
*/
function getCount($item): int {
  return count($item);
}

// Runtime: use interfaces for constraints
interface Printable {
  public function toString(): string;
}

function printItem(Printable $item): void {
  echo $item->toString();
}

Rust

// Trait bounds
fn max<T: PartialOrd>(a: T, b: T) -> T {
  if a > b { a } else { b }
}

// Multiple bounds
fn print_debug<T: std::fmt::Debug + Clone>(value: T) {
  println!("{:?}", value);
}

// Where clause (cleaner for complex bounds)
fn process<T>(value: T) -> String
where
  T: std::fmt::Display + Clone + Send,
{
  format!("{}", value)
}

// Associated type bounds
fn sum<I>(iter: I) -> i32
where
  I: Iterator<Item = i32>,
{
  iter.fold(0, |acc, x| acc + x)
}

Go

// Type constraints with interfaces
type Number interface {
  ~int | ~int64 | ~float64
}

func Max[T Number](a, b T) T {
  if a > b {
      return a
  }
  return b
}

// comparable constraint
func Contains[T comparable](slice []T, item T) bool {
  for _, v := range slice {
      if v == item {
          return true
      }
  }
  return false
}

// Custom constraint interface
type Stringer interface {
  String() string
}

func PrintAll[T Stringer](items []T) {
  for _, item := range items {
      fmt.Println(item.String())
  }
}

Python

from typing import TypeVar, Protocol

# Bound type variable
from typing import Comparable
T = TypeVar('T', bound=int)

def max_val(a: T, b: T) -> T:
  return a if a > b else b

# Protocol (structural typing)
class HasLength(Protocol):
  def __len__(self) -> int: ...

def log_length(item: HasLength) -> None:
  print(len(item))

log_length("hello")    # OK
log_length([1, 2, 3])  # OK

# Constrained TypeVar
T = TypeVar('T', int, float, str)

def add(a: T, b: T) -> T:
  return a + b

Zig

// Comptime constraints via compile errors
fn max(comptime T: type, a: T, b: T) T {
  // Check for ordering support at comptime
  switch (@typeInfo(T)) {
      .Int, .Float => {},
      else => @compileError("Type must be numeric"),
  }
  return if (a > b) a else b;
}

// Interface-like via function pointers
fn Sortable(comptime T: type) type {
  return struct {
      lessThan: fn (T, T) bool,
  };
}

// Trait-like pattern
fn print(value: anytype) void {
  // Works with any type that supports formatting
  std.debug.print("{any}\n", .{value});
}

C#

// Interface constraint
T Max<T>(T a, T b) where T : IComparable<T>
{
  return a.CompareTo(b) > 0 ? a : b;
}

// Multiple constraints
void Process<T>(T value)
  where T : class, IDisposable, new()
{
  using (value) { /* work */ }
}

// Constraint types:
// where T : struct        // Value type
// where T : class         // Reference type
// where T : new()         // Has parameterless ctor
// where T : BaseClass     // Derives from class
// where T : IInterface    // Implements interface
// where T : notnull       // Non-nullable (C# 8+)

// Generic math (C# 11+)
T Sum<T>(T[] values) where T : INumber<T>
{
  T sum = T.Zero;
  foreach (var v in values) sum += v;
  return sum;
}

C++

// Concepts (C++20)
template<typename T>
concept Comparable = requires(T a, T b) {
  { a < b } -> std::convertible_to<bool>;
  { a > b } -> std::convertible_to<bool>;
};

template<Comparable T>
T max_val(T a, T b) {
  return a > b ? a : b;
}

// requires clause
template<typename T>
  requires std::is_arithmetic_v<T>
T add(T a, T b) {
  return a + b;
}

// SFINAE (pre-C++20)
template<typename T,
  typename = std::enable_if_t<std::is_integral_v<T>>>
T square(T x) {
  return x * x;
}

C

// C has no type constraints
// Use preprocessor macros for type-generic code

#define MAX(a, b) ((a) > (b) ? (a) : (b))

int max_int = MAX(3, 5);
double max_double = MAX(3.14, 2.71);

// Type-generic math (C11)
#include <tgmath.h>

// sqrt works for float, double, long double
double d = sqrt(16.0);
float f = sqrt(16.0f);

// _Static_assert for compile-time checks
#define ASSERT_NUMERIC(x) _Static_assert( \
  _Generic((x),                          \
      int: 1, float: 1, double: 1,      \
      default: 0                         \
  ), "Must be numeric")

Java

// Upper bound
<T extends Comparable<T>> T max(T a, T b) {
  return a.compareTo(b) > 0 ? a : b;
}

// Multiple bounds
<T extends Comparable<T> & Serializable>
void process(T value) { }

// Bounded wildcards
// Upper bound: read-only
void printAll(List<? extends Number> list) {
  for (Number n : list) System.out.println(n);
}

// Lower bound: write-only
void addIntegers(List<? super Integer> list) {
  list.add(42);
}

// Type erasure: generics are compile-time only
// At runtime, Box<String> and Box<Integer>
// are both just Box

Ruby

# Ruby uses duck typing instead of constraints
# No compile-time type checking

# Convention: check capabilities at runtime
def process(item)
unless item.respond_to?(:length)
  raise ArgumentError, "must have length"
end
item.length
end

# RBS type signatures
# interface _Comparable
#   def <=>: (untyped) -> Integer
# end
#
# def max: [T < _Comparable] (T, T) -> T

# Sorbet generics with bounds
# sig do
#   type_parameters(:T)
#     .params(a: T, b: T)
#     .returns(T)
#     .where(T: Comparable)
# end
# def max(a, b); a > b ? a : b; end

Swift

// Protocol constraint
func max<T: Comparable>(_ a: T, _ b: T) -> T {
  return a > b ? a : b
}

// Multiple constraints
func process<T: Hashable & Codable>(_ value: T) {
  // T must be both Hashable and Codable
}

// Where clause
func allEqual<T>(_ a: T, _ b: T) -> Bool
  where T: Equatable {
  return a == b
}

// Associated type constraints
protocol Container {
  associatedtype Item: Equatable
  func contains(_ item: Item) -> Bool
}

// Protocol with Self constraint
protocol Copyable {
  func copy() -> Self
}

Key Takeaways

• Type erasure vs monomorphization -- Java and TypeScript erase generics at runtime (one shared representation); C++, Rust, Swift, and Go generate specialized code per type (monomorphization) for zero-cost abstraction.
• Constraint systems vary in expressiveness -- Rust's traits and C++ concepts offer rich bounds; Go's type sets (~int | ~float64) and Swift's protocol constraints are simpler; Java's extends and wildcards are more limited.
• Languages without generics use workarounds -- C uses void* and macros; JavaScript/Ruby rely on duck typing; PHP uses docblock annotations for static analysis; Zig uses comptime type parameters.
• Comptime generics (Zig) are unique -- Zig passes types as compile-time values, enabling zero-cost generics without a separate type parameter syntax; types are first-class at compile time.
• Python and PHP have late generics -- Python 3.12+ and PHPStan/Psalm add generic syntax; type checking is static only—runtime behavior is unchanged—unlike compiled languages.