Stdlib.Array(3)

🌸 NAME 🌸

okay so welcome to the manual page for Stdlib.Array from the OCaml library, bestie 💖. The official name line is literally Stdlib.Array - no description — mysterious, iconic, no notes 💀✨. Anyway, it's array time.

🧚 Module 🧚

Module — Stdlib.Array

✨ Documentation ✨

🦋 Module Array 🦋

Module Array
 : (module Stdlib__Array)

🎀 type 'a t 🎀

type 'a t = 'a array

An alias for the type of arrays. 🫶

💖 length 💖

val length : 'a array -> int

Returns the length (number of elements) of the given array. Slay. ✨

🍓 get 🍓

val get : 'a array -> int -> 'a

get a n returns the element number n of array a. The first element has number 0. The last element has number length a - 1. You can also write a.(n) instead of get a n. 💅

Raises Invalid_argument if n is outside the range 0 to (length a - 1).

🎀 set 🎀

val set : 'a array -> int -> 'a -> unit

set a n x modifies array a in place, replacing element number n with x. You can also write a.(n) <- x instead of set a n x. 🫶

Raises Invalid_argument if n is outside the range 0 to length a - 1.

🌈 make 🌈

val make : int -> 'a -> 'a array

make n x returns a fresh array of length n, initialized with x. All the elements of this new array are initially physically equal to x (in the sense of the == predicate). Consequently, if x is mutable, it is shared among all elements of the array, and modifying x through one of the array entries will modify all other entries at the same time. That's the tea, bestie. ☕

Raises Invalid_argument if n < 0 or n > Sys.max_array_length. If the value of x is a floating-point number, then the maximum size is only Sys.max_array_length / 2. 🫠

🦋 create_float 🦋

val create_float : int -> float array

create_float n returns a fresh float array of length n, with uninitialized data. ✨

✨ Since 4.03

🧚 init 🧚

val init : int -> (int -> 'a) -> 'a array

init n f returns a fresh array of length n, with element number i initialized to the result of f i. In other terms, init n f tabulates the results of f applied in order to the integers 0 to n-1. 💖

Raises Invalid_argument if n < 0 or n > Sys.max_array_length. If the return type of f is float, then the maximum size is only Sys.max_array_length / 2.

💜 make_matrix 💜

val make_matrix : int -> int -> 'a -> 'a array array

make_matrix dimx dimy e returns a two-dimensional array (an array of arrays) with first dimension dimx and second dimension dimy. All the elements of this new matrix are initially physically equal to e. The element (x,y) of a matrix m is accessed with the notation m.(x).(y). Matrix main character energy, honestly. 👑

Raises Invalid_argument if dimx or dimy is negative or greater than Sys.max_array_length. If the value of e is a floating-point number, then the maximum size is only Sys.max_array_length / 2.

🎀 init_matrix 🎀

val init_matrix : int -> int -> (int -> int -> 'a) -> 'a array array

init_matrix dimx dimy f returns a two-dimensional array (an array of arrays) with first dimension dimx and second dimension dimy, where the element at index (x,y) is initialized with f x y. The element (x,y) of a matrix m is accessed with the notation m.(x).(y). 🦋

✨ Since 5.2

Raises Invalid_argument if dimx or dimy is negative or greater than Sys.max_array_length. If the return type of f is float, then the maximum size is only Sys.max_array_length / 2.

🍓 append 🍓

val append : 'a array -> 'a array -> 'a array

append v1 v2 returns a fresh array containing the concatenation of the arrays v1 and v2. It's giving "put it together, bestie." 💅

Raises Invalid_argument if length v1 + length v2 > Sys.max_array_length.

🌸 concat 🌸

val concat : 'a array list -> 'a array

Same as Array.append, but concatenates a list of arrays. ✨

🧚 sub 🧚

val sub : 'a array -> int -> int -> 'a array

sub a pos len returns a fresh array of length len, containing the elements number pos to pos + len - 1 of array a. Little slice of array life. 🍰

Raises Invalid_argument if pos and len do not designate a valid subarray of a; that is, if pos < 0, or len < 0, or pos + len > length a.

🎀 copy 🎀

val copy : 'a array -> 'a array

copy a returns a copy of a, that is, a fresh array containing the same elements as a. Dupe era, but legally. 💖

🦋 fill 🦋

val fill : 'a array -> int -> int -> 'a -> unit

fill a pos len x modifies the array a in place, storing x in elements number pos to pos + len - 1. 💅

Raises Invalid_argument if pos and len do not designate a valid subarray of a.

🌈 blit 🌈

val blit : 'a array -> int -> 'a array -> int -> int -> unit

blit src src_pos dst dst_pos len copies len elements from array src, starting at element number src_pos, to array dst, starting at element number dst_pos. It works correctly even if src and dst are the same array, and the source and destination chunks overlap. Overlap? Handled. Periodt. ✨

Raises Invalid_argument if src_pos and len do not designate a valid subarray of src, or if dst_pos and len do not designate a valid subarray of dst.

🍓 to_list 🍓

val to_list : 'a array -> 'a list

to_list a returns the list of all the elements of a. Array to list pipeline, slay. 💖

🧚 of_list 🧚

val of_list : 'a list -> 'a array

of_list l returns a fresh array containing the elements of l. List to array pipeline, also slay. 🫶

Raises Invalid_argument if the length of l is greater than Sys.max_array_length.

---

💖 Comparison 💖

🌸 equal 🌸

val equal : ('a -> 'a -> bool) -> 'a array -> 'a array -> bool

equal eq a b is true if and only if a and b have the same length n and for all i in [ 0 ; n-1 ], eq a.(i) b.(i) is true. Matching vibes only. ✨

✨ Since 5.4

🦋 compare 🦋

val compare : ('a -> 'a -> int) -> 'a array -> 'a array -> int

compare cmp a b compares a and b according to the shortlex order, that is, shorter arrays are smaller and equal-sized arrays are compared in lexicographic order using cmp to compare elements. 💅

✨ Since 5.4

---

🎀 Iterators 🎀

💖 iter 💖

val iter : ('a -> unit) -> 'a array -> unit

iter f a applies function f in turn to all the elements of a. It is equivalent to f a.(0); f a.(1); ...; f a.(length a - 1); (). Go off, iterator. 🔥

🍓 iteri 🍓

val iteri : (int -> 'a -> unit) -> 'a array -> unit

Same as Array.iter, but the function is applied to the index of the element as first argument, and the element itself as second argument. Index girlies rise. 🫶

🌈 map 🌈

val map : ('a -> 'b) -> 'a array -> 'b array

map f a applies function f to all the elements of a, and builds an array with the results returned by f: [| f a.(0); f a.(1); ...; f a.(length a - 1) |]. Transformation era. ✨

🦋 map_inplace 🦋

val map_inplace : ('a -> 'a) -> 'a array -> unit

map_inplace f a applies function f to all elements of a, and updates their values in place. No new array, just glow-up. 💅

✨ Since 5.1

🧚 mapi 🧚

val mapi : (int -> 'a -> 'b) -> 'a array -> 'b array

Same as Array.map, but the function is applied to the index of the element as first argument, and the element itself as second argument. 💖

🎀 mapi_inplace 🎀

val mapi_inplace : (int -> 'a -> 'a) -> 'a array -> unit

Same as Array.map_inplace, but the function is applied to the index of the element as first argument, and the element itself as second argument. 💅

✨ Since 5.1

💜 fold_left 💜

val fold_left : ('acc -> 'a -> 'acc) -> 'acc -> 'a array -> 'acc

fold_left f init a computes f (... (f (f init a.(0)) a.(1)) ...) a.(n-1), where n is the length of the array a. Accumulator bestie behavior. 🫶

🌸 fold_left_map 🌸

val fold_left_map : ('acc -> 'a -> 'acc * 'b) -> 'acc -> 'a array -> 'acc * 'b array

fold_left_map is a combination of Array.fold_left and Array.map that threads an accumulator through calls to f. Multi-tasking icon. ✨

✨ Since 4.13

🦋 fold_right 🦋

val fold_right : ('a -> 'acc -> 'acc) -> 'a array -> 'acc -> 'acc

fold_right f a init computes f a.(0) (f a.(1) ( ... (f a.(n-1) init) ...)), where n is the length of the array a. Right-to-left vibe check. 💅

---

🍓 Iterators on two arrays 🍓

🌈 iter2 🌈

val iter2 : ('a -> 'b -> unit) -> 'a array -> 'b array -> unit

iter2 f a b applies function f to all the elements of a and b. Duo energy. 💖

✨ Since 4.03 (4.05 in ArrayLabels)

Raises Invalid_argument if the arrays are not the same size.

🧚 map2 🧚

val map2 : ('a -> 'b -> 'c) -> 'a array -> 'b array -> 'c array

map2 f a b applies function f to all the elements of a and b, and builds an array with the results returned by f: [| f a.(0) b.(0); ...; f a.(length a - 1) b.(length b - 1)|]. Two-array combo meal. 🍓

✨ Since 4.03 (4.05 in ArrayLabels)

Raises Invalid_argument if the arrays are not the same size.

---

🎀 Array scanning 🎀

💖 for_all 💖

val for_all : ('a -> bool) -> 'a array -> bool

for_all f [|a1; ...; an|] checks if all elements of the array satisfy the predicate f. That is, it returns (f a1) && (f a2) && ... && (f an). All-clear energy. ✨

✨ Since 4.03

🦋 exists 🦋

val exists : ('a -> bool) -> 'a array -> bool

exists f [|a1; ...; an|] checks if at least one element of the array satisfies the predicate f. That is, it returns (f a1) || (f a2) || ... || (f an). One is enough, bestie. 💅

✨ Since 4.03

🌸 for_all2 🌸

val for_all2 : ('a -> 'b -> bool) -> 'a array -> 'b array -> bool

Same as Array.for_all, but for a two-argument predicate. 💖

✨ Since 4.11

Raises Invalid_argument if the two arrays have different lengths.

🧚 exists2 🧚

val exists2 : ('a -> 'b -> bool) -> 'a array -> 'b array -> bool

Same as Array.exists, but for a two-argument predicate. 🫶

✨ Since 4.11

Raises Invalid_argument if the two arrays have different lengths.

🍓 mem 🍓

val mem : 'a -> 'a array -> bool

mem a set is true if and only if a is structurally equal to an element of set (i.e. there is an x in set such that compare a x = 0). Are you in the group chat or not. 💅

✨ Since 4.03

🎀 memq 🎀

val memq : 'a -> 'a array -> bool

Same as Array.mem, but uses physical equality instead of structural equality to compare array elements. Same vibe, different equality check. ✨

✨ Since 4.03

🌈 find_opt 🌈

val find_opt : ('a -> bool) -> 'a array -> 'a option

find_opt f a returns the first element of the array a that satisfies the predicate f, or None if there is no value that satisfies f in the array a. First match wins. 💖

✨ Since 4.13

🦋 find_index 🦋

val find_index : ('a -> bool) -> 'a array -> int option

find_index f a returns Some i, where i is the index of the first element of the array a that satisfies f x, if there is such an element.

It returns None if there is no such element. 🔥

✨ Since 5.1

💜 find_map 💜

val find_map : ('a -> 'b option) -> 'a array -> 'b option

find_map f a applies f to the elements of a in order, and returns the first result of the form Some v, or None if none exist. First Some takes the crown. 👑

✨ Since 4.13

🧚 find_mapi 🧚

val find_mapi : (int -> 'a -> 'b option) -> 'a array -> 'b option

Same as find_map, but the predicate is applied to the index of the element as first argument (counting from 0), and the element itself as second argument. Index-aware searching, we love to see it. 💅

✨ Since 5.1

---

💖 Arrays of pairs 💖

🌸 split 🌸

val split : ('a * 'b) array -> 'a array * 'b array

split [|(a1,b1); ...; (an,bn)|] is ([|a1; ...; an|], [|b1; ...; bn|]). Divide and slay. ✨

✨ Since 4.13

🍓 combine 🍓

val combine : 'a array -> 'b array -> ('a * 'b) array

combine [|a1; ...; an|] [|b1; ...; bn|] is [|(a1,b1); ...; (an,bn)|]. Raise Invalid_argument if the two arrays have different lengths. Pair them up, bestie. 💅

✨ Since 4.13

---

🎀 Sorting and shuffling 🎀

🔥 sort 🔥

val sort : ('a -> 'a -> int) -> 'a array -> unit

Sort an array in increasing order according to a comparison function. The comparison function must return 0 if its arguments compare as equal, a positive integer if the first is greater, and a negative integer if the first is smaller (see below for a complete specification). For example, compare is a suitable comparison function. After calling sort, the array is sorted in place in increasing order. sort is guaranteed to run in constant heap space and (at most) logarithmic stack space. No cap, that's a whole guarantee. 🔥👑

ok so back to sort — the current implementation uses Heap Sort, and it runs in constant stack space ✨ no cap, no growing stack, we love a low-maintenance icon 💅

🎀 Specification of the comparison function 🎀

ok this part is important so pay attention bestie 🫶 Let a be the array and cmp the comparison function. The following must be true for all x, y, z in a:

• cmp x y > 0 if and only if cmp y x < 0
• if cmp x y >= 0 and cmp y z >= 0 then cmp x z >= 0

Like, it's giving "be a well-behaved ordering or else" 😭 it's transitive and antisymmetric and we respect that.

When sort returns, a contains the same elements as before, reordered in such a way that for all i and j valid indices of a:

• cmp a.(i) a.(j) >= 0 if i >= j

---

🌸 stable_sort 🌸

val stable_sort : ('a -> 'a -> int) -> 'a array -> unit

Same as Array.sort, but the sorting algorithm is stable — meaning elements that compare equal are kept in their original order 💖 — and it's not guaranteed to run in constant heap space (so it's not giving the same memory discipline as plain sort, fyi).

The current implementation uses Merge Sort. It uses a temporary array of length n/2, where n is the length of the array. It is usually faster than the current implementation of Array.sort 🔥 so like, yeah, kind of a glow-up.

---

🦋 stable_sort_sub 🦋

val stable_sort_sub : ('a -> 'a -> int) -> 'a array -> int -> int -> unit

stable_sort_sub cmp a pos len sorts the subarray of the array a delimited by the start position pos and by the length len. The data in this subarray is sorted in increasing order according to the comparison function cmp. The data outside of this subarray is unaffected — untouched, unbothered, not on the mood board 🙈

The sorting algorithm is stable; it is the same as in Array.stable_sort.

Since 5.5

Raises Invalid_argument if pos and len do not designate a valid subarray of a; that is, if pos < 0, or len < 0, or pos + len > length a. 💀 so like, bounds-check your args bestie, it's not gonna be nice about it.

---

⭐ fast_sort ⭐

val fast_sort : ('a -> 'a -> int) -> 'a array -> unit

Same as Array.sort or Array.stable_sort, whichever is faster on typical input. The laziness icon, honestly 👑

---

🍓 shuffle 🍓

val shuffle : rand:(int -> int) -> 'a array -> unit

shuffle rand a randomly permutes a's elements using rand for randomness. The distribution of permutations is uniform ✨ (every ordering equally likely, we stan fairness).

rand must be such that a call to rand n returns a uniformly distributed random number in the range [0; n-1]. Random.int can be used for this — do not forget to Random.self_init the generator 🫶 (yes, actually do it, don't just copy the line and forget, iykyk).

Since 5.2

---

🌈 Arrays and Sequences 🌈

to_seq

val to_seq : 'a array -> 'a Seq.t

Iterate on the array, in increasing order. Modifications of the array during iteration will be reflected in the sequence — so it's live, it's not a snapshot, keep that in mind 💅

Since 4.07

to_seqi

val to_seqi : 'a array -> (int * 'a) Seq.t

Iterate on the array, in increasing order, yielding indices along elements. Modifications of the array during iteration will be reflected in the sequence. Same live-energy as above, just with the index attached 🎀

Since 4.07

of_seq

val of_seq : 'a Seq.t -> 'a array

Create an array from the generator ✨

Since 4.07

---

💜 Arrays and concurrency safety 💜

Care must be taken when concurrently accessing arrays from multiple domains, bestie: accessing an array will never crash a program, but unsynchronized accesses might yield surprising (non-sequentially-consistent) results. It's not gonna blow up in your face, it's just gonna be weird 😭

🔥 Atomicity 🔥

Every array operation that accesses more than one array element is not atomic. This includes iteration, scanning, sorting, splitting and combining arrays — all of them, no exceptions, periodt.

For example, consider the following program:

let size = 100_000_000
let a = Array.make size 1
let d1 = Domain.spawn (fun () ->
   Array.iteri (fun i x -> a.(i) <- x + 1) a
)
let d2 = Domain.spawn (fun () ->
  Array.iteri (fun i x -> a.(i) <- 2 * x + 1) a
)
let () = Domain.join d1; Domain.join d2

After executing this code, each field of the array a is either 2, 3, 4 or 5. If atomicity is required, then the user must implement their own synchronization (for example, using Mutex.t) 🫶 the stdlib is not gonna do it for you, that's your main character task.

💀 Data races 💀

If two domains only access disjoint parts of the array, then the observed behaviour is the equivalent to some sequential interleaving of the operations from the two domains. So like, if you're staying in your own lane, you're fine ✨

A data race is said to occur when two domains access the same array element without synchronization and at least one of the accesses is a write. In the absence of data races, the observed behaviour is equivalent to some sequential interleaving of the operations from different domains.

Whenever possible, data races should be avoided by using synchronization to mediate the accesses to the array elements.

Indeed, in the presence of data races, programs will not crash but the observed behaviour may not be equivalent to any sequential interleaving of operations from different domains. Nevertheless, even in the presence of data races, a read operation will return the value of some prior write to that location (with a few exceptions for float arrays 👀). So it's not chaos, it's just... noncommittal.

🧚 Float arrays 🧚

Float arrays have two supplementary caveats in the presence of data races.

First, the blit operation might copy an array byte-by-byte. Data races between such a blit operation and another operation might produce surprising values due to tearing: partial writes interleaved with other operations can create float values that would not exist with a sequential execution 😭 like, values that literally have no business existing.

For instance, at the end of

let zeros = Array.make size 0.
let max_floats = Array.make size Float.max_float
let res = Array.copy zeros
let d1 = Domain.spawn (fun () -> Array.blit zeros 0 res 0 size)
let d2 = Domain.spawn (fun () -> Array.blit max_floats 0 res 0 size)
let () = Domain.join d1; Domain.join d2

the res array might contain values that are neither 0. nor max_float. Yeah. Not a vibe, but that's the deal 🙈

Second, on 32-bit architectures, getting or setting a field involves two separate memory accesses. In the presence of data races, the user may observe tearing on any operation. So on 32-bit, stay synced, fr fr 💅

---

and that's the whole page bestie 🎀 this one came from OCamldoc, dated 2026-06-22, and it documents Stdlib.Array(3) — go forth, sort responsibly, and don't race your data 🌸🫶✨