Float.Array(3)

📦
version 5.5.0 · 19 versions
, in arch rolling

🦋 Float.Array(3) — OCaml library 🦋

okay bestie, welcome to the OCaml library section, page Float.Array(3) — this one is all about float arrays with a packed representation, and it is lowkey iconic for numerical code. let's get into it 💅

💖 NAME 💖

Float.Array — Float arrays with packed representation. that's the whole vibe, no cap.

📦 Module 📦

the module name in the source is literally:

Module   Float.Array

so yeah, this page is documenting Float.Array. 🫶

🌸 Documentation 🌸

inside the docs there's this stub:

Module Array
        : sig end

and then the real content starts:

Float arrays with packed representation. ✨

type t = floatarray

this is the type of "flat" float arrays. 🎀

a floatarray is an array data structure that contains a fixed number of elements of type float. the elements of a floatarray are guaranteed to be stored contiguously in memory, without any boxing. that's the whole point — it's packed and tidy. 💖

currently, the built-in type float array is optimized by default to also use an unboxed representation. however, this optimization may be disabled at configure-time when building the compiler. furthermore, operations on floatarray are a bit more efficient than those on float array, which require an extra dynamic check. so floatarray is giving main character energy in the efficiency department. 👑

array literals of type floatarray can be written using the same syntax as those of type float array as long as there is enough typing information in the environment so that the compiler can disambiguate between the two cases. see Section 12.25 'Type-directed disambiguation of array literals' of the manual for more details. 🧚

Since 4.08 ⭐

  • val length : t -> int — returns the length (number of elements) of the given floatarray. literally just the count, bestie. 🔢
  • val get : t -> int -> float — get a n returns the element number n of floatarray a. Raises Invalid_argument if n is outside the range 0 to (length a - 1). don't try to be cute with out-of-bounds indices, it will not slay. 💀
  • val set : t -> int -> float -> unit — set a n x modifies floatarray a in place, replacing element number n with x. Raises Invalid_argument if n is outside the range 0 to (length a - 1). same boundaries, same drama. 🌈
  • val make : int -> float -> t — make n x returns a fresh floatarray of length n, initialized with x. Raises Invalid_argument if n < 0 or n > Sys.max_floatarray_length. 🍓
  • val create : int -> t — create n returns a fresh floatarray of length n, with uninitialized data. Raises Invalid_argument if n < 0 or n > Sys.max_floatarray_length. uninitialized means exactly that — don't assume it's zeroed, bestie. 🙈
  • val init : int -> (int -> float) -> t — init n f returns a fresh floatarray 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 to the integers 0 to n-1. Raises Invalid_argument if n < 0 or n > Sys.max_floatarray_length. 🎀
  • val make_matrix : int -> int -> float -> t array — make_matrix dimx dimy e returns a two-dimensional array (an array of arrays) with first dimension dimx and second dimension dimy, where all elements are initialized with e. Since 5.2. Raises Invalid_argument if dimx or dimy is negative or greater than Sys.max_floatarray_length. ✨
  • val init_matrix : int -> int -> (int -> int -> float) -> t 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. Since 5.2. Raises Invalid_argument if dimx or dimy is negative or greater than Sys.max_floatarray_length. 🦋
  • val append : t -> t -> t — append v1 v2 returns a fresh floatarray containing the concatenation of the floatarrays v1 and v2. Raises Invalid_argument if length v1 + length v2 > Sys.max_floatarray_length. 💜
  • val concat : t list -> t — same as Float.Array.append, but concatenates a list of floatarrays. group-chat merge, but for arrays. 🫶
  • val sub : t -> int -> int -> t — sub a pos len returns a fresh floatarray of length len, containing the elements number pos to pos + len - 1 of floatarray a. 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. slice responsibly. 🔥
  • val copy : t -> t — copy a returns a copy of a, that is, a fresh floatarray containing the same elements as a. not the same object, bestie — a fresh one. 💖
  • val fill : t -> int -> int -> float -> unit — fill a pos len x modifies the floatarray 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. 🎨
  • val blit : t -> int -> t -> int -> int -> unit — blit src src_pos dst dst_pos len copies len elements from floatarray src, starting at element number src_pos, to floatarray dst, starting at element number dst_pos. it works correctly even if src and dst are the same floatarray, and the source and destination chunks overlap. overlap? handled. slay. ✨ 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. 💀
  • val to_list : t -> float list — to_list a returns the list of all the elements of a. 📋
  • val of_list : float list -> t — of_list l returns a fresh floatarray containing the elements of l. Raises Invalid_argument if the length of l is greater than Sys.max_floatarray_length. 🍓

💅 Comparison 💅

  • val equal : (float -> float -> bool) -> t -> t -> 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. Since 5.4 🌸
  • val compare : (float -> float -> int) -> t -> t -> 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 🦋

  • val iter : (float -> unit) -> t -> 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); (). yes, including that final (), bestie. 💅
  • val iteri : (int -> float -> unit) -> t -> unit — same as Float.Array.iter, but the function is applied with the index of the element as first argument, and the element itself as second argument. index first, element second. iykyk. 🔢
  • val map : (float -> float) -> t -> t — map f a applies function f to all the elements of a, and builds a floatarray with the results returned by f. ✨
  • val map_inplace : (float -> float) -> t -> unit — map_inplace f a applies function f to all elements of a, and updates their values in place. Since 5.1 💖
  • val mapi : (int -> float -> float) -> t -> t — same as Float.Array.map, but the function is applied to the index of the element as first argument, and the element itself as second argument. 🎀
  • val mapi_inplace : (int -> float -> float) -> t -> unit — same as Float.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 🫶
  • val fold_left : ('acc -> float -> 'acc) -> 'acc -> t -> '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 floatarray a. left fold go brrr. 🔥
  • val fold_right : (float -> 'acc -> 'acc) -> t -> '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 floatarray a. right fold, same energy, other direction. 💜

🌈 Iterators on two arrays 🌈

  • val iter2 : (float -> float -> unit) -> t -> t -> unit — iter2 f a b applies function f to all the elements of a and b. Raises Invalid_argument if the floatarrays are not the same size. matching lengths only, bestie. 💀
  • val map2 : (float -> float -> float) -> t -> t -> t — map2 f a b applies function f to all the elements of a and b, and builds a floatarray with the results returned by f: [| f a.(0) b.(0); ...; f a.(length a - 1) b.(length b - 1)|]. Raises Invalid_argument if the floatarrays are not the same size. ✨

🔍 Array scanning 🔍

  • val for_all : (float -> bool) -> t -> bool — for_all f [|a1; ...; an|] checks if all elements of the floatarray satisfy the predicate f. that is, it returns (f a1) && (f a2) && ... && (f an). all of them, or it's a no from me. ✅
  • val exists : (float -> bool) -> t -> bool — exists f [|a1; ...; an|] checks if at least one element of the floatarray satisfies the predicate f. that is, it returns (f a1) || (f a2) || ... || (f an). just one is enough to make it true, periodt. 🔥
  • val mem : float -> t -> bool — mem a set is true if and only if there is an element of set that is structurally equal to a, i.e. there is an x in set such that compare a x = 0. structural equality check, not IEEE. 🧐
  • val mem_ieee : float -> t -> bool — same as Float.Array.mem, but uses IEEE equality instead of structural equality. important distinction, bestie. ⚡

🎀 Array searching 🎀

  • val find_opt : (float -> bool) -> t -> float option — signature only in the original, so here it is exactly, no invented extras. 🫶
  • val find_index : (float -> bool) -> t -> 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 🔎
  • val find_map : (float -> 'a option) -> t -> 'a option — signature only in the original. keeping it raw and exact. ✨
  • val find_mapi : (int -> float -> 'a option) -> t -> 'a 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. Since 5.1 💖

👑 Sorting and shuffling 👑

  • val sort : (float -> float -> int) -> t -> unit — sort a floatarray 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. ⭐

the current implementation uses Heap Sort. it runs in constant stack space. no cap. 🔥

Specification of the comparison function: let a be the floatarray 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

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

that's the full spec, bestie — don't skip the inequalities. 💀

  • val stable_sort : (float -> float -> int) -> t -> unit — same as Float.Array.sort, but the sorting algorithm is stable (i.e. elements that compare equal are kept in their original order) and not guaranteed to run in constant heap space. the current implementation uses Merge Sort. it uses a temporary floatarray of length n/2, where n is the length of the floatarray. it is usually faster than the current implementation of Float.Array.sort. stable and speedy? we love to see it. 🦋
  • val fast_sort : (float -> float -> int) -> t -> unit — same as Float.Array.sort or Float.Array.stable_sort, whichever is faster on typical input. it's giving "pick the faster one" energy. ⚡
  • val shuffle : rand:(int -> int) -> t -> unit — shuffle rand a randomly permutes a's elements using rand for randomness. the distribution of permutations is uniform. 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). Since 5.2 🎲

🌊 Float arrays and Sequences 🌊

  • val to_seq : t -> float Seq.t — signature only in the original, kept exactly as-is. 🫶

Iterate on the floatarray, in increasing order. Modifications of the floatarray during iteration will be reflected in the sequence. ✨

  • val to_seqi : t -> (int * float) Seq.t — iterate on the floatarray, in increasing order, yielding indices along elements. Modifications of the floatarray during iteration will be reflected in the sequence. 🦋
  • val of_seq : float Seq.t -> t — create an array from the generator. 🎀
  • val map_to_array : (float -> 'a) -> t -> 'a array — map_to_array 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) |]. 💅
  • val map_from_array : ('a -> float) -> 'a array -> t — map_from_array f a applies function f to all the elements of a, and builds a floatarray with the results returned by f. 🌸

🌈 Arrays and concurrency safety 🌈

Care must be taken when concurrently accessing float arrays from multiple domains: accessing an array will never crash a program, but unsynchronized accesses might yield surprising (non-sequentially-consistent) results. 👑

⚛️ Atomicity ⚛️

Every float array operation that accesses more than one array element is not atomic. This includes iteration, scanning, sorting, splitting and combining arrays. 🫶

For example, consider the following program:

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

After executing this code, each field of the float 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). 🔥

💥 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. 🧚

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. 😭

😭 Tearing 😭

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. 🍓

For instance, at the end of

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

the res float array might contain values that are neither 0. nor max_float. 🙈

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. ⭐

That's the whole page, bestie! Footer vibes: OCamldoc, 2026-06-22, Float.Array(3) — slay and stay safe out there 💖

📦 where this page comes from