1BytesLabels(3) OCaml library BytesLabels(3)
2
6 BytesLabels - Byte sequence operations.
7
9 Module BytesLabels
10
12 Module BytesLabels
13 : sig end
14 Byte sequence operations.
17 A byte sequence is a mutable data structure that contains a
19 fixed-length sequence of bytes. Each byte can be indexed in constant
20 time for reading or writing.
21 Given a byte sequence s of length l , we can access each of the l bytes
23 of s via its index in the sequence. Indexes start at 0 , and we will
24 call an index valid in s if it falls within the range [0...l-1] (inclu‐
25 sive). A position is the point between two bytes or at the beginning or
26 end of the sequence. We call a position valid in s if it falls within
27 the range [0...l] (inclusive). Note that the byte at index n is between
28 positions n and n+1 .
29 Two parameters start and len are said to designate a valid range of s
31 if len >= 0 and start and start+len are valid positions in s .
32 Byte sequences can be modified in place, for instance via the set and
34 blit functions described below. See also strings (module String ),
35 which are almost the same data structure, but cannot be modified in
36 place.
37 Bytes are represented by the OCaml type char .
39 The labeled version of this module can be used as described in the Std‐
41 Labels module.
42 Since 4.02.0
45 val length : bytes -> int
52 Return the length (number of bytes) of the argument.
54 val get : bytes -> int -> char
58 get s n returns the byte at index n in argument s .
61 Raises Invalid_argument if n is not a valid index in s .
64 val set : bytes -> int -> char -> unit
68 set s n c modifies s in place, replacing the byte at index n with c .
71 Raises Invalid_argument if n is not a valid index in s .
74 val create : int -> bytes
78 create n returns a new byte sequence of length n . The sequence is
81 uninitialized and contains arbitrary bytes.
82 Raises Invalid_argument if n < 0 or n > Sys.max_string_length .
85 val make : int -> char -> bytes
89 make n c returns a new byte sequence of length n , filled with the byte
92 c .
93 Raises Invalid_argument if n < 0 or n > Sys.max_string_length .
96 val init : int -> f:(int -> char) -> bytes
100 init n f returns a fresh byte sequence of length n , with character i
103 initialized to the result of f i (in increasing index order).
104 Raises Invalid_argument if n < 0 or n > Sys.max_string_length .
107 val empty : bytes
111 A byte sequence of size 0.
113 val copy : bytes -> bytes
117 Return a new byte sequence that contains the same bytes as the argu‐
119 ment.
120 val of_string : string -> bytes
124 Return a new byte sequence that contains the same bytes as the given
126 string.
127 val to_string : bytes -> string
131 Return a new string that contains the same bytes as the given byte se‐
133 quence.
134 val sub : bytes -> pos:int -> len:int -> bytes
138 sub s ~pos ~len returns a new byte sequence of length len , containing
141 the subsequence of s that starts at position pos and has length len .
142 Raises Invalid_argument if pos and len do not designate a valid range
145 of s .
146 val sub_string : bytes -> pos:int -> len:int -> string
150 Same as BytesLabels.sub but return a string instead of a byte sequence.
152 val extend : bytes -> left:int -> right:int -> bytes
156 extend s ~left ~right returns a new byte sequence that contains the
159 bytes of s , with left uninitialized bytes prepended and right unini‐
160 tialized bytes appended to it. If left or right is negative, then bytes
161 are removed (instead of appended) from the corresponding side of s .
162 Since 4.05.0 in BytesLabels
165 Raises Invalid_argument if the result length is negative or longer than
168 Sys.max_string_length bytes.
169 val fill : bytes -> pos:int -> len:int -> char -> unit
173 fill s ~pos ~len c modifies s in place, replacing len characters with c
176 , starting at pos .
177 Raises Invalid_argument if pos and len do not designate a valid range
180 of s .
181 val blit : src:bytes -> src_pos:int -> dst:bytes -> dst_pos:int ->
185 len:int -> unit
186 blit ~src ~src_pos ~dst ~dst_pos ~len copies len bytes from sequence
189 src , starting at index src_pos , to sequence dst , starting at index
190 dst_pos . It works correctly even if src and dst are the same byte se‐
191 quence, and the source and destination intervals overlap.
192 Raises Invalid_argument if src_pos and len do not designate a valid
195 range of src , or if dst_pos and len do not designate a valid range of
196 dst .
197 val blit_string : src:string -> src_pos:int -> dst:bytes -> dst_pos:int
201 -> len:int -> unit
202 blit ~src ~src_pos ~dst ~dst_pos ~len copies len bytes from string src
205 , starting at index src_pos , to byte sequence dst , starting at index
206 dst_pos .
207 Since 4.05.0 in BytesLabels
210 Raises Invalid_argument if src_pos and len do not designate a valid
213 range of src , or if dst_pos and len do not designate a valid range of
214 dst .
215 val concat : sep:bytes -> bytes list -> bytes
219 concat ~sep sl concatenates the list of byte sequences sl , inserting
222 the separator byte sequence sep between each, and returns the result as
223 a new byte sequence.
224 Raises Invalid_argument if the result is longer than
227 Sys.max_string_length bytes.
228 val cat : bytes -> bytes -> bytes
232 cat s1 s2 concatenates s1 and s2 and returns the result as a new byte
235 sequence.
236 Since 4.05.0 in BytesLabels
239 Raises Invalid_argument if the result is longer than
242 Sys.max_string_length bytes.
243 val iter : f:(char -> unit) -> bytes -> unit
247 iter ~f s applies function f in turn to all the bytes of s . It is
250 equivalent to f (get s 0); f (get s 1); ...; f (get s
251 (length s - 1)); () .
252 val iteri : f:(int -> char -> unit) -> bytes -> unit
256 Same as BytesLabels.iter , but the function is applied to the index of
258 the byte as first argument and the byte itself as second argument.
259 val map : f:(char -> char) -> bytes -> bytes
263 map ~f s applies function f in turn to all the bytes of s (in increas‐
266 ing index order) and stores the resulting bytes in a new sequence that
267 is returned as the result.
268 val mapi : f:(int -> char -> char) -> bytes -> bytes
272 mapi ~f s calls f with each character of s and its index (in increasing
275 index order) and stores the resulting bytes in a new sequence that is
276 returned as the result.
277 val fold_left : f:('a -> char -> 'a) -> init:'a -> bytes -> 'a
281 fold_left f x s computes f (... (f (f x (get s 0)) (get s 1)) ...) (get
284 s (n-1)) , where n is the length of s .
285 Since 4.13.0
288 val fold_right : f:(char -> 'a -> 'a) -> bytes -> init:'a -> 'a
292 fold_right f s x computes f (get s 0) (f (get s 1) ( ... (f (get s
295 (n-1)) x) ...)) , where n is the length of s .
296 Since 4.13.0
299 val for_all : f:(char -> bool) -> bytes -> bool
303 for_all p s checks if all characters in s satisfy the predicate p .
306 Since 4.13.0
309 val exists : f:(char -> bool) -> bytes -> bool
313 exists p s checks if at least one character of s satisfies the predi‐
316 cate p .
317 Since 4.13.0
320 val trim : bytes -> bytes
324 Return a copy of the argument, without leading and trailing whitespace.
326 The bytes regarded as whitespace are the ASCII characters ' ' , '\012'
327 , '\n' , '\r' , and '\t' .
328 val escaped : bytes -> bytes
332 Return a copy of the argument, with special characters represented by
334 escape sequences, following the lexical conventions of OCaml. All
335 characters outside the ASCII printable range (32..126) are escaped, as
336 well as backslash and double-quote.
337 Raises Invalid_argument if the result is longer than
340 Sys.max_string_length bytes.
341 val index : bytes -> char -> int
345 index s c returns the index of the first occurrence of byte c in s .
348 Raises Not_found if c does not occur in s .
351 val index_opt : bytes -> char -> int option
355 index_opt s c returns the index of the first occurrence of byte c in s
358 or None if c does not occur in s .
359 Since 4.05
362 val rindex : bytes -> char -> int
366 rindex s c returns the index of the last occurrence of byte c in s .
369 Raises Not_found if c does not occur in s .
372 val rindex_opt : bytes -> char -> int option
376 rindex_opt s c returns the index of the last occurrence of byte c in s
379 or None if c does not occur in s .
380 Since 4.05
383 val index_from : bytes -> int -> char -> int
387 index_from s i c returns the index of the first occurrence of byte c in
390 s after position i . index s c is equivalent to index_from s 0 c .
391 Raises Invalid_argument if i is not a valid position in s .
394 Raises Not_found if c does not occur in s after position i .
397 val index_from_opt : bytes -> int -> char -> int option
401 index_from_opt s i c returns the index of the first occurrence of byte
404 c in s after position i or None if c does not occur in s after position
405 i . index_opt s c is equivalent to index_from_opt s 0 c .
406 Since 4.05
409 Raises Invalid_argument if i is not a valid position in s .
412 val rindex_from : bytes -> int -> char -> int
416 rindex_from s i c returns the index of the last occurrence of byte c in
419 s before position i+1 . rindex s c is equivalent to rindex_from s
420 (length s - 1) c .
421 Raises Invalid_argument if i+1 is not a valid position in s .
424 Raises Not_found if c does not occur in s before position i+1 .
427 val rindex_from_opt : bytes -> int -> char -> int option
431 rindex_from_opt s i c returns the index of the last occurrence of byte
434 c in s before position i+1 or None if c does not occur in s before po‐
435 sition i+1 . rindex_opt s c is equivalent to rindex_from s (length s -
436 1) c .
437 Since 4.05
440 Raises Invalid_argument if i+1 is not a valid position in s .
443 val contains : bytes -> char -> bool
447 contains s c tests if byte c appears in s .
450 val contains_from : bytes -> int -> char -> bool
454 contains_from s start c tests if byte c appears in s after position
457 start . contains s c is equivalent to contains_from
458 s 0 c .
459 Raises Invalid_argument if start is not a valid position in s .
462 val rcontains_from : bytes -> int -> char -> bool
466 rcontains_from s stop c tests if byte c appears in s before position
469 stop+1 .
470 Raises Invalid_argument if stop < 0 or stop+1 is not a valid position
473 in s .
474 val uppercase_ascii : bytes -> bytes
478 Return a copy of the argument, with all lowercase letters translated to
480 uppercase, using the US-ASCII character set.
481 Since 4.05.0
484 val lowercase_ascii : bytes -> bytes
488 Return a copy of the argument, with all uppercase letters translated to
490 lowercase, using the US-ASCII character set.
491 Since 4.05.0
494 val capitalize_ascii : bytes -> bytes
498 Return a copy of the argument, with the first character set to upper‐
500 case, using the US-ASCII character set.
501 Since 4.05.0
504 val uncapitalize_ascii : bytes -> bytes
508 Return a copy of the argument, with the first character set to lower‐
510 case, using the US-ASCII character set.
511 Since 4.05.0
514 type t = bytes
517 An alias for the type of byte sequences.
520 val compare : t -> t -> int
524 The comparison function for byte sequences, with the same specification
526 as compare . Along with the type t , this function compare allows the
527 module Bytes to be passed as argument to the functors Set.Make and
528 Map.Make .
529 val equal : t -> t -> bool
533 The equality function for byte sequences.
535 Since 4.05.0
538 val starts_with : prefix:bytes -> bytes -> bool
542 starts_with ~prefix s is true if and only if s starts with prefix .
545 Since 4.13.0
548 val ends_with : suffix:bytes -> bytes -> bool
552 ends_with ~suffix s is true if and only if s ends with suffix .
555 Since 4.13.0
558 Unsafe conversions (for advanced users)
563 This section describes unsafe, low-level conversion functions between
564 bytes and string . They do not copy the internal data; used improperly,
565 they can break the immutability invariant on strings provided by the
566 -safe-string option. They are available for expert library authors, but
567 for most purposes you should use the always-correct BytesLa‐
568 bels.to_string and BytesLabels.of_string instead.
569 val unsafe_to_string : bytes -> string
571 Unsafely convert a byte sequence into a string.
573 To reason about the use of unsafe_to_string , it is convenient to con‐
575 sider an "ownership" discipline. A piece of code that manipulates some
576 data "owns" it; there are several disjoint ownership modes, including:
577 -Unique ownership: the data may be accessed and mutated
579 -Shared ownership: the data has several owners, that may only access
581 it, not mutate it.
582 Unique ownership is linear: passing the data to another piece of code
584 means giving up ownership (we cannot write the data again). A unique
585 owner may decide to make the data shared (giving up mutation rights on
586 it), but shared data may not become uniquely-owned again.
587 unsafe_to_string s can only be used when the caller owns the byte se‐
590 quence s -- either uniquely or as shared immutable data. The caller
591 gives up ownership of s , and gains ownership of the returned string.
592 There are two valid use-cases that respect this ownership discipline:
594 1. Creating a string by initializing and mutating a byte sequence that
596 is never changed after initialization is performed.
597 let string_init len f : string =
600 let s = Bytes.create len in
601 for i = 0 to len - 1 do Bytes.set s i (f i) done;
602 Bytes.unsafe_to_string s
603 This function is safe because the byte sequence s will never be ac‐
606 cessed or mutated after unsafe_to_string is called. The string_init
607 code gives up ownership of s , and returns the ownership of the result‐
608 ing string to its caller.
609 Note that it would be unsafe if s was passed as an additional parameter
611 to the function f as it could escape this way and be mutated in the fu‐
612 ture -- string_init would give up ownership of s to pass it to f , and
613 could not call unsafe_to_string safely.
614 We have provided the String.init , String.map and String.mapi functions
616 to cover most cases of building new strings. You should prefer those
617 over to_string or unsafe_to_string whenever applicable.
618 2. Temporarily giving ownership of a byte sequence to a function that
620 expects a uniquely owned string and returns ownership back, so that we
621 can mutate the sequence again after the call ended.
622 let bytes_length (s : bytes) =
625 String.length (Bytes.unsafe_to_string s)
626 In this use-case, we do not promise that s will never be mutated after
629 the call to bytes_length s . The String.length function temporarily
630 borrows unique ownership of the byte sequence (and sees it as a string
631 ), but returns this ownership back to the caller, which may assume that
632 s is still a valid byte sequence after the call. Note that this is only
633 correct because we know that String.length does not capture its argu‐
634 ment -- it could escape by a side-channel such as a memoization combi‐
635 nator.
636 The caller may not mutate s while the string is borrowed (it has tempo‐
638 rarily given up ownership). This affects concurrent programs, but also
639 higher-order functions: if String.length returned a closure to be
640 called later, s should not be mutated until this closure is fully ap‐
641 plied and returns ownership.
642 val unsafe_of_string : string -> bytes
646 Unsafely convert a shared string to a byte sequence that should not be
648 mutated.
649 The same ownership discipline that makes unsafe_to_string correct ap‐
651 plies to unsafe_of_string : you may use it if you were the owner of the
652 string value, and you will own the return bytes in the same mode.
653 In practice, unique ownership of string values is extremely difficult
655 to reason about correctly. You should always assume strings are shared,
656 never uniquely owned.
657 For example, string literals are implicitly shared by the compiler, so
659 you never uniquely own them.
660 let incorrect = Bytes.unsafe_of_string "hello"
663 let s = Bytes.of_string "hello"
664 The first declaration is incorrect, because the string literal "hello"
667 could be shared by the compiler with other parts of the program, and
668 mutating incorrect is a bug. You must always use the second version,
669 which performs a copy and is thus correct.
670 Assuming unique ownership of strings that are not string literals, but
672 are (partly) built from string literals, is also incorrect. For exam‐
673 ple, mutating unsafe_of_string ("foo" ^ s) could mutate the shared
674 string "foo" -- assuming a rope-like representation of strings. More
675 generally, functions operating on strings will assume shared ownership,
676 they do not preserve unique ownership. It is thus incorrect to assume
677 unique ownership of the result of unsafe_of_string .
678 The only case we have reasonable confidence is safe is if the produced
680 bytes is shared -- used as an immutable byte sequence. This is possibly
681 useful for incremental migration of low-level programs that manipulate
682 immutable sequences of bytes (for example Marshal.from_bytes ) and pre‐
683 viously used the string type for this purpose.
684 val split_on_char : sep:char -> bytes -> bytes list
688 split_on_char sep s returns the list of all (possibly empty) subse‐
691 quences of s that are delimited by the sep character.
692 The function's output is specified by the following invariants:
694 -The list is not empty.
697 -Concatenating its elements using sep as a separator returns a byte se‐
699 quence equal to the input ( Bytes.concat (Bytes.make 1 sep)
700 (Bytes.split_on_char sep s) = s ).
701 -No byte sequence in the result contains the sep character.
703 Since 4.13.0
707 Iterators
712 val to_seq : t -> char Seq.t
713 Iterate on the string, in increasing index order. Modifications of the
715 string during iteration will be reflected in the sequence.
716 Since 4.07
719 val to_seqi : t -> (int * char) Seq.t
723 Iterate on the string, in increasing order, yielding indices along
725 chars
726 Since 4.07
729 val of_seq : char Seq.t -> t
733 Create a string from the generator
735 Since 4.07
738 UTF codecs and validations
743 UTF-8
744 val get_utf_8_uchar : t -> int -> Uchar.utf_decode
745 get_utf_8_uchar b i decodes an UTF-8 character at index i in b .
748 val set_utf_8_uchar : t -> int -> Uchar.t -> int
752 set_utf_8_uchar b i u UTF-8 encodes u at index i in b and returns the
755 number of bytes n that were written starting at i . If n is 0 there was
756 not enough space to encode u at i and b was left untouched. Otherwise a
757 new character can be encoded at i + n .
758 val is_valid_utf_8 : t -> bool
762 is_valid_utf_8 b is true if and only if b contains valid UTF-8 data.
765 UTF-16BE
770 val get_utf_16be_uchar : t -> int -> Uchar.utf_decode
771 get_utf_16be_uchar b i decodes an UTF-16BE character at index i in b .
774 val set_utf_16be_uchar : t -> int -> Uchar.t -> int
778 set_utf_16be_uchar b i u UTF-16BE encodes u at index i in b and returns
781 the number of bytes n that were written starting at i . If n is 0 there
782 was not enough space to encode u at i and b was left untouched. Other‐
783 wise a new character can be encoded at i + n .
784 val is_valid_utf_16be : t -> bool
788 is_valid_utf_16be b is true if and only if b contains valid UTF-16BE
791 data.
792 UTF-16LE
797 val get_utf_16le_uchar : t -> int -> Uchar.utf_decode
798 get_utf_16le_uchar b i decodes an UTF-16LE character at index i in b .
801 val set_utf_16le_uchar : t -> int -> Uchar.t -> int
805 set_utf_16le_uchar b i u UTF-16LE encodes u at index i in b and returns
808 the number of bytes n that were written starting at i . If n is 0 there
809 was not enough space to encode u at i and b was left untouched. Other‐
810 wise a new character can be encoded at i + n .
811 val is_valid_utf_16le : t -> bool
815 is_valid_utf_16le b is true if and only if b contains valid UTF-16LE
818 data.
819 Binary encoding/decoding of integers
824 The functions in this section binary encode and decode integers to and
825 from byte sequences.
826 All following functions raise Invalid_argument if the space needed at
828 index i to decode or encode the integer is not available.
829 Little-endian (resp. big-endian) encoding means that least (resp. most)
831 significant bytes are stored first. Big-endian is also known as net‐
832 work byte order. Native-endian encoding is either little-endian or
833 big-endian depending on Sys.big_endian .
834 32-bit and 64-bit integers are represented by the int32 and int64
836 types, which can be interpreted either as signed or unsigned numbers.
837 8-bit and 16-bit integers are represented by the int type, which has
839 more bits than the binary encoding. These extra bits are handled as
840 follows:
841 -Functions that decode signed (resp. unsigned) 8-bit or 16-bit integers
843 represented by int values sign-extend (resp. zero-extend) their result.
844 -Functions that encode 8-bit or 16-bit integers represented by int val‐
846 ues truncate their input to their least significant bytes.
847 val get_uint8 : bytes -> int -> int
850 get_uint8 b i is b 's unsigned 8-bit integer starting at byte index i .
853 Since 4.08
856 val get_int8 : bytes -> int -> int
860 get_int8 b i is b 's signed 8-bit integer starting at byte index i .
863 Since 4.08
866 val get_uint16_ne : bytes -> int -> int
870 get_uint16_ne b i is b 's native-endian unsigned 16-bit integer start‐
873 ing at byte index i .
874 Since 4.08
877 val get_uint16_be : bytes -> int -> int
881 get_uint16_be b i is b 's big-endian unsigned 16-bit integer starting
884 at byte index i .
885 Since 4.08
888 val get_uint16_le : bytes -> int -> int
892 get_uint16_le b i is b 's little-endian unsigned 16-bit integer start‐
895 ing at byte index i .
896 Since 4.08
899 val get_int16_ne : bytes -> int -> int
903 get_int16_ne b i is b 's native-endian signed 16-bit integer starting
906 at byte index i .
907 Since 4.08
910 val get_int16_be : bytes -> int -> int
914 get_int16_be b i is b 's big-endian signed 16-bit integer starting at
917 byte index i .
918 Since 4.08
921 val get_int16_le : bytes -> int -> int
925 get_int16_le b i is b 's little-endian signed 16-bit integer starting
928 at byte index i .
929 Since 4.08
932 val get_int32_ne : bytes -> int -> int32
936 get_int32_ne b i is b 's native-endian 32-bit integer starting at byte
939 index i .
940 Since 4.08
943 val get_int32_be : bytes -> int -> int32
947 get_int32_be b i is b 's big-endian 32-bit integer starting at byte in‐
950 dex i .
951 Since 4.08
954 val get_int32_le : bytes -> int -> int32
958 get_int32_le b i is b 's little-endian 32-bit integer starting at byte
961 index i .
962 Since 4.08
965 val get_int64_ne : bytes -> int -> int64
969 get_int64_ne b i is b 's native-endian 64-bit integer starting at byte
972 index i .
973 Since 4.08
976 val get_int64_be : bytes -> int -> int64
980 get_int64_be b i is b 's big-endian 64-bit integer starting at byte in‐
983 dex i .
984 Since 4.08
987 val get_int64_le : bytes -> int -> int64
991 get_int64_le b i is b 's little-endian 64-bit integer starting at byte
994 index i .
995 Since 4.08
998 val set_uint8 : bytes -> int -> int -> unit
1002 set_uint8 b i v sets b 's unsigned 8-bit integer starting at byte index
1005 i to v .
1006 Since 4.08
1009 val set_int8 : bytes -> int -> int -> unit
1013 set_int8 b i v sets b 's signed 8-bit integer starting at byte index i
1016 to v .
1017 Since 4.08
1020 val set_uint16_ne : bytes -> int -> int -> unit
1024 set_uint16_ne b i v sets b 's native-endian unsigned 16-bit integer
1027 starting at byte index i to v .
1028 Since 4.08
1031 val set_uint16_be : bytes -> int -> int -> unit
1035 set_uint16_be b i v sets b 's big-endian unsigned 16-bit integer start‐
1038 ing at byte index i to v .
1039 Since 4.08
1042 val set_uint16_le : bytes -> int -> int -> unit
1046 set_uint16_le b i v sets b 's little-endian unsigned 16-bit integer
1049 starting at byte index i to v .
1050 Since 4.08
1053 val set_int16_ne : bytes -> int -> int -> unit
1057 set_int16_ne b i v sets b 's native-endian signed 16-bit integer start‐
1060 ing at byte index i to v .
1061 Since 4.08
1064 val set_int16_be : bytes -> int -> int -> unit
1068 set_int16_be b i v sets b 's big-endian signed 16-bit integer starting
1071 at byte index i to v .
1072 Since 4.08
1075 val set_int16_le : bytes -> int -> int -> unit
1079 set_int16_le b i v sets b 's little-endian signed 16-bit integer start‐
1082 ing at byte index i to v .
1083 Since 4.08
1086 val set_int32_ne : bytes -> int -> int32 -> unit
1090 set_int32_ne b i v sets b 's native-endian 32-bit integer starting at
1093 byte index i to v .
1094 Since 4.08
1097 val set_int32_be : bytes -> int -> int32 -> unit
1101 set_int32_be b i v sets b 's big-endian 32-bit integer starting at byte
1104 index i to v .
1105 Since 4.08
1108 val set_int32_le : bytes -> int -> int32 -> unit
1112 set_int32_le b i v sets b 's little-endian 32-bit integer starting at
1115 byte index i to v .
1116 Since 4.08
1119 val set_int64_ne : bytes -> int -> int64 -> unit
1123 set_int64_ne b i v sets b 's native-endian 64-bit integer starting at
1126 byte index i to v .
1127 Since 4.08
1130 val set_int64_be : bytes -> int -> int64 -> unit
1134 set_int64_be b i v sets b 's big-endian 64-bit integer starting at byte
1137 index i to v .
1138 Since 4.08
1141 val set_int64_le : bytes -> int -> int64 -> unit
1145 set_int64_le b i v sets b 's little-endian 64-bit integer starting at
1148 byte index i to v .
1149 Since 4.08
1152 Byte sequences and concurrency safety
1157 Care must be taken when concurrently accessing byte sequences from mul‐
1158 tiple domains: accessing a byte sequence will never crash a program,
1159 but unsynchronized accesses might yield surprising (non-sequen‐
1160 tially-consistent) results.
1161 Atomicity
1164 Every byte sequence operation that accesses more than one byte is not
1165 atomic. This includes iteration and scanning.
1166 For example, consider the following program:
1168 let size = 100_000_000
1169 let b = Bytes.make size ' '
1170 let update b f () =
1171 Bytes.iteri (fun i x -> Bytes.set b i (Char.chr (f (Char.code x)))) b
1172 let d1 = Domain.spawn (update b (fun x -> x + 1))
1173 let d2 = Domain.spawn (update b (fun x -> 2 * x + 1))
1174 let () = Domain.join d1; Domain.join d2
1175 the bytes sequence b may contain a non-deterministic mixture of '!' ,
1177 'A' , 'B' , and 'C' values.
1178 After executing this code, each byte of the sequence b is either '!' ,
1180 'A' , 'B' , or 'C' . If atomicity is required, then the user must im‐
1181 plement their own synchronization (for example, using Mutex.t ).
1182 Data races
1185 If two domains only access disjoint parts of a byte sequence, then the
1186 observed behaviour is the equivalent to some sequential interleaving of
1187 the operations from the two domains.
1188 A data race is said to occur when two domains access the same byte
1190 without synchronization and at least one of the accesses is a write.
1191 In the absence of data races, the observed behaviour is equivalent to
1192 some sequential interleaving of the operations from different domains.
1193 Whenever possible, data races should be avoided by using synchroniza‐
1195 tion to mediate the accesses to the elements of the sequence.
1196 Indeed, in the presence of data races, programs will not crash but the
1198 observed behaviour may not be equivalent to any sequential interleaving
1199 of operations from different domains. Nevertheless, even in the pres‐
1200 ence of data races, a read operation will return the value of some
1201 prior write to that location.
1202 Mixed-size accesses
1205 Another subtle point is that if a data race involves mixed-size writes
1206 and reads to the same location, the order in which those writes and
1207 reads are observed by domains is not specified. For instance, the fol‐
1208 lowing code write sequentially a 32-bit integer and a char to the same
1209 index
1210 let b = Bytes.make 10 '\000'
1211 let d1 = Domain.spawn (fun () -> Bytes.set_int32_ne b 0 100; b.[0] <- 'd' )
1212 In this situation, a domain that observes the write of 'd' to b. 0 is
1215 not guaranteed to also observe the write to indices 1 , 2 , or 3 .
1216OCamldoc 2023-07-20 BytesLabels(3)