Upgrade to ocamlformat 0.27.0

.ocamlformat

@@ -1,4 +1,4 @@
 profile = default
-version = 0.26.2
+version = 0.27.0
 
 exp-grouping=preserve

src/kcas/kcas.ml

@@ -149,8 +149,8 @@ type 'a state = {
 }
 
 (** Tagged GADT for representing both the state of MCAS operations and of the
-    transaction log or splay [tree].  Different subsets of this GADT are used in
-    different contexts.  See the [root], [tree], and [which] existentials. *)
+    transaction log or splay [tree]. Different subsets of this GADT are used in
+    different contexts. See the [root], [tree], and [which] existentials. *)
 and _ tdt =
   | Before : [> `Before ] tdt
       (** The result has been determined to be the [before] value.
@@ -164,15 +164,15 @@ and _ tdt =
       mutable rot : rot;
           (** [rot] is for Root or Tree.
 
-              This field must be first, see [root_as_atomic] and
-              [tree_as_ref]. *)
+              This field must be first, see [root_as_atomic] and [tree_as_ref].
+          *)
       timeout : [ `Set | `Unset ] Timeout.t;
       mutable mode : Mode.t;
       mutable validate_counter : int;
       mutable post_commit : Action.t;
     }
       -> [> `Xt ] tdt
-      (** The result might not yet have been determined.  The [root] either says
+      (** The result might not yet have been determined. The [root] either says
           which it is or points to the root of the transaction log or [tree].
 
           Note that if/when local/stack allocation mode becomes available in

src/kcas_data/accumulator_intf.ml

@@ -3,7 +3,7 @@ module type Ops = sig
   type ('x, 'fn) fn
 
   val add : ('x, t -> int -> unit) fn
-  (** [add a n] increments the value of the accumulator [a] by [n].  [add]
+  (** [add a n] increments the value of the accumulator [a] by [n]. [add]
       operations can be performed scalably in parallel. *)
 
   val incr : ('x, t -> unit) fn

src/kcas_data/bits.mli

@@ -2,8 +2,8 @@ val max_0 : int -> int
 (** [max_0 n] is equivalent to [Int.max 0 n]. *)
 
 val is_pow_2 : int -> bool
-(** [is_pow_2 n] determines [n] is zero or of the form [1 lsl i] for some
-    [i]. *)
+(** [is_pow_2 n] determines [n] is zero or of the form [1 lsl i] for some [i].
+*)
 
 val ceil_pow_2_minus_1 : int -> int
 val ceil_pow_2 : int -> int

src/kcas_data/dllist.mli

@@ -12,8 +12,8 @@ open Kcas
     - The [length] operation is not provided.
     - The [set] operation is not provided.
 
-    A non-compositional {!take_all} operation is added for {{:
-    https://en.wikipedia.org/wiki/Privatization_(computer_programming)}privatization}
+    A non-compositional {!take_all} operation is added for
+    {{:https://en.wikipedia.org/wiki/Privatization_(computer_programming)}privatization}
     as well as conversions to a list of nodes ({!to_nodes_l} and {!to_nodes_r})
     and to a list of values ({!to_list_l} and {!to_list_r}).
 
@@ -46,7 +46,7 @@ val create : unit -> 'a t
 
 val create_node : 'a -> 'a node
 (** [create_node value] creates a new doubly-linked list node that is not in any
-    list.  The node can then e.g. be added to a list using {!move_l} or
+    list. The node can then e.g. be added to a list using {!move_l} or
     {!move_r}. *)
 
 val get : 'a node -> 'a

src/kcas_data/dllist_intf.ml

@@ -55,8 +55,8 @@ module type Ops = sig
   (** {3 Moving all nodes between lists} *)
 
   val swap : ('x, 'a t -> 'a t -> unit) fn
-  (** [swap l1 l2] exchanges the nodes of the doubly-linked lists [l1] and
-      [l2]. *)
+  (** [swap l1 l2] exchanges the nodes of the doubly-linked lists [l1] and [l2].
+  *)
 
   val transfer_l : ('x, 'a t -> 'a t -> unit) fn
   (** [transfer_l l1 l2] removes all nodes of [l1] and adds them to the left of

src/kcas_data/hashtbl.ml

@@ -495,7 +495,9 @@ let filter_map_inplace fn t =
   Kcas.Xt.commit { tx };
   Kcas.Xt.commit { tx = Xt.perform_pending t } |> ignore;
   (* Fenceless is safe as commit above has fences. *)
-  match Loc.fenceless_get raised with Done -> () | exn -> raise exn
+  match Loc.fenceless_get raised with
+  | Done -> ()
+  | exn -> raise exn
 
 let stats t =
   let length = ref 0 in

src/kcas_data/hashtbl.mli

@@ -5,10 +5,10 @@ open Kcas
     The interface provides a subset of the OCaml [Stdlib.Hashtbl] module with
     some changes:
 
-    - The functorial interface of the [Stdlib.Hashtbl] is not provided.
-      Instead the constructor functions, {!create}, {!of_seq}, and {!rebuild},
-      take an optional [HashedType] module as an argument.  By default {!create}
-      returns a randomized hash table.
+    - The functorial interface of the [Stdlib.Hashtbl] is not provided. Instead
+      the constructor functions, {!create}, {!of_seq}, and {!rebuild}, take an
+      optional [HashedType] module as an argument. By default {!create} returns
+      a randomized hash table.
     - The [add_seq] and [replace_seq] operations are not provided at all.
     - Non-instance specific operations related to randomization (e.g.
       [randomize], [is_randomized]) are not provided.
@@ -20,13 +20,13 @@ open Kcas
     {!fold}, and {!stats} are not provided.
 
     Please note that the design is intentionally based on [Stdlib.Hashtbl] and
-    copies its semantics as accurately as possible.  Some of the operations come
+    copies its semantics as accurately as possible. Some of the operations come
     with warnings.
 
-    The hash table implementation is designed to avoid starvation.  Read-only
-    accesses can generally proceed in parallel without interference.  Write
+    The hash table implementation is designed to avoid starvation. Read-only
+    accesses can generally proceed in parallel without interference. Write
     accesses that do not change the number of bindings can proceed in parallel
-    as long as they hit different internal buckets.  Write accesses that change
+    as long as they hit different internal buckets. Write accesses that change
     the number of bindings use a scalable {!Accumulator} and only make
     infrequent random checks to determine whether the hash table should be
     resized. *)
@@ -80,10 +80,10 @@ val of_seq :
   ('k * 'v) Seq.t ->
   ('k, 'v) t
 (** [of_seq assoc] creates a new hash table from the given association sequence
-    [assoc].  The associations are added in the same order as they appear in the
+    [assoc]. The associations are added in the same order as they appear in the
     sequence, using {!replace}, which means that if two pairs have the same key,
-    only the latest one will appear in the table.  See {!create} for the
-    optional arguments.
+    only the latest one will appear in the table. See {!create} for the optional
+    arguments.
 
     ⚠️ [of_seq (to_seq t)] does not necessarily copy the bindings of a hash table
     correctly. *)
@@ -109,7 +109,7 @@ val find : ('k, 'v) t -> 'k -> 'v
 
 val to_seq : ('k, 'v) t -> ('k * 'v) Seq.t
 (** [to_seq t] takes a snapshot of the keys and values in the hash table and
-    returns them as an association sequence.  Bindings of each individual key
+    returns them as an association sequence. Bindings of each individual key
     appear in the sequence in reverse order of their introduction.
 
     ⚠️ [of_seq (to_seq t)] does not necessarily copy the bindings of a hash table
@@ -134,12 +134,12 @@ val rebuild :
 (** [rebuild t] returns a copy of the given hash table [t] optionally rehashing
     all of the bindings.
 
-    See {!create} for descriptions of the optional arguments.  Unlike {!create},
+    See {!create} for descriptions of the optional arguments. Unlike {!create},
     [rebuild] uses the given hash table [t] as a template to get defaults for
     the optional arguments. *)
 
 val copy : ('k, 'v) t -> ('k, 'v) t
-(** [copy t] is equivalent to [rebuild t].  In other words, the returned hash
+(** [copy t] is equivalent to [rebuild t]. In other words, the returned hash
     table uses the same {!hashed_type} (and other parameters) as the given hash
     table [t]. *)
 
@@ -148,15 +148,16 @@ val iter : ('a -> 'b -> unit) -> ('a, 'b) t -> unit
 
 val filter_map_inplace : ('k -> 'v -> 'v option) -> ('k, 'v) t -> unit
 (** [filter_map_inplace f t] applies [f] to all bindings in the hash table [t]
-    and updates each binding depending on the result of [f].  If [f] returns
-    [None], the binding is discarded.  If [f] returns [Some new_value], the
+    and updates each binding depending on the result of [f]. If [f] returns
+    [None], the binding is discarded. If [f] returns [Some new_value], the
     binding is updated to associate the key to the [new_value].
 
     ⚠️ The given [f] may be called multiple times for the same bindings from
     multiple domains in parallel. *)
 
 val fold : ('a -> 'b -> 'c -> 'c) -> ('a, 'b) t -> 'c -> 'c
-(** [fold f t a] is equivalent to [Seq.fold_left (fun a (k, v) -> f k v a) a (to_seq t)]. *)
+(** [fold f t a] is equivalent to
+    [Seq.fold_left (fun a (k, v) -> f k v a) a (to_seq t)]. *)
 
 val stats : ('a, 'b) t -> Stdlib.Hashtbl.statistics
 (** [stats t] returns statistics about the hash table [t]. *)

src/kcas_data/kcas_data.mli

@@ -22,15 +22,15 @@
     explicitly documented.
 
     The main feature of these data structure implementations is their
-    compositionality.  If your application does not need compositionality, then
+    compositionality. If your application does not need compositionality, then
     other concurrency and parallelism safe data structure libraries may
     potentially offer better performance.
 
     But why should you care about composability?
 
     As an example, consider the implementation of a least-recently-used (LRU)
-    cache or a bounded associative map, but first, let's open the libraries
-    for convenience:
+    cache or a bounded associative map, but first, let's open the libraries for
+    convenience:
 
     {[
       open Kcas
@@ -41,35 +41,38 @@
     and a doubly-linked list and keep track of the amount of space in the cache:
 
     {[
-      type ('k, 'v) cache =
-        { space: int Loc.t;
-          table: ('k, 'k Dllist.node * 'v) Hashtbl.t;
-          order: 'k Dllist.t }
+      type ('k, 'v) cache = {
+        space : int Loc.t;
+        table : ('k, 'k Dllist.node * 'v) Hashtbl.t;
+        order : 'k Dllist.t;
+      }
     ]}
 
     On a cache lookup the doubly-linked list node corresponding to the accessed
     key is moved to the left end of the list:
 
     {[
-      let get_opt {table; order; _} key =
+      let get_opt { table; order; _ } key =
         Hashtbl.find_opt table key
         |> Option.map @@ fun (node, datum) ->
-           Dllist.move_l node order; datum
+           Dllist.move_l node order;
+           datum
     ]}
 
     On a cache update, in case of overflow, the association corresponding to the
     node on the right end of the list is dropped:
 
     {[
-      let set {table; order; space; _} key datum =
+      let set { table; order; space; _ } key datum =
         let node =
           match Hashtbl.find_opt table key with
           | None ->
-            if 0 = Loc.update space (fun n -> max 0 (n-1))
-            then Dllist.take_opt_r order
-                 |> Option.iter (Hashtbl.remove table);
-            Dllist.add_l key order
-          | Some (node, _) -> Dllist.move_l node order; node
+              if 0 = Loc.update space (fun n -> max 0 (n - 1)) then
+                Dllist.take_opt_r order |> Option.iter (Hashtbl.remove table);
+              Dllist.add_l key order
+          | Some (node, _) ->
+              Dllist.move_l node order;
+              node
         in
         Hashtbl.replace table key (node, datum)
     ]}
@@ -98,33 +101,34 @@
     operations. For the [get_opt] operation we end up with
 
     {[
-      let get_opt ~xt {table; order; _} key =
+      let get_opt ~xt { table; order; _ } key =
         Hashtbl.Xt.find_opt ~xt table key
         |> Option.map @@ fun (node, datum) ->
-           Dllist.Xt.move_l ~xt node order; datum
+           Dllist.Xt.move_l ~xt node order;
+           datum
     ]}
 
     and the [set] operation is just as easy to convert to a transactional
-    version.  One way to think about transactions is that they give us back the
+    version. One way to think about transactions is that they give us back the
     ability to compose programs such as the above. *)
 
 (** {1 [Stdlib] style data structures}
 
     The data structures in this section are designed to closely mimic the
-    corresponding unsynchronized data structures in the OCaml [Stdlib].  Each of
+    corresponding unsynchronized data structures in the OCaml [Stdlib]. Each of
     these provide a non-compositional, but concurrency and parallelism safe,
-    interface that is close to the [Stdlib] equivalent.  Additionally,
+    interface that is close to the [Stdlib] equivalent. Additionally,
     compositional transactional interfaces are provided for some operations.
 
     These implementations will use more space than the corresponding [Stdlib]
-    data structures.  Performance, when accessed concurrently, should be
+    data structures. Performance, when accessed concurrently, should be
     competitive or superior compared to naïve locking. *)
 
 module Hashtbl = Hashtbl
 module Queue = Queue
 module Stack = Stack
 
-(** {1 Communication and synchronization primitives}  *)
+(** {1 Communication and synchronization primitives} *)
 
 module Mvar = Mvar
 module Promise = Promise

src/kcas_data/mvar.mli

@@ -6,15 +6,15 @@ open Kcas
     with blocking semantics on both {!take} and {!put}.
 
     {b NOTE}: The current implementation is not guaranteed to be fair or
-    scalable.  In other words, when multiple producers block on {!put} or
+    scalable. In other words, when multiple producers block on {!put} or
     multiple consumers block on {!take} the operations are not queued and it is
     possible for a particular producer or consumer to starve. *)
 
 (** {1 Common interface} *)
 
 type !'a t
-(** The type of a synchronizing variable that may contain a value of type
-    ['a]. *)
+(** The type of a synchronizing variable that may contain a value of type ['a].
+*)
 
 val create : 'a option -> 'a t
 (** [create x_opt] returns a new synchronizing variable that will either be

src/kcas_data/promise.mli

@@ -18,7 +18,7 @@ open Kcas
 (** {1 Common interface} *)
 
 type !+'a t
-(** The type of a promise of a value of type ['a].  *)
+(** The type of a promise of a value of type ['a]. *)
 
 type !-'a u
 (** The type of a resolver of a value of type ['a]. *)

src/kcas_data/promise_intf.ml

@@ -7,7 +7,7 @@ module type Ops = sig
 
   val resolve : ('x, 'a u -> 'a -> unit) fn
   (** [resolve u v] resolves the promise corresponding to the resolver [u] to
-      the value [v].  Any awaiters of the corresponding promise are then
+      the value [v]. Any awaiters of the corresponding promise are then
       unblocked. *)
 
   val await : ('x, 'a t -> 'a) blocking_fn
@@ -25,7 +25,8 @@ module type Ops = sig
   (** {2 Result promises} *)
 
   val await_exn : ('x, 'a or_exn -> 'a) blocking_fn
-  (** [await_exn t] is equivalent to [match await t with v -> v | exception e -> raise e]. *)
+  (** [await_exn t] is equivalent to
+      [match await t with v -> v | exception e -> raise e]. *)
 
   val resolve_ok : ('x, ('a, 'b) result u -> 'a -> unit) fn
   (** [resolve_ok u v] is equivalent to [resolve u (Ok v)]. *)

src/kcas_data/queue.mli

@@ -3,12 +3,12 @@ open Kcas
 (** First-In First-Out (FIFO) queue.
 
     The interface provides a subset of the OCaml [Stdlib.Queue] module.
-    [transfer] and [add_seq] are not provided at all.  Compositional versions of
+    [transfer] and [add_seq] are not provided at all. Compositional versions of
     {!iter}, {!fold}, {!peek}, {!top}, and {!take} are not provided.
 
     The queue implementation is designed to avoid contention between a producer
-    and a consumer operating concurrently.  The implementation is also designed
-    to avoid starvation.  Performance in most concurrent use cases should be
+    and a consumer operating concurrently. The implementation is also designed
+    to avoid starvation. Performance in most concurrent use cases should be
     superior to what can be achieved with one or two locks. *)
 
 (** {1 Common interface} *)

src/kcas_data/stack.mli

@@ -3,7 +3,7 @@ open Kcas
 (** Last-In First-Out (LIFO) stack.
 
     The interface provides a subset of the OCaml [Stdlib.Stack] module.
-    [add_seq] is not provided at all.  Compositional versions of {!iter},
+    [add_seq] is not provided at all. Compositional versions of {!iter},
     {!fold}, {!pop}, and {!top} are not provided.
 
     The implementation is essentially a

test/kcas_data/linearizable_chaining_example.ml

@@ -1,19 +1,18 @@
 (** This demonstrates an approach to composing non-blocking linearizable data
     structures inspired by the paper
 
-      Concurrent Size
-      by Gal Sela and Erez Petrank
-      https://arxiv.org/pdf/2209.07100.pdf
+    Concurrent Size by Gal Sela and Erez Petrank
+    https://arxiv.org/pdf/2209.07100.pdf
 
     First a [Hashtbl] is implemented that allows [idempotent_add] and
-    [idempotent_remove] operations to be specified.  The hash table makes sure
+    [idempotent_remove] operations to be specified. The hash table makes sure
     that any operations that might witness the addition or the removal of a key
     will perform those operations before returning.
 
     Then a [Hashtbl_with_order] is implemented on top of the [Hashtbl] by
     specifying the [idempotent_add] and [idempotent_remove] operation such that
     they update a lock-free doubly-linked list to maintain a list of the keys in
-    the hash table in insertion [order].  In other words, we composed a hash
+    the hash table in insertion [order]. In other words, we composed a hash
     table with a doubly-linked list, both lock-free and linearizable, resulting
     in a lock-free linearizable hash table that maintains the insertion order.