From d7bbb65c48be26a287bb9aca3327f228e165f2b8 Mon Sep 17 00:00:00 2001
From: Gabriel Scherer <gabriel.scherer@gmail.com>
Date: Sat, 30 Jan 2021 22:05:54 +0100
Subject: [PATCH] benchmark various implementations of Bench.to_seq discussed
 in #1006

Results on my machine:

    enumeration (89.71 us) is 77.5% faster than
    batseq (399.39 us) which is 3.8% faster than
    too strict (415.29 us) which is probably (alpha=40.06%) same speed as
    simple (416.74 us)
    Saving times to times.flat

This suggests that using an enumeration indeed provides a large
performance advantage over the simple approach.
---
 Makefile                       |  1 +
 benchsuite/bench_set_to_seq.ml | 80 ++++++++++++++++++++++++++++++++++
 2 files changed, 81 insertions(+)
 create mode 100644 benchsuite/bench_set_to_seq.ml

diff --git a/Makefile b/Makefile
index f44f5bfdb..30aaa988c 100644
--- a/Makefile
+++ b/Makefile
@@ -67,6 +67,7 @@ BENCH_TARGETS += benchsuite/lines_of.native
 BENCH_TARGETS += benchsuite/bitset.native
 BENCH_TARGETS += benchsuite/bench_map.native
 BENCH_TARGETS += benchsuite/bench_nreplace.native
+BENCH_TARGETS += benchsuite/bench_set_to_seq.native
 TEST_TARGET = test-byte
 
 ifeq ($(BATTERIES_NATIVE_SHLIB), yes)
diff --git a/benchsuite/bench_set_to_seq.ml b/benchsuite/bench_set_to_seq.ml
new file mode 100644
index 000000000..4bda8a5c2
--- /dev/null
+++ b/benchsuite/bench_set_to_seq.ml
@@ -0,0 +1,80 @@
+(* cd .. && ocamlbuild -use-ocamlfind benchsuite/bench_set_to_seq.native && _build/benchsuite/bench_set_to_seq.native *)
+
+(* The purpose of this test is to compare different implementation of
+   BatSet.to_seq. *)
+
+(* the type BatSet.t is abstract,
+   we break the abstraction boundary locally to implement our versions outside the module. *)
+type 'a set =
+  | Empty
+  | Node of 'a set * 'a * 'a set * int
+external hide : 'a set -> 'a BatSet.t = "%identity"
+external reveal : 'a BatSet.t -> 'a set = "%identity"
+
+
+module TooStrict = struct
+  let rec to_seq m =
+    match m with
+    | Empty -> BatSeq.nil
+    | Node(l, v, r, _) ->
+       BatSeq.append (to_seq l) (fun () -> BatSeq.Cons (v, to_seq r))
+
+  let to_seq s = to_seq (reveal s)
+end
+
+module Simple = struct
+  let rec to_seq m =
+    fun () ->
+    match m with
+    | Empty -> BatSeq.Nil
+    | Node(l, v, r, _) ->
+       BatSeq.append (to_seq l) (fun () -> BatSeq.Cons (v, to_seq r)) ()
+
+  let to_seq s = to_seq (reveal s)
+end
+
+module Enumeration = struct
+  type 'a iter = E | C of 'a * 'a set * 'a iter
+
+  let rec cons_iter s t = match s with
+    | Empty -> t
+    | Node (l, e, r, _) -> cons_iter l (C (e, r, t))
+
+  let to_seq s =
+    let rec to_seq iter () =
+      match iter with
+      | E -> BatSeq.Nil
+      | C (e, r, t) ->
+         BatSeq.Cons (e, to_seq (cons_iter r t))
+    in
+    to_seq (cons_iter s E)
+
+  let to_seq s = to_seq (reveal s)
+end
+
+let test_input =
+  let s = ref BatSet.empty in
+  for i = 0 to 9999 do
+    s := BatSet.add i !s;
+  done;
+  !s
+
+let test to_seq =
+  test_input
+  |> to_seq
+  |> BatSeq.length
+
+let () =
+  assert (test TooStrict.to_seq = test BatSet.to_seq);
+  assert (test Simple.to_seq = test BatSet.to_seq);
+  assert (test Enumeration.to_seq = test BatSet.to_seq);
+  ()
+
+let () =
+  let repeat f n = for i = 1 to n do ignore (f ()) done in
+  Bench.bench_n [
+      "too strict", repeat (fun () -> test TooStrict.to_seq);
+      "simple", repeat (fun () -> test Simple.to_seq);
+      "enumeration", repeat (fun () -> test Enumeration.to_seq);
+      "batseq", repeat (fun () -> test BatSet.to_seq);
+    ] |>  Bench.run_outputs