From d2f9a457d0bb2fd11ac7d5f6587174a79ca9c4b6 Mon Sep 17 00:00:00 2001
From: Pierre Letouzey
Date: Thu, 19 May 2016 15:18:26 +0200
Subject: Unicode.ascii_of_ident is now truly injective

 A non-ASCII char is now converted to _UUxxxx_ with xxxx being its unicode index
 in hexa. And any preexisting _UU substring in the ident is converted to _UUU.
 The switch from __Uxxxx_ to _UUxxxx_ is cosmetic, it just helps the extraction
 (less __ in names). But the other part of the patch (detection of preexisting
 _UU substrings) is critical to make ascii_of_ident truly injective and avoid
 the following kind of proof of False via native_compute :

  Definition α := 1.
  Definition __U03b1_ := 2.
  Lemma oups : False.
  Proof.
   assert (α = __U03b1_). { native_compute. reflexivity. }
   discriminate.
  Qed.

Conflicts:
	lib/unicode.mli
---
 lib/unicode.ml  | 30 ++++++++++++++++++++----------
 lib/unicode.mli |  9 ++++++++-
 2 files changed, 28 insertions(+), 11 deletions(-)

diff --git a/lib/unicode.ml b/lib/unicode.ml
index cfaa73cc11..0dc4238ee4 100644
--- a/lib/unicode.ml
+++ b/lib/unicode.ml
@@ -235,14 +235,24 @@ let is_basic_ascii s =
   !ok
 
 let ascii_of_ident s =
-  if is_basic_ascii s then s else
-    let i = ref 0 and out = ref "" in
-    begin try while true do
+  let len = String.length s in
+  let has_UU i =
+    i+2 < len && s.[i]='_' && s.[i+1]='U' && s.[i+2]='U'
+  in
+  let i = ref 0 in
+  while !i < len && Char.code s.[!i] < 128 && not (has_UU !i) do
+    incr i
+  done;
+  if !i = len then s else
+    let out = Buffer.create (2*len) in
+    Buffer.add_substring out s 0 !i;
+    while !i < len do
       let j, n = next_utf8 s !i in
-      out :=
-        if n >= 128
-        then Printf.sprintf "%s__U%04x_" !out n
-        else Printf.sprintf "%s%c" !out s.[!i];
-      i := !i + j
-    done with End_of_input -> () end;
-    !out
+      if n >= 128 then
+        (Printf.bprintf out "_UU%04x_" n; i := !i + j)
+      else if has_UU !i then
+        (Buffer.add_string out "_UUU"; i := !i + 3)
+      else
+        (Buffer.add_char out s.[!i]; incr i)
+    done;
+    Buffer.contents out
diff --git a/lib/unicode.mli b/lib/unicode.mli
index 65e75a20d6..00211164fb 100644
--- a/lib/unicode.mli
+++ b/lib/unicode.mli
@@ -23,8 +23,15 @@ val ident_refutation : string -> (bool * string) option
 (** First char of a string, converted to lowercase *)
 val lowercase_first_char : string -> string
 
-(** For extraction, turn a unicode string into an ascii-only one *)
+(** Return [true] if all UTF-8 characters in the input string are just plain
+    ASCII characters. Returns [false] otherwise. *)
 val is_basic_ascii : string -> bool
+
+(** [ascii_of_ident s] maps UTF-8 string to a string composed solely from ASCII
+    characters. The non-ASCII characters are translated to ["_UUxxxx_"] where
+    {i xxxx} is the Unicode index of the character in hexadecimal (from four
+    to six hex digits). To avoid potential name clashes, any preexisting
+    substring ["_UU"] is turned into ["_UUU"]. *)
 val ascii_of_ident : string -> string
 
 (** Validate an UTF-8 string *)
-- 
cgit v1.2.3