From 5174ee7e118d2bc57fc2d8a6619101735af79b16 Mon Sep 17 00:00:00 2001
From: Matej Kosik
Date: Wed, 16 Dec 2015 12:55:40 +0100
Subject: COMMENTS: added to the "Unicode" module.

---
 lib/unicode.ml | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

(limited to 'lib/unicode.ml')

diff --git a/lib/unicode.ml b/lib/unicode.ml
index 1765e93dcd..05998bb804 100644
--- a/lib/unicode.ml
+++ b/lib/unicode.ml
@@ -18,7 +18,7 @@ exception Unsupported
    to simplify the masking process. (This choice seems to be a good
    trade-off between speed and space after some benchmarks.) *)
 
-(* A 256ko table, initially filled with zeros. *)
+(* A 256 KiB table, initially filled with zeros. *)
 let table = Array.make (1 lsl 17) 0
 
 (* Associate a 2-bit pattern to each status at position [i].
@@ -147,6 +147,11 @@ let utf8_of_unicode n =
       s
     end
 
+(* If [s] is some UTF-8 encoded string
+   and [i] is a position of some UTF-8 character within [s]
+   then [next_utf8 s i] returns [(j,n)] where:
+   - [j] indicates the position of the next UTF-8 character
+   - [n] represents the UTF-8 character at index [i] *)
 let next_utf8 s i =
   let err () = invalid_arg "utf8" in
   let l = String.length s - i in
-- 
cgit v1.2.3


From 508d5a99101097948b6de342295eec0d5c8cbe72 Mon Sep 17 00:00:00 2001
From: Pierre-Marie Pédrot
Date: Thu, 3 Mar 2016 18:59:51 +0100
Subject: Fixing bug #4105: poor escaping in the protocol between CoqIDE and
 coqtop.

Printing invalid UTF-8 string startled GTK too much, leading to CoqIDE dying
improperly. We now check that all strings outputed by Coq are proper UTF-8.
This is not perfect, as CoqIDE will sometimes truncate strings which contains
the null character, but at least it should not crash.
---
 lib/unicode.ml | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'lib/unicode.ml')

diff --git a/lib/unicode.ml b/lib/unicode.ml
index 1765e93dcd..cfaa73cc11 100644
--- a/lib/unicode.ml
+++ b/lib/unicode.ml
@@ -168,6 +168,13 @@ let next_utf8 s i =
        (c land 0x3F) lsl 6 + (d land 0x3F)
   else err ()
 
+let is_utf8 s =
+  let rec check i =
+    let (off, _) = next_utf8 s i in
+    check (i + off)
+  in
+  try check 0 with End_of_input -> true | Invalid_argument _ -> false
+
 (* Check the well-formedness of an identifier *)
 
 let initial_refutation j n s =
-- 
cgit v1.2.3