More cleanup of Util: utf8 aspects moved to a new file unicode.ml

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15818 85f007b7-540e-0410-9357-904b9bb8a0f7

1 files changed, 27 insertions, 0 deletions

diff --git a/lib/unicode.mli b/lib/unicode.mli
new file mode 100644
index 000000000..763123306
--- /dev/null
+++ b/lib/unicode.mli
@@ -0,0 +1,27 @@
+(************************************************************************)
+(*  v      *   The Coq Proof Assistant  /  The Coq Development Team     *)
+(* <O___,, *   INRIA - CNRS - LIX - LRI - PPS - Copyright 1999-2012     *)
+(*   \VV/  **************************************************************)
+(*    //   *      This file is distributed under the terms of the       *)
+(*         *       GNU Lesser General Public License Version 2.1        *)
+(************************************************************************)
+
+(** Unicode utilities *)
+
+type status = Letter | IdentPart | Symbol
+
+exception Unsupported
+
+(** Classify a unicode char into 3 classes, or raise [Unsupported] *)
+val classify : int -> status
+
+(** Check whether a given string be used as a legal identifier.
+    - [None] means yes
+    - [Some (b,s)] means no, with explanation [s] and severity [b] *)
+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 *)
+val ascii_of_ident : string -> string