Add more integer notations for parsing

CHANGELOG.md

@@ -1,6 +1,14 @@
 
 # [unreleased]
 
+## New features
+
+- Support more integer notations
+  ([PR#897](https://github.com/jasmin-lang/jasmin/pull/897)):
+    * Octal: `0O777`, `0o52`
+    * Binary: `0b11101`, `0B11100`
+    * `_` characters: `100_000_00___111`
+
 ## Bug fixes
 
 - Easycrypt extraction for CT : fix decreasing for loops

compiler/src/lexer.mll

@@ -108,6 +108,8 @@
 let blank    = [' ' '\t' '\r']
 let newline  = ['\n']
 let digit    = ['0'-'9']
+let octdigit = ['0'-'7']
+let bindigit = ['0'-'1']
 let hexdigit = ['0'-'9' 'a'-'f' 'A'-'F']
 let lower    = ['a'-'z']
 let upper    = ['A'-'Z']
@@ -133,11 +135,13 @@ rule main = parse
 
   | '"' (([^'"' '\\']|'\\' _)* as s) '"' { STRING (unescape (L.of_lexbuf lexbuf) s) }
 
-  (* Why this is needed *)
-  | ((*'-'?*) digit+) as s   
-      {INT s} 
+  | (digit+(('_')+ digit+)*) as s
 
-  | ('0' ['x' 'X'] hexdigit+) as s
+  | ('0' ['x' 'X'] hexdigit+(('_')+hexdigit+)*) as s
+
+  | ('0' ['b' 'B'] bindigit+(('_')+bindigit+)*) as s
+
+  | ('0' ['o' 'O'] octdigit+(('_')+octdigit+)*) as s
       {INT s}
 
   | ident as s

compiler/src/syntax.ml

@@ -1,4 +1,5 @@
 open Annotations
+open Utils
 (* -------------------------------------------------------------------- *)
 module L = Location
 
@@ -24,7 +25,9 @@ type castop1 = CSS of sowsize | CVS of svsize
 type castop = castop1 L.located option
 
 type int_representation = string
-let parse_int = Z.of_string
+let parse_int (i: int_representation) : Z.t =
+  let s = String.filter (( <> ) '_') i in
+  Z.of_string s
 
 let bits_of_wsize : wsize -> int = Annotations.int_of_ws 
 

compiler/tests/success/common/integer_notation.jazz

@@ -0,0 +1,22 @@
+/*
+Test for all valid integer syntaxes
+*/
+export fn test () -> reg u32 {
+    reg u32 y;
+    y = 0b11110000;
+    y = 0b111_111_11;
+    y = 0B111_00_11;
+
+    y = 0o01234567;
+    y = 0o765_4_321;
+    y = 0O76543210;
+
+    y = 1000000000;
+    y = 1000_0000_000;
+
+    y = 0x01234567;
+    y = 0x765_b_32aac;
+    y = 0X76aab3210;
+
+    return y;
+}