Support reading the new style of neverclaim output by Spin 6.24+.

* src/neverparse/neverclaimparse.yy, src/neverparse/neverclaimscan.ll:
Allow transitions between do..od, recognize atomic and assert.
* src/neverparse/parsedecl.hh: Pass the error_list to the lexer.
* src/tgbatest/neverclaimread.test: Add a test case.

src/neverparse/neverclaimparse.yy

@@ -1,5 +1,5 @@
 /* -*- coding: utf-8 -*-
-** Copyright (C) 2010, 2011, 2012 Laboratoire de Recherche et
+** Copyright (C) 2010, 2011, 2012, 2013 Laboratoire de Recherche et
 ** Développement de l'Epita (LRDE).
 **
 ** This file is part of Spot, a model checking library.
@@ -25,6 +25,7 @@
 %name-prefix "neverclaimyy"
 %debug
 %error-verbose
+%lex-param { spot::neverclaim_parse_error_list& error_list }
 
 %code requires
 {
@@ -59,20 +60,26 @@
    before parsedecl.hh uses it. */
 #include "parsedecl.hh"
 using namespace spot::ltl;
+static bool accept_all_needed = false;
+static bool accept_all_seen = false;
 }
 
 %token NEVER "never"
 %token SKIP "skip"
 %token IF "if"
 %token FI "fi"
+%token DO "do"
+%token OD "od"
 %token ARROW "->"
 %token GOTO "goto"
 %token FALSE "false"
+%token ATOMIC "atomic"
+%token ASSERT "assert"
 %token <str> FORMULA "boolean formula"
 %token <str> IDENT "identifier"
 %type <str> formula opt_dest
-%type <p> transition
+%type <p> transition src_dest
-%type <list> transitions
+%type <list> transitions transition_block
 %type <str> ident_list
 
 
@@ -112,7 +119,7 @@ ident_list:
   | ident_list IDENT ':'
     {
       result->add_state_alias(*$2, *$1);
-      // Keep any identifier that start with accept.
+      // Keep any identifier that starts with accept.
       if (strncmp("accept", $1->c_str(), 6))
         {
           delete $1;
@@ -125,9 +132,23 @@ ident_list:
         }
     }
 
+
+transition_block:
+  "if" transitions "fi"
+    {
+      $$ = $2;
+    }
+  | "do" transitions "od"
+    {
+      $$ = $2;
+    }
+
 state:
   ident_list "skip"
     {
+      if (*$1 == "accept_all")
+	accept_all_seen = true;
+
       spot::state_explicit_string::transition* t = result->create_transition(*$1, *$1);
       bool acc = !strncmp("accept", $1->c_str(), 6);
       if (acc)
@@ -137,11 +158,11 @@ state:
     }
   | ident_list { delete $1; }
   | ident_list "false" { delete $1; }
-  | ident_list "if" transitions "fi"
+  | ident_list transition_block
     {
       std::list<pair>::iterator it;
       bool acc = !strncmp("accept", $1->c_str(), 6);
-      for (it = $3->begin(); it != $3->end(); ++it)
+      for (it = $2->begin(); it != $2->end(); ++it)
       {
 	spot::state_explicit_string::transition* t =
 	  result->create_transition(*$1, *it->second);
@@ -153,10 +174,10 @@ state:
       }
       // Free the list
       delete $1;
-      for (std::list<pair>::iterator it = $3->begin();
+      for (std::list<pair>::iterator it = $2->begin();
-	   it != $3->end(); ++it)
+	   it != $2->end(); ++it)
 	delete it->second;
-      delete $3;
+      delete $2;
     }
 
 
@@ -184,41 +205,57 @@ opt_dest:
     {
       $$ = $3;
     }
+  | "->" "assert" FORMULA
+    {
+      delete $3;
+      $$ = new std::string("accept_all");
+      accept_all_needed = true;
+    }
 
-transition:
+src_dest: formula opt_dest
-  ':' ':' formula opt_dest
     {
       // If there is no destination, do ignore the transition.
       // This happens for instance with
       //   if
       //   :: false
       //   fi
-      if (!$4)
+      if (!$2)
 	{
-	  delete $3;
+	  delete $1;
 	  $$ = 0;
 	}
       else
 	{
 	  spot::ltl::parse_error_list pel;
 	  const spot::ltl::formula* f =
-	    spot::ltl::parse_boolean(*$3, pel, parse_environment,
+	    spot::ltl::parse_boolean(*$1, pel, parse_environment,
 				     debug_level(), true);
-	  delete $3;
+	  delete $1;
 	  for(spot::ltl::parse_error_list::const_iterator i = pel.begin();
 	  i != pel.end(); ++i)
 	    {
 	      // Adjust the diagnostic to the current position.
-	      location here = @3;
+	      location here = @1;
 	      here.end.line = here.begin.line + i->first.end.line - 1;
 	      here.end.column = here.begin.column + i->first.end.column -1;
 	      here.begin.line += i->first.begin.line - 1;
 	      here.begin.column += i->first.begin.column - 1;
 	      error(here, i->second);
 	    }
-	  $$ = new pair(f, $4);
+	  $$ = new pair(f, $2);
 	}
     }
+
+
+transition:
+  ':' ':' "atomic" '{' src_dest '}'
+    {
+      $$ = $5;
+    }
+  | ':' ':' src_dest
+    {
+      $$ = $3;
+    }
 %%
 
 void
@@ -250,6 +287,17 @@ namespace spot
     parser.set_debug_level(debug);
     parser.parse();
     neverclaimyyclose();
+
+    if (accept_all_needed && !accept_all_seen)
+      {
+	spot::state_explicit_string::transition* t =
+	  result->create_transition("accept_all", "accept_all");
+	result->add_acceptance_condition
+	  (t, spot::ltl::constant::true_instance());
+      }
+    accept_all_needed = false;
+    accept_all_seen = false;
+
     return result;
   }
 }

src/neverparse/neverclaimscan.ll

@@ -1,5 +1,5 @@
-/* Copyright (C) 2010, 2011 Laboratoire de Recherche et Développement de
+/* Copyright (C) 2010, 2011, 2013 Laboratoire de Recherche et
-** l'Epita (LRDE).
+** Développement de l'Epita (LRDE).
 **
 ** This file is part of Spot, a model checking library.
 **
@@ -23,6 +23,7 @@
 %{
 #include <string>
 #include "neverparse/parsedecl.hh"
+#include "misc/escape.hh"
 
 #define YY_USER_ACTION \
   yylloc->columns(yyleng);
@@ -30,9 +31,13 @@
 #define YY_NEVER_INTERACTIVE 1
 
 typedef neverclaimyy::parser::token token;
+static int parent_level = 0;
+static bool missing_parent = false;
 
 %}
 
+%x in_par
+
 eol      \n|\r|\n\r|\r\n
 
 %%
@@ -50,11 +55,46 @@ eol      \n|\r|\n\r|\r\n
 "skip"			return token::SKIP;
 "if"			return token::IF;
 "fi"			return token::FI;
+"do"			return token::DO;
+"od"			return token::OD;
 "->"			return token::ARROW;
 "goto"			return token::GOTO;
 "false"|"0"		return token::FALSE;
+"atomic"		return token::ATOMIC;
+"assert"		return token::ASSERT;
 
-("!"[ \t]*)?"(".*")"|"true"|"1"   {
+("!"[ \t]*)?"("		{
+			  parent_level = 1;
+			  BEGIN(in_par);
+			  yylval->str = new std::string(yytext,yyleng);
+			}
+
+<in_par>{
+	 "("		{
+			  ++parent_level;
+			  yylval->str->append(yytext, yyleng);
+			}
+	 ")"		{
+	                  yylval->str->append(yytext, yyleng);
+			  if (!--parent_level)
+			    {
+                              BEGIN(0);
+			      spot::trim(*yylval->str);
+			      return token::FORMULA;
+			    }
+			}
+         [^()]+		yylval->str->append(yytext, yyleng);
+	 <<EOF>>	{
+			  unput(')');
+			  if (!missing_parent)
+                             error_list.push_back(
+			       spot::neverclaim_parse_error(*yylloc,
+ 				"missing closing parenthese"));
+				  missing_parent = true;
+			}
+}
+
+"true"|"1"		{
                           yylval->str = new std::string(yytext, yyleng);
 			  return token::FORMULA;
                         }
@@ -95,5 +135,6 @@ namespace spot
   neverclaimyyclose()
   {
     fclose(yyin);
+    missing_parent = false;
   }
 }

src/neverparse/parsedecl.hh

@@ -1,5 +1,5 @@
-// Copyright (C) 2010 Laboratoire de Recherche et Développement de
+// Copyright (C) 2010, 2013 Laboratoire de Recherche et Développement
-// l'EPITA.
+// de l'EPITA.
 //
 // This file is part of Spot, a model checking library.
 //
@@ -24,8 +24,9 @@
 #include "location.hh"
 
 # define YY_DECL \
-  int neverclaimyylex (neverclaimyy::parser::semantic_type *yylval, \
+  int neverclaimyylex(neverclaimyy::parser::semantic_type *yylval, \
-		       neverclaimyy::location *yylloc)
+                      neverclaimyy::location *yylloc, \
+                      spot::neverclaim_parse_error_list& error_list)
 YY_DECL;
 
 namespace spot

src/tgbatest/neverclaimread.test

@@ -1,6 +1,6 @@
 #!/bin/sh
 # -*- coding: utf-8 -*-
-# Copyright (C) 2010, 2011, 2012 Laboratoire de Recherche et
+# Copyright (C) 2010, 2011, 2012, 2013 Laboratoire de Recherche et
 # Développement de l'Epita (LRDE).
 #
 # This file is part of Spot, a model checking library.
@@ -68,6 +68,51 @@ diff stdout expected
 rm input stdout expected
 
 
+# Same test, but with the newer syntax output since Spin 6.24
+cat >input <<EOF
+never {
+T2_init:
+do
+:: (1) -> goto T2_init
+:: (p1 && p0) -> goto T1
+od;
+T1:
+do
+:: atomic { (p1 && (! p0)) -> assert(!(p1 && (! p0))) }
+:: !(p1) -> goto T1
+:: ! (p1) -> goto T2_init
+od;
+}
+EOF
+
+run 0 ../ltl2tgba -XN input > stdout
+
+cat >expected <<EOF
+digraph G {
+  0 [label="", style=invis, height=0]
+  0 -> 1
+  1 [label="T2_init"]
+  1 -> 1 [label="1\n"]
+  1 -> 2 [label="p0 & p1\n"]
+  2 [label="T1"]
+  2 -> 3 [label="p1 & !p0\n"]
+  2 -> 2 [label="!p1\n"]
+  2 -> 1 [label="!p1\n"]
+  3 [label="accept_all", peripheries=2]
+  3 -> 3 [label="1\n{Acc[1]}"]
+}
+EOF
+
+# Sort out some possible inversions in the output.
+# (The order is not guaranteed by SPOT.)
+sed -e 's/!p0 & p1/p1 \& !p0/g' -e 's/p1 & p0/p0 \& p1/g' stdout \
+    > tmp_ && mv tmp_ stdout
+diff stdout expected
+
+rm input stdout expected
+
+
+
 # Test broken guards in input
 cat >input <<EOF
 never {
@@ -93,9 +138,8 @@ EOF
 # that "p1 && " is an atomic property.
 
 run 2 ../ltl2tgba -XN input > stdout 2>stderr
-cat >expected <<EOF
+cat >expected <<\EOF
-input:9.16: syntax error, unexpected closing parenthesis
+input:9.16: syntax error, unexpected $undefined, expecting fi or ':'
-input:9.16: ignoring trailing garbage
 EOF
 grep input: stderr >> stderrfilt
 diff stderrfilt expected