Merge branch 'branch-2.6' into maint

* origin/branch-2.6: (24 commits)
  tests: calc: modernize the use of locations
  tests: remove useless location initializations
  lalr1.cc: always initialize yylval.
  tests: check that C and C++ objects can be linked together.
  yacc.c: also disable -Wuninitialized.
  glr.cc, yacc.c: initialize yylloc properly
  yacc.c, glr.c: a better YY_LOCATION_PRINT
  yacc.c: simplify initialization
  doc: formatting changes
  c++: fix position operator signatures
  tests: remove useless location initialization.
  tests: fix locations in C
  tests: handle %parse-param in the generated yyerror
  tests: simplifications
  grammars: fix display of nul character in error message
  tests: sort
  tests: cosmetic changes
  comment changes
  autoconf: update
  gnulib: update
  ...

Conflicts:
	NEWS
	gnulib
	tests/named-refs.at
	tests/regression.at

NEWS

@@ -51,6 +51,25 @@ GNU Bison NEWS
   The reductions are now explicitly represented as transitions to other
   diamond shaped nodes.
 
+* Noteworthy changes in release ?.? (????-??-??) [?]
+
+  We consider compiler warnings about Bison generated parsers to be bugs.
+  Rather than working around them in your own project, please consider
+  reporting them to us.
+
+** Bug fixes
+
+  Warnings about uninitialized yylval and/or yylloc for push parsers with a
+  pure interface have been fixed for GCC 4.0 up to 4.8, and Clang 2.9 to
+  3.2.
+
+  Other issues in the test suite have been addressed.
+
+  Nul characters are correctly displayed in error messages.
+
+  When possible, yylloc is correctly initialized before calling yylex.  It
+  is no longer necessary to initialize it in the %initial-action.
+
 * Noteworthy changes in release 2.6.4 (2012-10-23) [stable]
 
   Bison 2.6.3's --version was incorrect.  This release fixes this issue.

THANKS

@@ -64,6 +64,7 @@ Laurent Mascherpa         laurent.mascherpa@epita.fr
 Lie Yan                   lie.yan@kaust.edu.sa
 Magnus Fromreide          magfr@lysator.liu.se
 Marc Autret               autret_m@epita.fr
+Marc Mendiola             mmendiol@usc.edu
 Martin Jacobs             martin.jacobs@arcor.de
 Martin Mokrejs            mmokrejs@natur.cuni.cz
 Martin Nylin              martin.nylin@linuxmail.org
@@ -88,6 +89,7 @@ Paul Hilfinger            Hilfinger@CS.Berkeley.EDU
 Per Allansson             per@appgate.com
 Peter Fales               psfales@lucent.com
 Peter Hamorsky            hamo@upjs.sk
+Peter Simons              simons@cryp.to
 Piotr Gackiewicz          gacek@intertel.com.pl
 Quoc Peyrot               chojin@lrde.epita.fr
 R Blake                   blakers@mac.com

data/c.m4

@@ -643,3 +643,46 @@ m4_define([b4_yylloc_default_define],
     while (YYID (0))
 #endif
 ]])
+
+# b4_yy_location_print_define
+# ---------------------------
+# Define YY_LOCATION_PRINT.
+m4_define([b4_yy_location_print_define],
+[b4_locations_if([[
+/* YY_LOCATION_PRINT -- Print the location on the stream.
+   This macro was not mandated originally: define only if we know
+   we won't break user code: when these are the locations we know.  */
+
+#ifndef YY_LOCATION_PRINT
+# if defined ]b4_api_PREFIX[LTYPE_IS_TRIVIAL && ]b4_api_PREFIX[LTYPE_IS_TRIVIAL
+#  define YY_LOCATION_PRINT(File, Loc)                                   \
+  do {                                                                   \
+    fprintf (File, "%d.%d", (Loc).first_line, (Loc).first_column);       \
+    if ((Loc).first_line < (Loc).last_line)                              \
+      fprintf (File, "-%d.%d", (Loc).last_line,  (Loc).last_column - 1); \
+    else if ((Loc).first_column < (Loc).last_column - 1)                 \
+      fprintf (File, "-%d", (Loc).last_column - 1);                      \
+  } while (0)
+# else
+#  define YY_LOCATION_PRINT(File, Loc) ((void) 0)
+# endif
+#endif]],
+[[/* This macro is provided for backward compatibility. */
+#ifndef YY_LOCATION_PRINT
+# define YY_LOCATION_PRINT(File, Loc) ((void) 0)
+#endif]])
+])
+
+# b4_yyloc_default
+# ----------------
+# Expand to a possible default value for yylloc.
+m4_define([b4_yyloc_default],
+[[
+# if defined ]b4_api_PREFIX[LTYPE_IS_TRIVIAL && ]b4_api_PREFIX[LTYPE_IS_TRIVIAL
+  = { ]m4_join([, ],
+               m4_defn([b4_location_initial_line]),
+               m4_defn([b4_location_initial_column]),
+               m4_defn([b4_location_initial_line]),
+               m4_defn([b4_location_initial_column]))[ }
+# endif
+]])

data/glr.c

@@ -223,11 +223,11 @@ b4_percent_code_get([[top]])[
 #endif
 
 /* Default (constant) value used for initialization for null
-   right-hand sides.  Unlike the standard yacc.c template,
+   right-hand sides.  Unlike the standard yacc.c template, here we set
-   here we set the default value of $$ to a zeroed-out value.
+   the default value of $$ to a zeroed-out value.  Since the default
-   Since the default value is undefined, this behavior is
+   value is undefined, this behavior is technically correct.  */
-   technically correct.  */
+static YYSTYPE yyval_default;]b4_locations_if([[
-static YYSTYPE yyval_default;
+static YYLTYPE yyloc_default][]b4_yyloc_default;])[
 
 /* Copy the second part of user declarations.  */
 ]b4_user_post_prologue
@@ -475,28 +475,10 @@ static const ]b4_int_type_for([b4_stos])[ yystos[] =
 #define YYTERROR 1
 
 ]b4_locations_if([[
-#ifndef YYLLOC_DEFAULT
 ]b4_yylloc_default_define[
 # define YYRHSLOC(Rhs, K) ((Rhs)[K].yystate.yyloc)
-
+]])[
-/* YY_LOCATION_PRINT -- Print the location on the stream.
+]b4_yy_location_print_define[
-   This macro was not mandated originally: define only if we know
-   we won't break user code: when these are the locations we know.  */
-
-# define YY_LOCATION_PRINT(File, Loc)                   \
-    fprintf (File, "%d.%d-%d.%d",                       \
-             (Loc).first_line, (Loc).first_column,      \
-             (Loc).last_line,  (Loc).last_column)
-#endif
-]],[
-#ifndef YYLLOC_DEFAULT
-# define YYLLOC_DEFAULT(Current, Rhs, N) ((void) 0)
-#endif
-])[
-
-#ifndef YY_LOCATION_PRINT
-# define YY_LOCATION_PRINT(File, Loc) ((void) 0)
-#endif
 
 /* YYLEX -- calling `yylex' with the right arguments.  */
 #define YYLEX ]b4_c_function_call([yylex], [int], b4_lex_param)[
@@ -2295,14 +2277,9 @@ yyrecoverSyntaxError (yyGLRStack* yystackp]b4_user_formals[)
   YYDPRINTF ((stderr, "Starting parse\n"));
 
   yychar = YYEMPTY;
-  yylval = yyval_default;
+  yylval = yyval_default;]b4_locations_if([
-]b4_locations_if([
+  yylloc = yyloc_default;])[
-#if defined ]b4_api_PREFIX[LTYPE_IS_TRIVIAL && ]b4_api_PREFIX[LTYPE_IS_TRIVIAL
+]m4_ifdef([b4_initial_action], [
-  yylloc.first_line   = yylloc.last_line   = ]b4_location_initial_line[;
-  yylloc.first_column = yylloc.last_column = ]b4_location_initial_column[;
-#endif
-])
-m4_ifdef([b4_initial_action], [
 b4_dollar_pushdef([yylval], [], [yylloc])dnl
 /* User initialization code.  */
 b4_user_initial_action

data/glr.cc

@@ -87,12 +87,17 @@ m4_define([b4_yy_symbol_print_generate],
 ]b4_parse_param_use[]dnl
 [  yyparser.yy_symbol_print_ (yytype, yyvaluep]b4_locations_if([, yylocationp])[);
 }
-]])
+]])[
 
+# Hijack the initial action to initialize the locations.
+]b4_locations_if([b4_percent_define_ifdef([[api.location.type]], [],
+[m4_define([b4_initial_action],
+[yylloc.initialize ();]m4_ifdef([b4_initial_action], [
+m4_defn([b4_initial_action])]))])])[
 
 # Hijack the post prologue to insert early definition of YYLLOC_DEFAULT
 # and declaration of yyerror.
-m4_append([b4_post_prologue],
+]m4_append([b4_post_prologue],
 [b4_syncline([@oline@], [@ofile@])[
 ]b4_yylloc_default_define[
 #define YYRHSLOC(Rhs, K) ((Rhs)[K].yystate.yyloc)

data/lalr1.cc

@@ -512,17 +512,18 @@ do {					\
     int yychar = yyempty_;
     int yytoken = 0;
 
-    /* State.  */
+    // State.
     int yyn;
     int yylen = 0;
     int yystate = 0;
 
-    /* Error handling.  */
+    // Error handling.
     int yynerrs_ = 0;
     int yyerrstatus_ = 0;
 
     /// Semantic value of the lookahead.
-    semantic_type yylval;
+    static semantic_type yyval_default;
+    semantic_type yylval = yyval_default;
     /// Location of the lookahead.
     location_type yylloc;
     /// The locations where the error started and ended.

data/location.cc

@@ -87,7 +87,7 @@ b4_copyright([Positions for Bison parsers in C++],
   };
 
   /// Add and assign a position.
-  inline const position&
+  inline position&
   operator+= (position& res, const int width)
   {
     res.columns (width);
@@ -103,7 +103,7 @@ b4_copyright([Positions for Bison parsers in C++],
   }
 
   /// Add and assign a position.
-  inline const position&
+  inline position&
   operator-= (position& res, const int width)
   {
     return res += -width;

data/yacc.c

@@ -171,10 +171,11 @@ m4_define([b4_declare_scanner_communication_variables], [[
 int yychar;
 
 ]b4_pure_if([[
-#if defined __GNUC__ && (4 < __GNUC__ + (6 <= __GNUC_MINOR__))
+#if defined __GNUC__ && 407 <= __GNUC__ * 100 + __GNUC_MINOR__
 /* Suppress an incorrect diagnostic about yylval being uninitialized.  */
 # define YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN \
     _Pragma ("GCC diagnostic push") \
+    _Pragma ("GCC diagnostic ignored \"-Wuninitialized\"")\
     _Pragma ("GCC diagnostic ignored \"-Wmaybe-uninitialized\"")
 # define YY_IGNORE_MAYBE_UNINITIALIZED_END \
     _Pragma ("GCC diagnostic pop")
@@ -182,21 +183,22 @@ int yychar;
 /* Default value used for initialization, for pacifying older GCCs
    or non-GCC compilers.  */
 static YYSTYPE yyval_default;
-# define YYLVAL_INITIALIZE() (yylval = yyval_default)
+# define YY_INITIAL_VALUE(Value) = Value
 #endif]])[
-#ifndef YYLVAL_INITIALIZE
-# define YYLVAL_INITIALIZE()
-#endif
 #ifndef YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN
 # define YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN
 # define YY_IGNORE_MAYBE_UNINITIALIZED_END
 #endif
+#ifndef YY_INITIAL_VALUE
+# define YY_INITIAL_VALUE(Value) /* Nothing. */
+#endif
 
 /* The semantic value of the lookahead symbol.  */
-YYSTYPE yylval;]b4_locations_if([[
+YYSTYPE yylval YY_INITIAL_VALUE(yyval_default);]b4_locations_if([[
 
 /* Location data for the lookahead symbol.  */
-YYLTYPE yylloc;]])b4_pure_if([], [[
+YYLTYPE yylloc][]b4_yyloc_default[;
+]])b4_pure_if([], [[
 
 /* Number of syntax errors so far.  */
 int yynerrs;]])])
@@ -264,7 +266,7 @@ typedef struct ]b4_prefix[pstate ]b4_prefix[pstate;
   [[b4_prefix[pstate *ps]], [[ps]]]b4_pure_if([,
   [[[int pushed_char]], [[pushed_char]]],
   [[b4_api_PREFIX[STYPE const *pushed_val]], [[pushed_val]]]b4_locations_if([,
-  [[b4_api_PREFIX[LTYPE const *pushed_loc]], [[pushed_loc]]]])])m4_ifset([b4_parse_param], [,
+  [[b4_api_PREFIX[LTYPE *pushed_loc]], [[pushed_loc]]]])])m4_ifset([b4_parse_param], [,
   b4_parse_param]))
 b4_pull_if([b4_c_function_decl([b4_prefix[pull_parse]], [[int]],
   [[b4_prefix[pstate *ps]], [[ps]]]m4_ifset([b4_parse_param], [,
@@ -765,33 +767,15 @@ do                                                              \
     }								\
 while (YYID (0))
 
-
+/* Error token number */
 #define YYTERROR	1
 #define YYERRCODE	256
 
 ]b4_locations_if([[
 ]b4_yylloc_default_define[
 #define YYRHSLOC(Rhs, K) ((Rhs)[K])
-
+]])[
-/* YY_LOCATION_PRINT -- Print the location on the stream.
+]b4_yy_location_print_define[
-   This macro was not mandated originally: define only if we know
-   we won't break user code: when these are the locations we know.  */
-
-#ifndef YY_LOCATION_PRINT
-# if defined ]b4_api_PREFIX[LTYPE_IS_TRIVIAL && ]b4_api_PREFIX[LTYPE_IS_TRIVIAL
-#  define YY_LOCATION_PRINT(File, Loc)			\
-     fprintf (File, "%d.%d-%d.%d",			\
-	      (Loc).first_line, (Loc).first_column,	\
-	      (Loc).last_line,  (Loc).last_column)
-# else
-#  define YY_LOCATION_PRINT(File, Loc) ((void) 0)
-# endif
-#endif]],
-[[/* This macro is provided for backward compatibility. */
-#ifndef YY_LOCATION_PRINT
-# define YY_LOCATION_PRINT(File, Loc) ((void) 0)
-#endif]])[
-
 
 /* YYLEX -- calling `yylex' with the right arguments.  */
 #ifdef YYLEX_PARAM
@@ -1426,7 +1410,7 @@ b4_c_function_def([[yyparse]], [[int]], b4_parse_param)[
   yypstate *yyps_local;]b4_pure_if([[
   int yychar;
   YYSTYPE yylval;]b4_locations_if([[
-  YYLTYPE yylloc;]])])[
+  YYLTYPE yylloc][]b4_yyloc_default[;]])])[
   if (yyps)
     yyps_local = yyps;
   else
@@ -1506,7 +1490,7 @@ b4_c_function_def([[yyparse]], [[int]], b4_parse_param)[
   [[[yypstate *yyps]], [[yyps]]]b4_pure_if([,
   [[[int yypushed_char]], [[yypushed_char]]],
   [[[YYSTYPE const *yypushed_val]], [[yypushed_val]]]b4_locations_if([,
-  [[[YYLTYPE const *yypushed_loc]], [[yypushed_loc]]]])])m4_ifset([b4_parse_param], [,
+  [[[YYLTYPE *yypushed_loc]], [[yypushed_loc]]]])])m4_ifset([b4_parse_param], [,
   b4_parse_param]))], [[
 
 
@@ -1557,9 +1541,9 @@ b4_c_function_def([[yyparse]], [[int]], b4_parse_param)[
       goto yyread_pushed_token;
     }]])[
 
-  yyss = yyssa;
+  yyssp = yyss = yyssa;
-  yyvs = yyvsa;]b4_locations_if([[
+  yyvsp = yyvs = yyvsa;]b4_locations_if([[
-  yyls = yylsa;]])[
+  yylsp = yyls = yylsa;]])[
   yystacksize = YYINITDEPTH;]b4_lac_if([[
 
   yyes = yyesa;
@@ -1573,31 +1557,17 @@ b4_c_function_def([[yyparse]], [[int]], b4_parse_param)[
   yyerrstatus = 0;
   yynerrs = 0;
   yychar = YYEMPTY; /* Cause a token to be read.  */
-
+]m4_ifdef([b4_initial_action], [
-  /* Initialize stack pointers.
-     Waste one element of value and location stack
-     so that they stay on the same level as the state stack.
-     The wasted elements are never initialized.  */
-  yyssp = yyss;
-  yyvsp = yyvs;]b4_locations_if([[
-  yylsp = yyls;]])[
-
-  YYLVAL_INITIALIZE ();]b4_locations_if([[
-#if defined ]b4_api_PREFIX[LTYPE_IS_TRIVIAL && ]b4_api_PREFIX[LTYPE_IS_TRIVIAL
-  /* Initialize the default location before parsing starts.  */
-  yylloc.first_line   = yylloc.last_line   = ]b4_location_initial_line[;
-  yylloc.first_column = yylloc.last_column = ]b4_location_initial_column[;
-#endif]])
-m4_ifdef([b4_initial_action],[
 b4_dollar_pushdef([m4_define([b4_dollar_dollar_used])yylval], [],
-                  [m4_define([b4_at_dollar_used])yylloc])dnl
+               [m4_define([b4_at_dollar_used])dnl
+b4_push_if([b4_pure_if([*])yypushed_loc], [yylloc])])dnl
 /* User initialization code.  */
 b4_user_initial_action
 b4_dollar_popdef[]dnl
 m4_ifdef([b4_dollar_dollar_used],[[  yyvsp[0] = yylval;
-]])dnl
-m4_ifdef([b4_at_dollar_used], [[  yylsp[0] = yylloc;
 ]])])dnl
+b4_locations_if([[  yylsp[0] = ]b4_push_if([b4_pure_if([*])yypushed_loc], [yylloc])[;
+]])dnl
 [  goto yysetstate;
 
 /*------------------------------------------------------------.

src/scan-gram.l

@@ -291,7 +291,7 @@ splice	 (\\[ \f\t\v]*\n)*
   }
 
   . {
-    complain_at (*loc, _("invalid character: %s"), quote (yytext));
+    complain_at (*loc, _("invalid character: %s"), quote_mem (yytext, yyleng));
   }
 
   <<EOF>> {
@@ -375,7 +375,7 @@ splice	 (\\[ \f\t\v]*\n)*
   }
   . {
     complain_at (*loc, _("invalid character in bracketed name: %s"),
-		 quote (yytext));
+		 quote_mem (yytext, yyleng));
   }
   <<EOF>> {
     BEGIN bracketed_id_context_state;

tests/actions.at

@@ -69,6 +69,76 @@ AT_PARSER_CHECK([./input], 0,
 AT_CLEANUP
 
 
+## ------------------ ##
+## Initial location.  ##
+## ------------------ ##
+
+# AT_TEST(SKELETON-NAME, DIRECTIVES)
+# ----------------------------------
+# Check the the initial location is correct.
+m4_pushdef([AT_TEST],
+[AT_SETUP([Initial location: $1 $2])
+
+AT_BISON_OPTION_PUSHDEFS([%locations %skeleton "$1" $2 %parse-param { int x }])
+AT_DATA_GRAMMAR([[input.y]],
+[[%defines /* FIXME: Required by lalr1.cc in Bison 2.6. */
+%locations
+%debug
+%skeleton "$1"
+$2
+%parse-param { int x } // Useless, but used to force yyerror purity.
+%code
+{
+# include <stdio.h>
+# include <stdlib.h> // getenv
+]AT_YYERROR_DECLARE[
+]AT_YYLEX_DECLARE[
+}
+%%
+exp: { ]AT_SKEL_CC_IF([[std::cerr << @$ << std::endl]],
+                      [[YY_LOCATION_PRINT(stderr, @$); fputc ('\n', stderr)]])[; }
+%%
+]AT_YYERROR_DEFINE[
+
+]AT_YYLEX_PROTOTYPE[
+{]AT_PURE_IF([
+  YYUSE(lvalp);
+  YYUSE(llocp);], [AT_SKEL_CC_IF([
+  YYUSE(lvalp);
+  YYUSE(llocp);])])[
+  return 'x';
+}
+
+int
+main (void)
+{]AT_SKEL_CC_IF([[
+  yy::parser p (0);
+  p.set_debug_level (!!getenv("YYDEBUG"));
+  return p.parse ();]], [[
+  yydebug = !!getenv("YYDEBUG");
+  return !!yyparse (0);]])[
+}
+]])
+
+AT_FULL_COMPILE([input])
+AT_PARSER_CHECK([./input], 1, [],
+[[1.1
+1.1: syntax error
+]])
+AT_BISON_OPTION_POPDEFS
+AT_CLEANUP
+])
+
+## FIXME: test Java, and iterate over skeletons.
+AT_TEST([yacc.c])
+AT_TEST([yacc.c], [%define api.pure])
+AT_TEST([yacc.c], [%define api.push-pull both])
+AT_TEST([yacc.c], [%define api.push-pull both %define api.pure])
+AT_TEST([glr.c])
+AT_TEST([lalr1.cc])
+AT_TEST([glr.cc])
+
+m4_popdef([AT_TEST])
 
 
 
@@ -159,8 +229,7 @@ main (void)
 }
 ]])
 
-AT_BISON_CHECK([[-o input.c input.y]])
+AT_FULL_COMPILE([input])
-AT_COMPILE([[input]])
 AT_PARSER_CHECK([[./input]], [[0]],
 [[6
 ]])
@@ -305,7 +374,7 @@ line:
       $$ = -1;
       V(line,  $$, @$, ": ");
       V('(',   $1, @1, " ");
-      fprintf (stderr, "error (@%d-%d) ", RANGE(@2));
+      fprintf (stderr, "error (@%d-%d) ", RANGE (@2));
       V(')',   $3, @3, "\n");
     }
 ;
@@ -557,7 +626,7 @@ AT_BISON_OPTION_POPDEFS
 # AT_CHECK_PRINTER_AND_DESTRUCTOR([BISON-OPTIONS], [UNION-FLAG], [SKIP_FLAG])
 # ---------------------------------------------------------------------------
 m4_define([AT_CHECK_PRINTER_AND_DESTRUCTOR],
-[AT_SETUP([Printers and Destructors $2: $1])
+[AT_SETUP([Printers and Destructors$2]m4_ifval([$1], [[: $1]]))
 
 $3
 _AT_CHECK_PRINTER_AND_DESTRUCTOR($[1], $[2], $[3], $[4],
@@ -572,13 +641,13 @@ AT_CLEANUP
 
 
 AT_CHECK_PRINTER_AND_DESTRUCTOR([])
-AT_CHECK_PRINTER_AND_DESTRUCTOR([], [with union])
+AT_CHECK_PRINTER_AND_DESTRUCTOR([], [ with union])
 
 AT_CHECK_PRINTER_AND_DESTRUCTOR([%defines %skeleton "lalr1.cc"])
-AT_CHECK_PRINTER_AND_DESTRUCTOR([%defines %skeleton "lalr1.cc"], [with union])
+AT_CHECK_PRINTER_AND_DESTRUCTOR([%defines %skeleton "lalr1.cc"], [ with union])
 
 AT_CHECK_PRINTER_AND_DESTRUCTOR([%glr-parser])
-AT_CHECK_PRINTER_AND_DESTRUCTOR([%glr-parser], [with union])
+AT_CHECK_PRINTER_AND_DESTRUCTOR([%glr-parser], [ with union])
 
 
 
@@ -595,10 +664,6 @@ AT_DATA_GRAMMAR([[input.y]],
 [[%error-verbose
 %debug
 %locations
-%initial-action {
-  @$.first_line = @$.last_line = 1;
-  @$.first_column = @$.last_column = 1;
-}
 
 %{
 # include <stdio.h>
@@ -656,29 +721,29 @@ AT_PARSER_CHECK([./input], 1,
 ]],
 [[Starting parse
 Entering state 0
-Reading a token: Next token is token 'a' (1.1-1.1: <> printer for 'a' @ 1)
+Reading a token: Next token is token 'a' (1.1: <> printer for 'a' @ 1)
-Shifting token 'a' (1.1-1.1: <> printer for 'a' @ 1)
+Shifting token 'a' (1.1: <> printer for 'a' @ 1)
 Entering state 1
-Reading a token: Next token is token 'b' (1.2-1.2: 'b'/'c' printer for 'b' @ 2)
+Reading a token: Next token is token 'b' (1.2: 'b'/'c' printer for 'b' @ 2)
-Shifting token 'b' (1.2-1.2: 'b'/'c' printer for 'b' @ 2)
+Shifting token 'b' (1.2: 'b'/'c' printer for 'b' @ 2)
 Entering state 3
-Reading a token: Next token is token 'c' (1.3-1.3: 'b'/'c' printer for 'c' @ 3)
+Reading a token: Next token is token 'c' (1.3: 'b'/'c' printer for 'c' @ 3)
-Shifting token 'c' (1.3-1.3: 'b'/'c' printer for 'c' @ 3)
+Shifting token 'c' (1.3: 'b'/'c' printer for 'c' @ 3)
 Entering state 5
-Reading a token: Next token is token 'd' (1.4-1.4: <> printer for 'd' @ 4)
+Reading a token: Next token is token 'd' (1.4: <> printer for 'd' @ 4)
-Shifting token 'd' (1.4-1.4: <> printer for 'd' @ 4)
+Shifting token 'd' (1.4: <> printer for 'd' @ 4)
 Entering state 6
 Reading a token: Now at end of input.
-1.5-4: syntax error, unexpected $end, expecting 'e'
+1.5: syntax error, unexpected $end, expecting 'e'
-Error: popping token 'd' (1.4-1.4: <> printer for 'd' @ 4)
+Error: popping token 'd' (1.4: <> printer for 'd' @ 4)
 Stack now 0 1 3 5
-Error: popping token 'c' (1.3-1.3: 'b'/'c' printer for 'c' @ 3)
+Error: popping token 'c' (1.3: 'b'/'c' printer for 'c' @ 3)
 Stack now 0 1 3
-Error: popping token 'b' (1.2-1.2: 'b'/'c' printer for 'b' @ 2)
+Error: popping token 'b' (1.2: 'b'/'c' printer for 'b' @ 2)
 Stack now 0 1
-Error: popping token 'a' (1.1-1.1: <> printer for 'a' @ 1)
+Error: popping token 'a' (1.1: <> printer for 'a' @ 1)
 Stack now 0
-Cleanup: discarding lookahead token $end (1.5-1.5: )
+Cleanup: discarding lookahead token $end (1.5: )
 Stack now 0
 ]])
 
@@ -829,10 +894,6 @@ AT_DATA_GRAMMAR([[input]]$1[[.y]],
 [[%error-verbose
 %debug
 %locations
-%initial-action {
-  @$.first_line = @$.last_line = 1;
-  @$.first_column = @$.last_column = 1;
-}
 
 %{
 # include <stdio.h>
@@ -899,16 +960,16 @@ AT_PARSER_CHECK([./input$1], 0,
 ]],
 [[Starting parse
 Entering state 0
-Reducing stack by rule 1 (line 46):
+Reducing stack by rule 1 (line 42):
--> $$ = nterm start (1.1-1.1: <]]kind[[> for 'S' @ 1)
+-> $$ = nterm start (1.1: <]]kind[[> for 'S' @ 1)
 Stack now 0
 Entering state 1
 Reading a token: Now at end of input.
-Shifting token END (1.1-1.1: <]]kind[[> for 'E' @ 1)
+Shifting token END (1.1: <]]kind[[> for 'E' @ 1)
 Entering state 2
 Stack now 0 1 2
-Cleanup: popping token END (1.1-1.1: <]]kind[[> for 'E' @ 1)
+Cleanup: popping token END (1.1: <]]kind[[> for 'E' @ 1)
-Cleanup: popping nterm start (1.1-1.1: <]]kind[[> for 'S' @ 1)
+Cleanup: popping nterm start (1.1: <]]kind[[> for 'S' @ 1)
 ]])
 
 m4_popdef([kind])
@@ -1237,7 +1298,7 @@ AT_CHECK_ACTION_LOCATIONS([[%printer]])
 # Check that we can used qualified $$ (v.g., $<type>$) not only in
 # rule actions, but also where $$ is valid: %printer and %destructor.
 #
-# FIXME: Not actually checking %desctructor, but it's the same code as
+# FIXME: Not actually checking %destructor, but it's the same code as
 # %printer...
 #
 # To do that, use a semantic value that has two fields (sem_type),
@@ -1295,17 +1356,6 @@ AT_DATA_GRAMMAR([[input.y]],
 %printer { report (yyo, $<ival>$, $$      ); } <fval>;
 %printer { report (yyo, $<ival>$, $<fval>$); } <>;
 
-]AT_SKEL_CC_IF([[
-/* The lalr1.cc skeleton, for backward compatibility, defines
-   a constructor for position that initializes the filename.  The
-   glr.cc skeleton does not (and in fact cannot: location/position
-   are stored in a union, from which objects with constructors are
-   excluded in C++). */
-%initial-action {
-  @$.initialize ();
-}
-]])[
-
 %initial-action
 {
   $<ival>$ = 42;

tests/calc.at

@@ -178,18 +178,7 @@ read_signed_integer (]AT_YYLEX_FORMALS[)
 
 ]AT_YYLEX_PROTOTYPE[
 {
-  static int init = 1;
   int c;
-
-  if (init)
-    {
-      init = 0;
-]AT_LOCATION_IF([
-      AT_LOC_LAST_COLUMN = 1;
-      AT_LOC_LAST_LINE = 1;
-])[
-    }
-
   /* Skip current token, then white spaces.  */
   do
     {
@@ -261,6 +250,8 @@ AT_SKEL_CC_IF(
 {
   semantic_value ival;
 };
+%printer { ]AT_SKEL_CC_IF([[yyoutput << $$]],
+                          [[fprintf (yyoutput, "%d", $$)]])[; } <ival>;
 
 %code provides
 {
@@ -280,24 +271,16 @@ AT_SKEL_CC_IF(
 FILE *input;
 static int power (int base, int exponent);
 
-]AT_SKEL_CC_IF(,
+]AT_YYERROR_DECLARE[
-[static void yyerror (AT_YYERROR_ARG_LOC_IF([AT_YYLTYPE *llocp, ])
-                     AT_PARAM_IF([semantic_value *result, int *count, ])
-                     const char *s
-                     );])[
 ]AT_YYLEX_DECLARE_EXTERN[
 }
 
-]AT_SKEL_CC_IF([AT_LOCATION_TYPE_IF([], [[
+]AT_SKEL_CC_IF([AT_LOCATION_TYPE_IF([[
-/* The lalr1.cc skeleton, for backward compatibility, defines
+%initial-action
-   a constructor for position that initializes the filename.  The
+{
-   glr.cc skeleton does not (and in fact cannot: location/position
+  @$.first.l = @$.first.c = 1;
-   are stored in a union, from which objects with constructors are
+  @$.last = @$.first;
-   excluded in C++). */
+}]])])[
-%initial-action {
-  @$.initialize ();
-}
-]])])[
 
 /* Bison Declarations */
 %token CALC_EOF 0 "end of input"
@@ -365,27 +348,8 @@ power (int base, int exponent)
       o << '-' << s.last.c - 1;
     return o;
   }
-]])
+]])])[
-AT_YYERROR_DEFINE],
+]AT_YYERROR_DEFINE[
-[/* A C error reporting function.  */
-static void
-yyerror (AT_YYERROR_ARG_LOC_IF([AT_YYLTYPE *llocp, ])
-         AT_PARAM_IF([semantic_value *result, int *count, ])
-         const char *s)
-{
-AT_PARAM_IF([(void) result; (void) count;])
-AT_YYERROR_SEES_LOC_IF([
-  fprintf (stderr, "%d.%d",
-           AT_LOC_FIRST_LINE, AT_LOC_FIRST_COLUMN);
-  if (AT_LOC_FIRST_LINE != AT_LOC_LAST_LINE)
-    fprintf (stderr, "-%d.%d",
-             AT_LOC_LAST_LINE,  AT_LOC_LAST_COLUMN - 1);
-  else if (AT_LOC_FIRST_COLUMN != AT_LOC_LAST_COLUMN - 1)
-    fprintf (stderr, "-%d",
-             AT_LOC_LAST_COLUMN - 1);
-  fprintf (stderr, ": ");])
-  fprintf (stderr, "%s\n", s);
-}])[
 ]AT_DEFINES_IF([],
 [AT_CALC_LEX
 AT_CALC_MAIN])])
@@ -550,7 +514,7 @@ _AT_CHECK_CALC([$1],
                [842])
 
 # Some syntax errors.
-_AT_CHECK_CALC_ERROR([$1], [1], [0 0], [15],
+_AT_CHECK_CALC_ERROR([$1], [1], [1 2], [15],
                      [1.3: syntax error, unexpected number])
 _AT_CHECK_CALC_ERROR([$1], [1], [1//2], [20],
                      [1.3: syntax error, unexpected '/', expecting number or '-' or '(' or '!'])
@@ -594,10 +558,10 @@ calc: error: 4444 != 1])
 
 # The same, but this time exercising explicitly triggered syntax errors.
 # POSIX says the lookahead causing the error should not be discarded.
-_AT_CHECK_CALC_ERROR([$1], [0], [(!) + (0 0) = 1], [102],
+_AT_CHECK_CALC_ERROR([$1], [0], [(!) + (1 2) = 1], [102],
 [1.10: syntax error, unexpected number
 calc: error: 2222 != 1])
-_AT_CHECK_CALC_ERROR([$1], [0], [(- *) + (0 0) = 1], [113],
+_AT_CHECK_CALC_ERROR([$1], [0], [(- *) + (1 2) = 1], [113],
 [1.4: syntax error, unexpected '*', expecting number or '-' or '(' or '!'
 1.12: syntax error, unexpected number
 calc: error: 2222 != 1])

tests/cxx-type.at

@@ -376,7 +376,7 @@ m4_define([_AT_GLR_STDERR],
 ]])
 
 m4_define([_AT_GLR_STDERR_WITH_LOC],
-[[17.5-4: syntax error
+[[17.5: syntax error
 ]])
 
 m4_define([_AT_VERBOSE_GLR_STDERR],
@@ -384,7 +384,7 @@ m4_define([_AT_VERBOSE_GLR_STDERR],
 ]])
 
 m4_define([_AT_VERBOSE_GLR_STDERR_WITH_LOC],
-[[17.5-4: syntax error, unexpected ID, expecting '=' or '+' or ')'
+[[17.5: syntax error, unexpected ID, expecting '=' or '+' or ')'
 ]])
 
 ## ---------------------------------------------------- ##

tests/glr-regression.at

@@ -1543,13 +1543,6 @@ AT_DATA_GRAMMAR([glr-regr17.y],
   ]AT_YYLEX_DECLARE[
 %}
 
-%initial-action {
-  @$.first_line = 1;
-  @$.first_column = 1;
-  @$.last_line = 1;
-  @$.last_column = 1;
-}
-
 %%
 
 /* Tests the case of an empty RHS that has inherited the location of the

tests/headers.at

@@ -240,9 +240,6 @@ AT_TEST([x7], [%define api.push-pull both])
 AT_TEST([x8], [%define api.pure %define api.push-pull both])
 #AT_TEST([x5], [%locations %language "c++" %glr-parser])
 
-AT_COMPILE_CXX([parser], [[x[1-8].o -DCC_IS_CXX=$CC_IS_CXX main.cc]])
-AT_CHECK([./parser], [0], [[expout]])
-
 # Check that the headers are self-contained, and protected against
 # multiple inclusions.  While at it, check they are sane for C++.
 for h in *.h *.hh
@@ -256,6 +253,12 @@ EOF
   AT_COMPILE_CXX([$h.o])
 done
 
+# Do this late, so that other checks have been performed.
+AT_SKIP_IF_CANNOT_LINK_C_AND_CXX
+
+AT_COMPILE_CXX([parser], [[x[1-8].o -DCC_IS_CXX=$CC_IS_CXX main.cc]])
+AT_CHECK([./parser], [0], [[expout]])
+
 m4_popdef([AT_TEST])
 
 AT_CLEANUP

tests/input.at

@@ -20,6 +20,67 @@ AT_BANNER([[Input Processing.]])
 # Mostly test that we are robust to mistakes.
 
 
+## ---------------- ##
+## Invalid inputs.  ##
+## ---------------- ##
+
+AT_SETUP([Invalid inputs])
+
+AT_DATA([input.y],
+[[\000\001\002\377?
+%%
+?
+default: 'a' }
+%&
+%a-does-not-exist
+%-
+%{
+]])
+AT_CHECK([[$PERL -pi -e 's/\\(\d{3})/chr(oct($1))/ge' input.y || exit 77]])
+
+AT_BISON_CHECK([input.y], [1], [],
+[[input.y:1.1: error: invalid character: '\0'
+input.y:1.1: error: invalid character: '\001'
+input.y:1.1: error: invalid character: '\002'
+input.y:1.1: error: invalid character: '\377'
+input.y:1.2: error: invalid character: '?'
+input.y:3.1: error: invalid character: '?'
+input.y:4.14: error: invalid character: '}'
+input.y:5.1: error: invalid character: '%'
+input.y:5.2: error: invalid character: '&'
+input.y:6.1-17: error: invalid directive: '%a-does-not-exist'
+input.y:7.1: error: invalid character: '%'
+input.y:7.2: error: invalid character: '-'
+input.y:8.1-9.0: error: missing '%}' at end of file
+input.y:8.1-9.0: error: syntax error, unexpected %{...%}
+]])
+
+AT_CLEANUP
+
+
+AT_SETUP([Invalid inputs with {}])
+
+# We used to SEGV here.  See
+# http://lists.gnu.org/archive/html/bug-bison/2005-07/msg00053.html
+
+AT_DATA([input.y],
+[[
+%destructor
+%initial-action
+%lex-param
+%parse-param
+%printer
+%union
+]])
+
+AT_BISON_CHECK([input.y], [1], [],
+[[input.y:3.1-15: error: syntax error, unexpected %initial-action, expecting {...}
+]])
+
+AT_CLEANUP
+
+
+
 ## ------------ ##
 ## Invalid $n.  ##
 ## ------------ ##

tests/local.at

@@ -119,6 +119,7 @@ m4_pushdef([AT_SKEL_CC_IF],
 [m4_bmatch([$3], [%language "[Cc]\+\+"\|%skeleton "[a-z0-9]+\.cc"], [$1], [$2])])
 m4_pushdef([AT_SKEL_JAVA_IF],
 [m4_bmatch([$3], [%language "[Jj][Aa][Vv][Aa]"\|%skeleton "[a-z0-9]+\.java"], [$1], [$2])])
+# The target language: "c", "c++", or "java".
 m4_pushdef([AT_LANG],
 [AT_SKEL_JAVA_IF([java],
                  [AT_SKEL_CC_IF([c++],
@@ -140,6 +141,12 @@ m4_pushdef([AT_LOCATION_TYPE_IF],
 [m4_bmatch([$3], [%define \(api\.location\.type\|location_type\)], [$1], [$2])])
 m4_pushdef([AT_PARAM_IF],
 [m4_bmatch([$3], [%parse-param], [$1], [$2])])
+# Comma-terminated list of formals parse-parameters.
+# E.g., %parse-param { int x } {int y} -> "int x, int y, ".
+m4_pushdef([AT_PARSE_PARAMS])
+m4_bpatsubst([$3], [%parse-param { *\([^{}]*[^{} ]\) *}],
+             [m4_append([AT_PARSE_PARAMS], [\1, ])])
+
 m4_pushdef([AT_PURE_IF],
 [m4_bmatch([$3], [%define  *api\.pure\|%pure-parser],
            [m4_bmatch([$3], [%define  *api\.pure *"?false"?], [$2], [$1])],
@@ -241,6 +248,7 @@ m4_popdef([AT_GLR_OR_PARAM_IF])
 m4_popdef([AT_PURE_AND_LOC_IF])
 m4_popdef([AT_LOCATION_TYPE_IF])
 m4_popdef([AT_LOCATION_IF])
+m4_popdef([AT_PARSE_PARAMS])
 m4_popdef([AT_PARAM_IF])
 m4_popdef([AT_LEXPARAM_IF])
 m4_popdef([AT_YACC_IF])
@@ -350,7 +358,7 @@ static
 # Must be called inside a AT_BISON_OPTION_PUSHDEFS/POPDEFS pair.
 m4_define([AT_YYERROR_FORMALS],
 [m4_case(AT_LANG,
-[c], [AT_YYERROR_ARG_LOC_IF([AT_YYLTYPE *llocp, ])[const char *msg]])[]dnl
+[c], [AT_YYERROR_ARG_LOC_IF([AT_YYLTYPE *llocp, ])AT_PARSE_PARAMS [const char *msg]])[]dnl
 ])
 
 m4_define([AT_YYERROR_PROTOTYPE],
@@ -374,16 +382,11 @@ m4_define([AT_YYERROR_DEFINE],
 /* A C error reporting function.  */
 static
 ]AT_YYERROR_PROTOTYPE[
-{
+{]m4_bpatsubst(m4_defn([AT_PARSE_PARAMS]),
-]AT_YYERROR_SEES_LOC_IF([[
+              [[^,]+[^A-Za-z_0-9]\([A-Za-z_][A-Za-z_0-9]*\), *], [
-  fprintf (stderr, "%d.%d",
+  YYUSE(\1);])dnl
-           ]AT_LOC_FIRST_LINE[, ]AT_LOC_FIRST_COLUMN[);
+AT_YYERROR_SEES_LOC_IF([[
-  if (]AT_LOC_FIRST_LINE[ != ]AT_LOC_LAST_LINE[)
+  YY_LOCATION_PRINT (stderr, ]AT_LOC[);
-    fprintf (stderr, "-%d.%d",
-             ]AT_LOC_LAST_LINE[,  ]AT_LOC_LAST_COLUMN[ - 1);
-  else if (]AT_LOC_FIRST_COLUMN[ != ]AT_LOC_LAST_COLUMN[ - 1)
-    fprintf (stderr, "-%d",
-             ]AT_LOC_LAST_COLUMN[ - 1);
   fprintf (stderr, ": ");]])[
   fprintf (stderr, "%s\n", msg);
 }]],
@@ -680,6 +683,44 @@ m4_define([AT_FULL_COMPILE],
 ])
 
 
+
+# AT_SKIP_IF_CANNOT_LINK_C_AND_CXX
+# --------------------------------
+# Check that we can link together C and C++ objects.
+m4_define([AT_SKIP_IF_CANNOT_LINK_C_AND_CXX],
+[AT_DATA([c-and-cxx.h],
+[[#ifdef __cplusplus
+extern "C"
+{
+#endif
+  int fortytwo (void);
+#ifdef __cplusplus
+}
+#endif
+]])
+AT_DATA([c-only.c],
+[[#include "c-and-cxx.h"
+int
+main (void)
+{
+  return fortytwo () == 42 ? 0 : 1;
+}
+]])
+AT_DATA([cxx-only.cc],
+[[#include "c-and-cxx.h"
+int fortytwo ()
+{
+  return 42;
+}
+]])
+AT_COMPILE([c-only.o], [c-only.c])
+AT_COMPILE_CXX([cxx-only.o], [cxx-only.cc])
+AT_CHECK([$CXX $CXXFLAGS $CPPFLAGS $LDFLAGS c-only.o cxx-only.o -o c-and-cxx ||
+          exit 77], [ignore], [ignore])
+AT_CHECK([./c-and-cxx])
+])
+
+
 ## ---------------------------- ##
 ## Running a generated parser.  ##
 ## ---------------------------- ##

tests/named-refs.at

@@ -443,14 +443,19 @@ AT_SETUP([Stray symbols in brackets])
 AT_DATA_GRAMMAR([test.y],
 [[
 %%
-start: foo[ /* aaa */ *&-.+ ] bar
+start: foo[ /* aaa */ *&-.+\000\001\002\377 ] bar
   { s = $foo; }
 ]])
+AT_CHECK([[$PERL -pi -e 's/\\(\d{3})/chr(oct($1))/ge' test.y || exit 77]])
 AT_BISON_CHECK([-o test.c test.y], 1, [],
 [[test.y:11.23: error: invalid character in bracketed name: '*'
 test.y:11.24: error: invalid character in bracketed name: '&'
 test.y:11.25: error: invalid character in bracketed name: '-'
 test.y:11.27: error: invalid character in bracketed name: '+'
+test.y:11.28: error: invalid character in bracketed name: '\0'
+test.y:11.28: error: invalid character in bracketed name: '\001'
+test.y:11.28: error: invalid character in bracketed name: '\002'
+test.y:11.28: error: invalid character in bracketed name: '\377'
 ]])
 AT_CLEANUP
 

tests/regression.at

@@ -1597,14 +1597,14 @@ AT_CLEANUP
 m4_pushdef([AT_TEST],
 [AT_SETUP([[Lex and parse params: $1]])
 
-AT_BISON_OPTION_PUSHDEFS([%locations %skeleton $1])
+AT_BISON_OPTION_PUSHDEFS([%locations %skeleton "$1" %parse-param { int x } %parse-param { int y }])
 
 ## FIXME: Improve parsing of parse-param and use the generated
 ## yyerror.
 AT_DATA_GRAMMAR([input.y],
 [[%defines
 %locations
-%skeleton $1
+%skeleton "$1"
 %union { int ival; }
 %parse-param { int x }
 // Spaces, tabs, and new lines.
@@ -1618,26 +1618,18 @@ AT_DATA_GRAMMAR([input.y],
 #include <stdio.h>
 #include <stdlib.h>
 
-]AT_SKEL_CC_IF([], [[
+  ]AT_YYERROR_DECLARE[
-static
-void
-yyerror (int x, int y, const char *msg)
-{
-  fprintf (stderr, "x: %d, y: %d, %s\n", x, y, msg);
-}]])[
-
   ]AT_YYLEX_DECLARE[
 %}
 
 %%
 exp: 'a' { fprintf (stdout, "x: %d, y: %d\n", x, y); };
 %%
+]AT_YYERROR_DEFINE[
 ]AT_YYLEX_DEFINE(["a"])[
 
 ]AT_SKEL_CC_IF(
-[AT_YYERROR_DEFINE
+[int
-
-int
 yyparse (int x, int y)
 {
   yy::parser parser(x, y);
@@ -1661,9 +1653,9 @@ AT_CLEANUP
 ])
 
 ## FIXME: test Java, and iterate over skeletons.
-AT_TEST("yacc.c")
+AT_TEST([yacc.c])
-AT_TEST("glr.c")
+AT_TEST([glr.c])
-AT_TEST("lalr1.cc")
+AT_TEST([lalr1.cc])
-AT_TEST("glr.cc")
+AT_TEST([glr.cc])
 
 m4_popdef([AT_TEST])