portability: fix spawning on at least FreeBSD 8 and FreeBSD 9.

[bison.git] / src / output.c

diff --git a/src/output.c b/src/output.c
index afcec94956b05cce53ea896b3533438c82c5d076..86d9728b7113be5a9ff79904e523ed5939bd11ed 100644 (file)
--- a/src/output.c
+++ b/src/output.c
@@ -1,7 +1,7 @@
 /* Output the generated parsing program for Bison.
 
 /* Output the generated parsing program for Bison.
 

-   Copyright (C) 1984, 1986, 1989, 1992, 2000, 2001, 2002, 2003, 2004,
-   2005, 2006, 2007, 2008 Free Software Foundation, Inc.
+   Copyright (C) 1984, 1986, 1989, 1992, 2000-2010 Free Software
+   Foundation, Inc.

 
    This file is part of Bison, the GNU Compiler Compiler.
 
 
    This file is part of Bison, the GNU Compiler Compiler.
 


@@ -24,15 +24,16 @@
 #include <configmake.h>
 #include <error.h>
 #include <get-errno.h>
 #include <configmake.h>
 #include <error.h>
 #include <get-errno.h>

+#include <pipe.h>

 #include <quotearg.h>
 #include <quotearg.h>

-#include <subpipe.h>

 #include <timevar.h>
 #include <timevar.h>

+#include <wait-process.h>

 
 #include "complain.h"
 #include "files.h"
 #include "getargs.h"
 #include "gram.h"
 
 #include "complain.h"
 #include "files.h"
 #include "getargs.h"
 #include "gram.h"

-#include "muscle_tab.h"
+#include "muscle-tab.h"

 #include "output.h"
 #include "reader.h"
 #include "scan-code.h"    /* max_left_semantic_context */
 #include "output.h"
 #include "reader.h"
 #include "scan-code.h"    /* max_left_semantic_context */


@@ -40,6 +41,7 @@
 #include "symtab.h"
 #include "tables.h"
 
 #include "symtab.h"
 #include "tables.h"
 

+# define ARRAY_CARDINALITY(Array) (sizeof (Array) / sizeof *(Array))

 
 static struct obstack format_obstack;
 
 
 static struct obstack format_obstack;
 


@@ -104,7 +106,6 @@ GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_int_table, int)
 GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_base_table, base_number)
 GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_rule_number_table, rule_number)
 GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_symbol_number_table, symbol_number)
 GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_base_table, base_number)
 GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_rule_number_table, rule_number)
 GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_symbol_number_table, symbol_number)

-GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_item_number_table, item_number)

 GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_state_number_table, state_number)
 
 
 GENERATE_MUSCLE_INSERT_TABLE(muscle_insert_state_number_table, state_number)
 
 


@@ -201,38 +202,27 @@ prepare_symbols (void)
 }
 
 
 }
 
 

-/*-------------------------------------------------------------.
-| Prepare the muscles related to the rules: rhs, prhs, r1, r2, |
-| rline, dprec, merger.                                        |
-`-------------------------------------------------------------*/
+/*----------------------------------------------------------------.
+| Prepare the muscles related to the rules: r1, r2, rline, dprec, |
+| merger.                                                         |
+`----------------------------------------------------------------*/

 
 static void
 prepare_rules (void)
 {
 
 static void
 prepare_rules (void)
 {

-  rule_number r;
-  unsigned int i = 0;
-  item_number *rhs = xnmalloc (nritems, sizeof *rhs);
-  unsigned int *prhs = xnmalloc (nrules, sizeof *prhs);

   unsigned int *rline = xnmalloc (nrules, sizeof *rline);
   symbol_number *r1 = xnmalloc (nrules, sizeof *r1);
   unsigned int *r2 = xnmalloc (nrules, sizeof *r2);
   int *dprec = xnmalloc (nrules, sizeof *dprec);
   int *merger = xnmalloc (nrules, sizeof *merger);
 
   unsigned int *rline = xnmalloc (nrules, sizeof *rline);
   symbol_number *r1 = xnmalloc (nrules, sizeof *r1);
   unsigned int *r2 = xnmalloc (nrules, sizeof *r2);
   int *dprec = xnmalloc (nrules, sizeof *dprec);
   int *merger = xnmalloc (nrules, sizeof *merger);
 

+  rule_number r;

   for (r = 0; r < nrules; ++r)
     {
   for (r = 0; r < nrules; ++r)
     {

-      item_number *rhsp = NULL;
-      /* Index of rule R in RHS. */
-      prhs[r] = i;
-      /* RHS of the rule R. */
-      for (rhsp = rules[r].rhs; *rhsp >= 0; ++rhsp)
-       rhs[i++] = *rhsp;

       /* LHS of the rule R. */
       r1[r] = rules[r].lhs->number;
       /* Length of rule R's RHS. */
       /* LHS of the rule R. */
       r1[r] = rules[r].lhs->number;
       /* Length of rule R's RHS. */

-      r2[r] = i - prhs[r];
-      /* Separator in RHS. */
-      rhs[i++] = -1;
+      r2[r] = rule_rhs_length(&rules[r]);

       /* Line where rule was defined. */
       rline[r] = rules[r].location.start.line;
       /* Dynamic precedence (GLR).  */
       /* Line where rule was defined. */
       rline[r] = rules[r].location.start.line;
       /* Dynamic precedence (GLR).  */


@@ -240,10 +230,7 @@ prepare_rules (void)
       /* Merger-function index (GLR).  */
       merger[r] = rules[r].merger;
     }
       /* Merger-function index (GLR).  */
       merger[r] = rules[r].merger;
     }

-  aver (i == nritems);

 
 

-  muscle_insert_item_number_table ("rhs", rhs, ritem[0], 1, nritems);
-  muscle_insert_unsigned_int_table ("prhs", prhs, 0, 0, nrules);

   muscle_insert_unsigned_int_table ("rline", rline, 0, 0, nrules);
   muscle_insert_symbol_number_table ("r1", r1, 0, 0, nrules);
   muscle_insert_unsigned_int_table ("r2", r2, 0, 0, nrules);
   muscle_insert_unsigned_int_table ("rline", rline, 0, 0, nrules);
   muscle_insert_symbol_number_table ("r1", r1, 0, 0, nrules);
   muscle_insert_unsigned_int_table ("r2", r2, 0, 0, nrules);


@@ -253,8 +240,6 @@ prepare_rules (void)
   MUSCLE_INSERT_INT ("rules_number", nrules);
   MUSCLE_INSERT_INT ("max_left_semantic_context", max_left_semantic_context);
 
   MUSCLE_INSERT_INT ("rules_number", nrules);
   MUSCLE_INSERT_INT ("max_left_semantic_context", max_left_semantic_context);
 

-  free (rhs);
-  free (prhs);

   free (rline);
   free (r1);
   free (r2);
   free (rline);
   free (r1);
   free (r2);


@@ -287,7 +272,7 @@ prepare_states (void)
 | Compare two symbols by type-name, and then by number.  |
 `-------------------------------------------------------*/
 
 | Compare two symbols by type-name, and then by number.  |
 `-------------------------------------------------------*/
 

-int
+static int

 symbol_type_name_cmp (const symbol **lhs, const symbol **rhs)
 {
   int res = UNIQSTR_CMP((*lhs)->type_name, (*rhs)->type_name);
 symbol_type_name_cmp (const symbol **lhs, const symbol **rhs)
 {
   int res = UNIQSTR_CMP((*lhs)->type_name, (*rhs)->type_name);


@@ -302,7 +287,7 @@ symbol_type_name_cmp (const symbol **lhs, const symbol **rhs)
 `----------------------------------------------------------------*/
 
 static symbol **
 `----------------------------------------------------------------*/
 
 static symbol **

-symbols_by_type_name ()
+symbols_by_type_name (void)

 {
   typedef int (*qcmp_type) (const void *, const void *);
   symbol **res = xmemdup (symbols, nsyms * sizeof *res);
 {
   typedef int (*qcmp_type) (const void *, const void *);
   symbol **res = xmemdup (symbols, nsyms * sizeof *res);


@@ -396,49 +381,35 @@ merger_output (FILE *out)
 }
 
 
 }
 
 

-/*----------------------------------.
-| Whether S is a valid identifier.  |
-`----------------------------------*/
-
-static bool
-is_identifier (uniqstr s)
-{
-  static char const alphanum[26 + 26 + 1 + 10] =
-    "abcdefghijklmnopqrstuvwxyz"
-    "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
-    "_"
-    "0123456789";
-  if (!s || ! memchr (alphanum, *s, sizeof alphanum - 10))
-    return false;
-  for (++s; *s; ++s)
-    if (! memchr (alphanum, *s, sizeof alphanum))
-      return false;
-  return true;
-}
-
-/*---------------------------------------.
-| Output the symbol definitions to OUT.  |
-`---------------------------------------*/
+/*---------------------------------------------.
+| Prepare the muscles for symbol definitions.  |
+`---------------------------------------------*/

 
 static void
 
 static void

-symbol_definitions_output (FILE *out)
+prepare_symbol_definitions (void)

 {
   int i;
   for (i = 0; i < nsyms; ++i)
     {
       symbol *sym = symbols[i];
       const char *key;
 {
   int i;
   for (i = 0; i < nsyms; ++i)
     {
       symbol *sym = symbols[i];
       const char *key;

+      const char *value;

 
 #define SET_KEY(Entry)                                                  \
       obstack_fgrow2 (&format_obstack, "symbol(%d, %s)", i, Entry);     \
       obstack_1grow (&format_obstack, 0);                               \
       key = obstack_finish (&format_obstack);
 
 
 #define SET_KEY(Entry)                                                  \
       obstack_fgrow2 (&format_obstack, "symbol(%d, %s)", i, Entry);     \
       obstack_1grow (&format_obstack, 0);                               \
       key = obstack_finish (&format_obstack);
 

-      // Whether the tag is a valid identifier.
-      SET_KEY("tag_is_id");
-      MUSCLE_INSERT_INT (key, is_identifier(sym->tag));
+      // Whether the symbol has an identifier.
+      value = symbol_id_get (sym);
+      SET_KEY("has_id");
+      MUSCLE_INSERT_INT (key, !!value);
+
+      // Its identifier.
+      SET_KEY("id");
+      MUSCLE_INSERT_STRING (key, value ? value : "");

 
 

-      // The inner tag.
+      // Its tag.  Typically for documentation purpose.

       SET_KEY("tag");
       MUSCLE_INSERT_STRING (key, sym->tag);
 
       SET_KEY("tag");
       MUSCLE_INSERT_STRING (key, sym->tag);
 


@@ -452,12 +423,34 @@ symbol_definitions_output (FILE *out)
       SET_KEY("number");
       MUSCLE_INSERT_INT (key, sym->number);
 
       SET_KEY("number");
       MUSCLE_INSERT_INT (key, sym->number);
 

-      SET_KEY("has_type_name");
+      SET_KEY("has_type");

       MUSCLE_INSERT_INT (key, !!sym->type_name);
 
       MUSCLE_INSERT_INT (key, !!sym->type_name);
 

-      SET_KEY("type_name");
+      SET_KEY("type");

       MUSCLE_INSERT_STRING (key, sym->type_name ? sym->type_name : "");
 
       MUSCLE_INSERT_STRING (key, sym->type_name ? sym->type_name : "");
 

+#define CODE_PROP(PropName)                                             \
+      do {                                                              \
+        code_props const *p = symbol_ ## PropName ## _get (sym);        \
+        SET_KEY("has_" #PropName);                                      \
+        MUSCLE_INSERT_INT (key, !!p->code);                             \
+                                                                        \
+        if (p->code)                                                    \
+          {                                                             \
+            SET_KEY(#PropName "_file");                                 \
+            MUSCLE_INSERT_STRING (key, p->location.start.file);         \
+                                                                        \
+            SET_KEY(#PropName "_line");                                 \
+            MUSCLE_INSERT_INT (key, p->location.start.line);            \
+                                                                        \
+            SET_KEY(#PropName);                                         \
+            MUSCLE_INSERT_STRING_RAW (key, p->code);                    \
+          }                                                             \
+      } while (0)
+
+      CODE_PROP(destructor);
+      CODE_PROP(printer);
+#undef CODE_PROP

 #undef SET_KEY
     }
 }
 #undef SET_KEY
     }
 }


@@ -478,73 +471,19 @@ token_definitions_output (FILE *out)
     {
       symbol *sym = symbols[i];
       int number = sym->user_token_number;
     {
       symbol *sym = symbols[i];
       int number = sym->user_token_number;

+      uniqstr id = symbol_id_get (sym);

 
 

-      /* At this stage, if there are literal aliases, they are part of
-        SYMBOLS, so we should not find symbols which are the aliases
-        here.  */
-      aver (number != USER_NUMBER_ALIAS);
-
-      /* Skip error token.  */
-      if (sym == errtoken)
-       continue;
-
-      /* If this string has an alias, then it is necessarily the alias
-        which is to be output.  */
-      if (sym->alias)
-       sym = sym->alias;
-
-      /* Don't output literal chars or strings (when defined only as a
-        string).  Note that must be done after the alias resolution:
-        think about `%token 'f' "f"'.  */
-      if (sym->tag[0] == '\'' || sym->tag[0] == '\"')
-       continue;
-
-      /* Don't #define nonliteral tokens whose names contain periods
-        or '$' (as does the default value of the EOF token).  */
-      if (strchr (sym->tag, '.') || strchr (sym->tag, '$'))
-       continue;
-
-      fprintf (out, "%s[[[%s]], %d]",
-              sep, sym->tag, number);
-      sep = ",\n";
-    }
-  fputs ("])\n\n", out);
-}
-
-
-/*---------------------------------------------------.
-| Output the symbol destructors or printers to OUT.  |
-`---------------------------------------------------*/
+      /* At this stage, if there are literal string aliases, they are
+         part of SYMBOLS, so we should not find their aliased symbols
+         here.  */
+      aver (number != USER_NUMBER_HAS_STRING_ALIAS);

 
 

-static void
-symbol_code_props_output (FILE *out, char const *what,
-                          code_props const *(*get)(symbol const *))
-{
-  int i;
-  char const *sep = "";
-
-  fputs ("m4_define([b4_symbol_", out);
-  fputs (what, out);
-  fputs ("], \n[", out);
-  for (i = 0; i < nsyms; ++i)
-    {
-      symbol *sym = symbols[i];
-      char const *code = (*get) (sym)->code;
-      if (code)
+      /* Skip error token and tokens without identifier.  */
+      if (sym != errtoken && id)

         {
         {

-          location loc = (*get) (sym)->location;
-          /* Filename, lineno,
-             Symbol-name, Symbol-number,
-             code, optional typename.  */
-          fprintf (out, "%s[", sep);
+          fprintf (out, "%s[[[%s]], %d]",
+                   sep, id, number);

           sep = ",\n";
           sep = ",\n";

-          escaped_output (out, loc.start.file);
-          fprintf (out, ", %d, ", loc.start.line);
-          escaped_output (out, sym->tag);
-          fprintf (out, ", %d, [[%s]]", sym->number, code);
-          if (sym->type_name)
-            fprintf (out, ", [[%s]]", sym->type_name);
-          fputc (']', out);

         }
     }
   fputs ("])\n\n", out);
         }
     }
   fputs ("])\n\n", out);


@@ -608,9 +547,6 @@ muscles_output (FILE *out)
 {
   fputs ("m4_init()\n", out);
   merger_output (out);
 {
   fputs ("m4_init()\n", out);
   merger_output (out);

-  symbol_code_props_output (out, "destructors", &symbol_destructor_get);
-  symbol_code_props_output (out, "printers", &symbol_printer_get);
-  symbol_definitions_output (out);

   symbol_numbers_output (out);
   token_definitions_output (out);
   type_names_output (out);
   symbol_numbers_output (out);
   token_definitions_output (out);
   type_names_output (out);


@@ -628,7 +564,7 @@ output_skeleton (void)
 {
   FILE *in;
   int filter_fd[2];
 {
   FILE *in;
   int filter_fd[2];

-  char const *argv[9];
+  char const *argv[10];

   pid_t pid;
 
   /* Compute the names of the package data dir and skeleton files.  */
   pid_t pid;
 
   /* Compute the names of the package data dir and skeleton files.  */


@@ -679,6 +615,19 @@ output_skeleton (void)
   {
     int i = 0;
     argv[i++] = m4;
   {
     int i = 0;
     argv[i++] = m4;

+
+    /* When POSIXLY_CORRECT is set, GNU M4 1.6 and later disable GNU
+       extensions, which Bison's skeletons depend on.  With older M4,
+       it has no effect.  M4 1.4.12 added a -g/--gnu command-line
+       option to make it explicit that a program wants GNU M4
+       extensions even when POSIXLY_CORRECT is set.
+
+       See the thread starting at
+       <http://lists.gnu.org/archive/html/bug-bison/2008-07/msg00000.html>
+       for details.  */
+    if (*M4_GNU_OPTION)
+      argv[i++] = M4_GNU_OPTION;
+

     argv[i++] = "-I";
     argv[i++] = pkgdatadir;
     if (trace_flag & trace_m4)
     argv[i++] = "-I";
     argv[i++] = pkgdatadir;
     if (trace_flag & trace_m4)


@@ -688,28 +637,12 @@ output_skeleton (void)
     argv[i++] = full_m4bison;
     argv[i++] = full_skeleton;
     argv[i++] = NULL;
     argv[i++] = full_m4bison;
     argv[i++] = full_skeleton;
     argv[i++] = NULL;

+    aver (i <= ARRAY_CARDINALITY (argv));

   }
   }

-  /* When POSIXLY_CORRECT is set, some future versions of GNU M4 (most likely
-     2.0) may drop some of the GNU extensions that Bison's skeletons depend
-     upon.  So that the next release of Bison is forward compatible with those
-     future versions of GNU M4, we unset POSIXLY_CORRECT here.
-
-     FIXME: A user might set POSIXLY_CORRECT to affect processes run from
-     macros like m4_syscmd in a custom skeleton.  For now, Bison makes no
-     promises about the behavior of custom skeletons, so this scenario is not a
-     concern.  However, we eventually want to eliminate this shortcoming.  The
-     next release of GNU M4 (1.4.12 or 1.6) will accept the -g command-line
-     option as a no-op, and later releases will accept it to indicate that
-     POSIXLY_CORRECT should be ignored.  Once the GNU M4 versions that accept
-     -g are pervasive, Bison should use -g instead of unsetting
-     POSIXLY_CORRECT.

 
 

-     See the thread starting at
-     <http://lists.gnu.org/archive/html/bug-bison/2008-07/msg00000.html>
-     for details.  */
-  unsetenv ("POSIXLY_CORRECT");
-  init_subpipe ();
-  pid = create_subpipe (argv, filter_fd);
+  /* The ugly cast is because gnulib gets the const-ness wrong.  */
+  pid = create_pipe_bidi ("m4", m4, (char **)(void*)argv, false, true,
+                          true, filter_fd);

   free (full_m4sugar);
   free (full_m4bison);
   free (full_skeleton);
   free (full_m4sugar);
   free (full_m4bison);
   free (full_skeleton);


@@ -717,7 +650,7 @@ output_skeleton (void)
   if (trace_flag & trace_muscles)
     muscles_output (stderr);
   {
   if (trace_flag & trace_muscles)
     muscles_output (stderr);
   {

-    FILE *out = fdopen (filter_fd[0], "w");
+    FILE *out = fdopen (filter_fd[1], "w");

     if (! out)
       error (EXIT_FAILURE, get_errno (),
              "fdopen");
     if (! out)
       error (EXIT_FAILURE, get_errno (),
              "fdopen");


@@ -727,14 +660,17 @@ output_skeleton (void)
 
   /* Read and process m4's output.  */
   timevar_push (TV_M4);
 
   /* Read and process m4's output.  */
   timevar_push (TV_M4);

-  end_of_output_subpipe (pid, filter_fd);
-  in = fdopen (filter_fd[1], "r");
+  in = fdopen (filter_fd[0], "r");

   if (! in)
     error (EXIT_FAILURE, get_errno (),
           "fdopen");
   scan_skel (in);
   if (! in)
     error (EXIT_FAILURE, get_errno (),
           "fdopen");
   scan_skel (in);

+  /* scan_skel should have read all of M4's output.  Otherwise, when we
+     close the pipe, we risk letting M4 report a broken-pipe to the
+     Bison user.  */
+  aver (feof (in));

   xfclose (in);
   xfclose (in);

-  reap_subpipe (pid, m4);
+  wait_subprocess (pid, "m4", false, false, true, true, NULL);

   timevar_pop (TV_M4);
 }
 
   timevar_pop (TV_M4);
 }
 


@@ -751,11 +687,8 @@ prepare (void)
     use_push_for_pull_flag = true;
 
   /* Flags. */
     use_push_for_pull_flag = true;
 
   /* Flags. */

-  MUSCLE_INSERT_BOOL ("debug_flag", debug_flag);

   MUSCLE_INSERT_BOOL ("defines_flag", defines_flag);
   MUSCLE_INSERT_BOOL ("defines_flag", defines_flag);

-  MUSCLE_INSERT_BOOL ("error_verbose_flag", error_verbose);

   MUSCLE_INSERT_BOOL ("glr_flag", glr_parser);
   MUSCLE_INSERT_BOOL ("glr_flag", glr_parser);

-  MUSCLE_INSERT_BOOL ("locations_flag", locations_flag);

   MUSCLE_INSERT_BOOL ("nondeterministic_flag", nondeterministic_parser);
   MUSCLE_INSERT_BOOL ("synclines_flag", !no_lines_flag);
   MUSCLE_INSERT_BOOL ("tag_seen_flag", tag_seen);
   MUSCLE_INSERT_BOOL ("nondeterministic_flag", nondeterministic_parser);
   MUSCLE_INSERT_BOOL ("synclines_flag", !no_lines_flag);
   MUSCLE_INSERT_BOOL ("tag_seen_flag", tag_seen);


@@ -808,6 +741,7 @@ output (void)
   prepare_rules ();
   prepare_states ();
   prepare_actions ();
   prepare_rules ();
   prepare_states ();
   prepare_actions ();

+  prepare_symbol_definitions ();

 
   prepare ();
 
 
   prepare ();