doc: fix environment issues.

[bison.git] / src / scan-gram.l

diff --git a/src/scan-gram.l b/src/scan-gram.l
index 6df72afe0a3da5310dd79b3088fff36b50fd0c45..e2e08f9ab9dbf62253ee67e27d4c59c0d5840c52 100644 (file)
--- a/src/scan-gram.l
+++ b/src/scan-gram.l
@@ -1,12 +1,12 @@
 /* Bison Grammar Scanner                             -*- C -*-
 
 /* Bison Grammar Scanner                             -*- C -*-
 

-   Copyright (C) 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
+   Copyright (C) 2002-2012 Free Software Foundation, Inc.

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

-   This program is free software; you can redistribute it and/or modify
+   This program is free software: you can redistribute it and/or modify

    it under the terms of the GNU General Public License as published by
    it under the terms of the GNU General Public License as published by

-   the Free Software Foundation; either version 2 of the License, or
+   the Free Software Foundation, either version 3 of the License, or

    (at your option) any later version.
 
    This program is distributed in the hope that it will be useful,
    (at your option) any later version.
 
    This program is distributed in the hope that it will be useful,


@@ -15,12 +15,9 @@
    GNU General Public License for more details.
 
    You should have received a copy of the GNU General Public License
    GNU General Public License for more details.
 
    You should have received a copy of the GNU General Public License

-   along with this program; if not, write to the Free Software
-   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
-   02110-1301  USA
-*/
+   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */

 
 

-%option debug nodefault nounput noyywrap never-interactive
+%option debug nodefault noinput nounput noyywrap never-interactive

 %option prefix="gram_" outfile="lex.yy.c"
 
 %{
 %option prefix="gram_" outfile="lex.yy.c"
 
 %{


@@ -30,24 +27,26 @@
 #define gram_wrap() 1
 
 #define FLEX_PREFIX(Id) gram_ ## Id
 #define gram_wrap() 1
 
 #define FLEX_PREFIX(Id) gram_ ## Id

-#include "flex-scanner.h"
+#include <src/flex-scanner.h>

 
 

-#include "complain.h"
-#include "files.h"
-#include "gram.h"
-#include "quotearg.h"
-#include "reader.h"
-#include "uniqstr.h"
+#include <src/complain.h>
+#include <src/files.h>
+#include <src/gram.h>
+#include <quotearg.h>
+#include <src/reader.h>
+#include <src/uniqstr.h>

 
 

+#include <ctype.h>

 #include <mbswidth.h>
 #include <quote.h>
 #include <mbswidth.h>
 #include <quote.h>

+#include <streq.h>

 
 

-#include "scan-gram.h"
+#include <src/scan-gram.h>

 
 #define YY_DECL GRAM_LEX_DECL
 
 
 #define YY_DECL GRAM_LEX_DECL
 

-#define YY_USER_INIT                                   \
-   code_start = scanner_cursor = loc->start;           \
+#define YY_USER_INIT                                    \
+   code_start = scanner_cursor = loc->start;            \

 
 /* Location of scanner cursor.  */
 static boundary scanner_cursor;
 
 /* Location of scanner cursor.  */
 static boundary scanner_cursor;


@@ -57,8 +56,32 @@ static boundary scanner_cursor;
 static size_t no_cr_read (FILE *, char *, size_t);
 #define YY_INPUT(buf, result, size) ((result) = no_cr_read (yyin, buf, size))
 
 static size_t no_cr_read (FILE *, char *, size_t);
 #define YY_INPUT(buf, result, size) ((result) = no_cr_read (yyin, buf, size))
 

+#define RETURN_PERCENT_PARAM(Value)                     \
+  RETURN_VALUE(PERCENT_PARAM, param, param_ ## Value)
+
+#define RETURN_PERCENT_FLAG(Value)                              \
+  RETURN_VALUE(PERCENT_FLAG, uniqstr, uniqstr_new (Value))
+
+#define RETURN_VALUE(Token, Field, Value)       \
+  do {                                          \
+    val->Field = Value;                         \
+    return Token;                               \
+  } while (0)
+
+#define ROLLBACK_CURRENT_TOKEN                                  \
+  do {                                                          \
+    scanner_cursor.column -= mbsnwidth (yytext, yyleng, 0);     \
+    yyless (0);                                                 \
+  } while (0)
+

 /* A string representing the most recently saved token.  */
 /* A string representing the most recently saved token.  */

-char *last_string;
+static char *last_string;
+
+/* Bracketed identifier. */
+static uniqstr bracketed_id_str = 0;
+static location bracketed_id_loc;
+static boundary bracketed_id_start;
+static int bracketed_id_context_state = 0;

 
 void
 gram_scanner_last_string_free (void)
 
 void
 gram_scanner_last_string_free (void)


@@ -78,43 +101,51 @@ static void unexpected_newline (boundary, char const *);
  /* Strings and characters in directives/rules. */
 %x SC_ESCAPED_STRING SC_ESCAPED_CHARACTER
  /* A identifier was just read in directives/rules.  Special state
  /* Strings and characters in directives/rules. */
 %x SC_ESCAPED_STRING SC_ESCAPED_CHARACTER
  /* A identifier was just read in directives/rules.  Special state

-    to capture the sequence `identifier :'. */
+    to capture the sequence 'identifier :'. */

 %x SC_AFTER_IDENTIFIER
 %x SC_AFTER_IDENTIFIER

+ /* A complex tag, with nested angles brackets. */
+%x SC_TAG

 
 

- /* Three types of user code:
-    - prologue (code between `%{' `%}' in the first section, before %%);
+ /* Four types of user code:
+    - prologue (code between '%{' '%}' in the first section, before %%);

     - actions, printers, union, etc, (between braced in the middle section);
     - actions, printers, union, etc, (between braced in the middle section);

-    - epilogue (everything after the second %%). */
-%x SC_PROLOGUE SC_BRACED_CODE SC_EPILOGUE
+    - epilogue (everything after the second %%).
+    - predicate (code between '%?{' and '{' in middle section); */
+%x SC_PROLOGUE SC_BRACED_CODE SC_EPILOGUE SC_PREDICATE

  /* C and C++ comments in code. */
 %x SC_COMMENT SC_LINE_COMMENT
  /* Strings and characters in code. */
 %x SC_STRING SC_CHARACTER
  /* C and C++ comments in code. */
 %x SC_COMMENT SC_LINE_COMMENT
  /* Strings and characters in code. */
 %x SC_STRING SC_CHARACTER

+ /* Bracketed identifiers support. */
+%x SC_BRACKETED_ID SC_RETURN_BRACKETED_ID

 
 

-letter   [.abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_]
-id       {letter}({letter}|[0-9])*
-directive %{letter}({letter}|[0-9]|-)*
-int      [0-9]+
+letter    [.abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_]
+id        {letter}({letter}|[-0-9])*
+directive %{id}
+int       [0-9]+

 
 /* POSIX says that a tag must be both an id and a C union member, but
    historically almost any character is allowed in a tag.  We disallow
 
 /* POSIX says that a tag must be both an id and a C union member, but
    historically almost any character is allowed in a tag.  We disallow

-   NUL and newline, as this simplifies our implementation.  */
-tag     [^\0\n>]+
+   NUL, as this simplifies our implementation.  We disallow angle
+   bracket to match them in nested pairs: several languages use them
+   for generics/template types.  */
+tag      [^\0<>]+

 
 /* Zero or more instances of backslash-newline.  Following GCC, allow
    white space between the backslash and the newline.  */
 
 /* Zero or more instances of backslash-newline.  Following GCC, allow
    white space between the backslash and the newline.  */

-splice  (\\[ \f\t\v]*\n)*
+splice   (\\[ \f\t\v]*\n)*

 
 %%
 %{
 
 %%
 %{

-  /* Nesting level of the current code in braces.  */
-  int braces_level IF_LINT (= 0);
+  /* Nesting level.  Either for nested braces, or nested angle brackets
+     (but not mixed).  */
+  int nesting PACIFY_CC (= 0);

 
   /* Parent context state, when applicable.  */
 
   /* Parent context state, when applicable.  */

-  int context_state IF_LINT (= 0);
+  int context_state PACIFY_CC (= 0);

 
   /* Location of most recent identifier, when applicable.  */
 
   /* Location of most recent identifier, when applicable.  */

-  location id_loc IF_LINT (= empty_location);
+  location id_loc PACIFY_CC (= empty_location);

 
   /* Where containing code started, when applicable.  Its initial
      value is relevant only when yylex is invoked in the SC_EPILOGUE
 
   /* Where containing code started, when applicable.  Its initial
      value is relevant only when yylex is invoked in the SC_EPILOGUE


@@ -123,7 +154,7 @@ splice       (\\[ \f\t\v]*\n)*
 
   /* Where containing comment or string or character literal started,
      when applicable.  */
 
   /* Where containing comment or string or character literal started,
      when applicable.  */

-  boundary token_start IF_LINT (= scanner_cursor);
+  boundary token_start PACIFY_CC (= scanner_cursor);

 %}
 
 
 %}
 
 


@@ -131,10 +162,10 @@ splice     (\\[ \f\t\v]*\n)*
   | Scanning white space.  |
   `-----------------------*/
 
   | Scanning white space.  |
   `-----------------------*/
 

-<INITIAL,SC_AFTER_IDENTIFIER>
+<INITIAL,SC_AFTER_IDENTIFIER,SC_BRACKETED_ID,SC_RETURN_BRACKETED_ID>

 {
   /* Comments and white space.  */
 {
   /* Comments and white space.  */

-  ","         warn_at (*loc, _("stray `,' treated as white space"));
+  ","          warn_at (*loc, _("stray ',' treated as white space"));

   [ \f\n\t\v]  |
   "//".*       ;
   "/*" {
   [ \f\n\t\v]  |
   "//".*       ;
   "/*" {


@@ -154,53 +185,59 @@ splice     (\\[ \f\t\v]*\n)*
   /*----------------------------.
   | Scanning Bison directives.  |
   `----------------------------*/
   /*----------------------------.
   | Scanning Bison directives.  |
   `----------------------------*/

+
+  /* For directives that are also command line options, the regex must be
+        "%..."
+     after "[-_]"s are removed, and the directive must match the --long
+     option name, with a single string argument.  Otherwise, add exceptions
+     to ../build-aux/cross-options.pl.  */
+

 <INITIAL>
 {
 <INITIAL>
 {

-  "%after-header"                   return PERCENT_AFTER_HEADER;
-  "%before-header"                  return PERCENT_BEFORE_HEADER;
-  "%binary"                        return PERCENT_NONASSOC;
-  "%debug"                         return PERCENT_DEBUG;
-  "%default"[-_]"prec"             return PERCENT_DEFAULT_PREC;
-  "%define"                        return PERCENT_DEFINE;
-  "%defines"                       return PERCENT_DEFINES;
-  "%destructor"                            return PERCENT_DESTRUCTOR;
-  "%dprec"                         return PERCENT_DPREC;
-  "%end-header"                     return PERCENT_END_HEADER;
-  "%error"[-_]"verbose"            return PERCENT_ERROR_VERBOSE;
-  "%expect"                        return PERCENT_EXPECT;
-  "%expect"[-_]"rr"                return PERCENT_EXPECT_RR;
-  "%file-prefix"                   return PERCENT_FILE_PREFIX;
+  "%binary"                         return PERCENT_NONASSOC;
+  "%code"                           return PERCENT_CODE;
+  "%debug"                          RETURN_PERCENT_FLAG("parse.trace");
+  "%default"[-_]"prec"              return PERCENT_DEFAULT_PREC;
+  "%define"                         return PERCENT_DEFINE;
+  "%defines"                        return PERCENT_DEFINES;
+  "%destructor"                     return PERCENT_DESTRUCTOR;
+  "%dprec"                          return PERCENT_DPREC;
+  "%error"[-_]"verbose"             return PERCENT_ERROR_VERBOSE;
+  "%expect"                         return PERCENT_EXPECT;
+  "%expect"[-_]"rr"                 return PERCENT_EXPECT_RR;
+  "%file-prefix"                    return PERCENT_FILE_PREFIX;

   "%fixed"[-_]"output"[-_]"files"   return PERCENT_YACC;
   "%fixed"[-_]"output"[-_]"files"   return PERCENT_YACC;

-  "%initial-action"                return PERCENT_INITIAL_ACTION;
-  "%glr-parser"                    return PERCENT_GLR_PARSER;
-  "%left"                          return PERCENT_LEFT;
-  "%lex-param"                     return PERCENT_LEX_PARAM;
-  "%locations"                     return PERCENT_LOCATIONS;
-  "%merge"                         return PERCENT_MERGE;
-  "%name"[-_]"prefix"              return PERCENT_NAME_PREFIX;
-  "%no"[-_]"default"[-_]"prec"     return PERCENT_NO_DEFAULT_PREC;
-  "%no"[-_]"lines"                 return PERCENT_NO_LINES;
-  "%nonassoc"                      return PERCENT_NONASSOC;
-  "%nondeterministic-parser"               return PERCENT_NONDETERMINISTIC_PARSER;
-  "%nterm"                         return PERCENT_NTERM;
-  "%output"                        return PERCENT_OUTPUT;
-  "%parse-param"                   return PERCENT_PARSE_PARAM;
-  "%prec"                          return PERCENT_PREC;
-  "%printer"                       return PERCENT_PRINTER;
-  "%pure"[-_]"parser"              return PERCENT_PURE_PARSER;
-  "%require"                       return PERCENT_REQUIRE;
-  "%right"                         return PERCENT_RIGHT;
-  "%skeleton"                      return PERCENT_SKELETON;
-  "%start"                         return PERCENT_START;
-  "%start-header"                   return PERCENT_START_HEADER;
-  "%symbol-default"                 return PERCENT_SYMBOL_DEFAULT;
-  "%term"                          return PERCENT_TOKEN;
-  "%token"                         return PERCENT_TOKEN;
-  "%token"[-_]"table"              return PERCENT_TOKEN_TABLE;
-  "%type"                          return PERCENT_TYPE;
-  "%union"                         return PERCENT_UNION;
-  "%verbose"                       return PERCENT_VERBOSE;
-  "%yacc"                          return PERCENT_YACC;
+  "%initial-action"                 return PERCENT_INITIAL_ACTION;
+  "%glr-parser"                     return PERCENT_GLR_PARSER;
+  "%language"                       return PERCENT_LANGUAGE;
+  "%left"                           return PERCENT_LEFT;
+  "%lex-param"                      RETURN_PERCENT_PARAM(lex);
+  "%locations"                      RETURN_PERCENT_FLAG("locations");
+  "%merge"                          return PERCENT_MERGE;
+  "%name"[-_]"prefix"               return PERCENT_NAME_PREFIX;
+  "%no"[-_]"default"[-_]"prec"      return PERCENT_NO_DEFAULT_PREC;
+  "%no"[-_]"lines"                  return PERCENT_NO_LINES;
+  "%nonassoc"                       return PERCENT_NONASSOC;
+  "%nondeterministic-parser"        return PERCENT_NONDETERMINISTIC_PARSER;
+  "%nterm"                          return PERCENT_NTERM;
+  "%output"                         return PERCENT_OUTPUT;
+  "%param"                          RETURN_PERCENT_PARAM(both);
+  "%parse-param"                    RETURN_PERCENT_PARAM(parse);
+  "%prec"                           return PERCENT_PREC;
+  "%precedence"                     return PERCENT_PRECEDENCE;
+  "%printer"                        return PERCENT_PRINTER;
+  "%pure"[-_]"parser"               RETURN_PERCENT_FLAG("api.pure");
+  "%require"                        return PERCENT_REQUIRE;
+  "%right"                          return PERCENT_RIGHT;
+  "%skeleton"                       return PERCENT_SKELETON;
+  "%start"                          return PERCENT_START;
+  "%term"                           return PERCENT_TOKEN;
+  "%token"                          return PERCENT_TOKEN;
+  "%token"[-_]"table"               return PERCENT_TOKEN_TABLE;
+  "%type"                           return PERCENT_TYPE;
+  "%union"                          return PERCENT_UNION;
+  "%verbose"                        return PERCENT_VERBOSE;
+  "%yacc"                           return PERCENT_YACC;

 
   {directive} {
     complain_at (*loc, _("invalid directive: %s"), quote (yytext));
 
   {directive} {
     complain_at (*loc, _("invalid directive: %s"), quote (yytext));


@@ -213,6 +250,7 @@ splice       (\\[ \f\t\v]*\n)*
   {id} {
     val->uniqstr = uniqstr_new (yytext);
     id_loc = *loc;
   {id} {
     val->uniqstr = uniqstr_new (yytext);
     id_loc = *loc;

+    bracketed_id_str = NULL;

     BEGIN SC_AFTER_IDENTIFIER;
   }
 
     BEGIN SC_AFTER_IDENTIFIER;
   }
 


@@ -225,11 +263,17 @@ splice     (\\[ \f\t\v]*\n)*
     return INT;
   }
 
     return INT;
   }
 

-  /* Characters.  We don't check there is only one.  */
-  "'"        STRING_GROW; token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER;
+  /* Identifiers may not start with a digit.  Yet, don't silently
+     accept "1FOO" as "1 FOO".  */
+  {int}{id} {
+    complain_at (*loc, _("invalid identifier: %s"), quote (yytext));
+  }
+
+  /* Characters.  */
+  "'"         token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER;

 
   /* Strings. */
 
   /* Strings. */

-  "\""       token_start = loc->start; BEGIN SC_ESCAPED_STRING;
+  "\""        token_start = loc->start; BEGIN SC_ESCAPED_STRING;

 
   /* Prologue. */
   "%{"        code_start = loc->start; BEGIN SC_PROLOGUE;
 
   /* Prologue. */
   "%{"        code_start = loc->start; BEGIN SC_PROLOGUE;


@@ -237,18 +281,32 @@ splice     (\\[ \f\t\v]*\n)*
   /* Code in between braces.  */
   "{" {
     STRING_GROW;
   /* Code in between braces.  */
   "{" {
     STRING_GROW;

-    braces_level = 0;
+    nesting = 0;

     code_start = loc->start;
     BEGIN SC_BRACED_CODE;
   }
 
     code_start = loc->start;
     BEGIN SC_BRACED_CODE;
   }
 

+  /* Semantic predicate. */
+  "%?"[ \f\n\t\v]*"{" {
+    nesting = 0;
+    code_start = loc->start;
+    BEGIN SC_PREDICATE;
+  }
+

   /* A type. */
   /* A type. */

+  "<*>"       return TAG_ANY;
+  "<>"        return TAG_NONE;

   "<"{tag}">" {
     obstack_grow (&obstack_for_string, yytext + 1, yyleng - 2);
     STRING_FINISH;
     val->uniqstr = uniqstr_new (last_string);
     STRING_FREE;
   "<"{tag}">" {
     obstack_grow (&obstack_for_string, yytext + 1, yyleng - 2);
     STRING_FINISH;
     val->uniqstr = uniqstr_new (last_string);
     STRING_FREE;

-    return TYPE;
+    return TAG;
+  }
+  "<"         {
+    nesting = 0;
+    token_start = loc->start;
+    BEGIN SC_TAG;

   }
 
   "%%" {
   }
 
   "%%" {


@@ -258,6 +316,13 @@ splice      (\\[ \f\t\v]*\n)*
     return PERCENT_PERCENT;
   }
 
     return PERCENT_PERCENT;
   }
 

+  "[" {
+    bracketed_id_str = NULL;
+    bracketed_id_start = loc->start;
+    bracketed_id_context_state = YY_START;
+    BEGIN SC_BRACKETED_ID;
+  }
+

   . {
     complain_at (*loc, _("invalid character: %s"), quote (yytext));
   }
   . {
     complain_at (*loc, _("invalid character: %s"), quote (yytext));
   }


@@ -269,64 +334,144 @@ splice    (\\[ \f\t\v]*\n)*
 }
 
 
 }
 
 

+  /*--------------------------------------------------------------.
+  | Supporting \0 complexifies our implementation for no expected |
+  | added value.                                                  |
+  `--------------------------------------------------------------*/
+
+<SC_ESCAPED_CHARACTER,SC_ESCAPED_STRING,SC_TAG>
+{
+  \0        complain_at (*loc, _("invalid null character"));
+}
+
+

   /*-----------------------------------------------------------------.
   | Scanning after an identifier, checking whether a colon is next.  |
   `-----------------------------------------------------------------*/
 
 <SC_AFTER_IDENTIFIER>
 {
   /*-----------------------------------------------------------------.
   | Scanning after an identifier, checking whether a colon is next.  |
   `-----------------------------------------------------------------*/
 
 <SC_AFTER_IDENTIFIER>
 {

+  "[" {
+    if (bracketed_id_str)
+      {
+        ROLLBACK_CURRENT_TOKEN;
+        BEGIN SC_RETURN_BRACKETED_ID;
+        *loc = id_loc;
+        return ID;
+      }
+    else
+      {
+        bracketed_id_start = loc->start;
+        bracketed_id_context_state = YY_START;
+        BEGIN SC_BRACKETED_ID;
+      }
+  }

   ":" {
   ":" {

+    BEGIN (bracketed_id_str ? SC_RETURN_BRACKETED_ID : INITIAL);

     *loc = id_loc;
     *loc = id_loc;

-    BEGIN INITIAL;

     return ID_COLON;
   }
   . {
     return ID_COLON;
   }
   . {

-    scanner_cursor.column -= mbsnwidth (yytext, yyleng, 0);
-    yyless (0);
+    ROLLBACK_CURRENT_TOKEN;
+    BEGIN (bracketed_id_str ? SC_RETURN_BRACKETED_ID : INITIAL);

     *loc = id_loc;
     *loc = id_loc;

-    BEGIN INITIAL;

     return ID;
   }
   <<EOF>> {
     return ID;
   }
   <<EOF>> {

+    BEGIN (bracketed_id_str ? SC_RETURN_BRACKETED_ID : INITIAL);

     *loc = id_loc;
     *loc = id_loc;

-    BEGIN INITIAL;

     return ID;
   }
 }
 
     return ID;
   }
 }
 

+  /*--------------------------------.
+  | Scanning bracketed identifiers. |
+  `--------------------------------*/
+
+<SC_BRACKETED_ID>
+{
+  {id} {
+    if (bracketed_id_str)
+      {
+        complain_at (*loc, _("unexpected identifier in bracketed name: %s"),
+                     quote (yytext));
+      }
+    else
+      {
+        bracketed_id_str = uniqstr_new (yytext);
+        bracketed_id_loc = *loc;
+      }
+  }
+  "]" {
+    BEGIN bracketed_id_context_state;
+    if (bracketed_id_str)
+      {
+        if (INITIAL == bracketed_id_context_state)
+          {
+            val->uniqstr = bracketed_id_str;
+            bracketed_id_str = 0;
+            *loc = bracketed_id_loc;
+            return BRACKETED_ID;
+          }
+      }
+    else
+      complain_at (*loc, _("an identifier expected"));
+  }
+  . {
+    complain_at (*loc, _("invalid character in bracketed name: %s"),
+                 quote (yytext));
+  }
+  <<EOF>> {
+    BEGIN bracketed_id_context_state;
+    unexpected_eof (bracketed_id_start, "]");
+  }
+}
+
+<SC_RETURN_BRACKETED_ID>
+{
+  . {
+    ROLLBACK_CURRENT_TOKEN;
+    val->uniqstr = bracketed_id_str;
+    bracketed_id_str = 0;
+    *loc = bracketed_id_loc;
+    BEGIN INITIAL;
+    return BRACKETED_ID;
+  }
+}
+

 
   /*---------------------------------------------------------------.
 
   /*---------------------------------------------------------------.

-  | Scanning a Yacc comment.  The initial `/ *' is already eaten.  |
+  | Scanning a Yacc comment.  The initial '/ *' is already eaten.  |

   `---------------------------------------------------------------*/
 
 <SC_YACC_COMMENT>
 {
   "*/"     BEGIN context_state;
   `---------------------------------------------------------------*/
 
 <SC_YACC_COMMENT>
 {
   "*/"     BEGIN context_state;

-  .|\n    ;
+  .|\n     ;

   <<EOF>>  unexpected_eof (token_start, "*/"); BEGIN context_state;
 }
 
 
   /*------------------------------------------------------------.
   <<EOF>>  unexpected_eof (token_start, "*/"); BEGIN context_state;
 }
 
 
   /*------------------------------------------------------------.

-  | Scanning a C comment.  The initial `/ *' is already eaten.  |
+  | Scanning a C comment.  The initial '/ *' is already eaten.  |

   `------------------------------------------------------------*/
 
 <SC_COMMENT>
 {
   "*"{splice}"/"  STRING_GROW; BEGIN context_state;
   `------------------------------------------------------------*/
 
 <SC_COMMENT>
 {
   "*"{splice}"/"  STRING_GROW; BEGIN context_state;

-  <<EOF>>        unexpected_eof (token_start, "*/"); BEGIN context_state;
+  <<EOF>>         unexpected_eof (token_start, "*/"); BEGIN context_state;

 }
 
 
   /*--------------------------------------------------------------.
 }
 
 
   /*--------------------------------------------------------------.

-  | Scanning a line comment.  The initial `//' is already eaten.  |
+  | Scanning a line comment.  The initial '//' is already eaten.  |

   `--------------------------------------------------------------*/
 
 <SC_LINE_COMMENT>
 {
   `--------------------------------------------------------------*/
 
 <SC_LINE_COMMENT>
 {

-  "\n"          STRING_GROW; BEGIN context_state;
-  {splice}      STRING_GROW;
-  <<EOF>>       BEGIN context_state;
+  "\n"           STRING_GROW; BEGIN context_state;
+  {splice}       STRING_GROW;
+  <<EOF>>        BEGIN context_state;

 }
 
 
 }
 
 


@@ -358,41 +503,88 @@ splice     (\\[ \f\t\v]*\n)*
 
   /*----------------------------------------------------------.
   | Scanning a Bison character literal, decoding its escapes. |
 
   /*----------------------------------------------------------.
   | Scanning a Bison character literal, decoding its escapes. |

-  | The initial quote is already eaten.                              |
+  | The initial quote is already eaten.                       |

   `----------------------------------------------------------*/
 
 <SC_ESCAPED_CHARACTER>
 {
   "'"|"\n" {
   `----------------------------------------------------------*/
 
 <SC_ESCAPED_CHARACTER>
 {
   "'"|"\n" {

-    if (yytext[0] == '\n')
-      unexpected_newline (token_start, "'");
-    STRING_GROW;

     STRING_FINISH;
     loc->start = token_start;
     STRING_FINISH;
     loc->start = token_start;

-    val->character = last_string[1];
+    val->character = last_string[0];
+    {
+      /* FIXME: Eventually, make these errors.  */
+      if (last_string[0] == '\0')
+        {
+          warn_at (*loc, _("empty character literal"));
+          /* '\0' seems dangerous even if we are about to complain.  */
+          val->character = '\'';
+        }
+      else if (last_string[1] != '\0')
+        warn_at (*loc, _("extra characters in character literal"));
+    }
+    if (yytext[0] == '\n')
+      unexpected_newline (token_start, "'");

     STRING_FREE;
     BEGIN INITIAL;
     return CHAR;
   }
   <<EOF>> {
     STRING_FREE;
     BEGIN INITIAL;
     return CHAR;
   }
   <<EOF>> {

-    unexpected_eof (token_start, "'");

     STRING_FINISH;
     loc->start = token_start;
     STRING_FINISH;
     loc->start = token_start;

-    if (strlen(last_string) > 1)
-      val->character = last_string[1];
-    else
-      val->character = last_string[0];
+    val->character = last_string[0];
+    {
+      /* FIXME: Eventually, make these errors.  */
+      if (last_string[0] == '\0')
+        {
+          warn_at (*loc, _("empty character literal"));
+          /* '\0' seems dangerous even if we are about to complain.  */
+          val->character = '\'';
+        }
+      else if (last_string[1] != '\0')
+        warn_at (*loc, _("extra characters in character literal"));
+    }
+    unexpected_eof (token_start, "'");

     STRING_FREE;
     BEGIN INITIAL;
     return CHAR;
   }
 }
 
     STRING_FREE;
     BEGIN INITIAL;
     return CHAR;
   }
 }
 

-<SC_ESCAPED_CHARACTER,SC_ESCAPED_STRING>
+  /*-----------------------------------------------------------.
+  | Scanning a Bison nested tag.  The initial angle bracket is |
+  | already eaten.                                             |
+  `-----------------------------------------------------------*/
+
+<SC_TAG>

 {
 {

-  \0       complain_at (*loc, _("invalid null character"));
-}
+  ">" {
+    --nesting;
+    if (nesting < 0)
+      {
+        STRING_FINISH;
+        loc->start = token_start;
+        val->uniqstr = uniqstr_new (last_string);
+        STRING_FREE;
+        BEGIN INITIAL;
+        return TAG;
+      }
+    STRING_GROW;
+  }
+
+  [^<>]+ STRING_GROW;
+  "<"+   STRING_GROW; nesting += yyleng;

 
 

+  <<EOF>> {
+    unexpected_eof (token_start, ">");
+    STRING_FINISH;
+    loc->start = token_start;
+    val->uniqstr = uniqstr_new (last_string);
+    STRING_FREE;
+    BEGIN INITIAL;
+    return TAG;
+  }
+}

 
   /*----------------------------.
   | Decode escaped characters.  |
 
   /*----------------------------.
   | Decode escaped characters.  |


@@ -402,10 +594,9 @@ splice      (\\[ \f\t\v]*\n)*
 {
   \\[0-7]{1,3} {
     unsigned long int c = strtoul (yytext + 1, NULL, 8);
 {
   \\[0-7]{1,3} {
     unsigned long int c = strtoul (yytext + 1, NULL, 8);

-    if (UCHAR_MAX < c)
-      complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext));
-    else if (! c)
-      complain_at (*loc, _("invalid null character: %s"), quote (yytext));
+    if (!c || UCHAR_MAX < c)
+      complain_at (*loc, _("invalid number after \\-escape: %s"),
+                   yytext+1);

     else
       obstack_1grow (&obstack_for_string, c);
   }
     else
       obstack_1grow (&obstack_for_string, c);
   }


@@ -413,37 +604,40 @@ splice     (\\[ \f\t\v]*\n)*
   \\x[0-9abcdefABCDEF]+ {
     verify (UCHAR_MAX < ULONG_MAX);
     unsigned long int c = strtoul (yytext + 2, NULL, 16);
   \\x[0-9abcdefABCDEF]+ {
     verify (UCHAR_MAX < ULONG_MAX);
     unsigned long int c = strtoul (yytext + 2, NULL, 16);

-    if (UCHAR_MAX < c)
-      complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext));
-    else if (! c)
-      complain_at (*loc, _("invalid null character: %s"), quote (yytext));
+    if (!c || UCHAR_MAX < c)
+      complain_at (*loc, _("invalid number after \\-escape: %s"),
+                   yytext+1);

     else
       obstack_1grow (&obstack_for_string, c);
   }
 
     else
       obstack_1grow (&obstack_for_string, c);
   }
 

-  \\a  obstack_1grow (&obstack_for_string, '\a');
-  \\b  obstack_1grow (&obstack_for_string, '\b');
-  \\f  obstack_1grow (&obstack_for_string, '\f');
-  \\n  obstack_1grow (&obstack_for_string, '\n');
-  \\r  obstack_1grow (&obstack_for_string, '\r');
-  \\t  obstack_1grow (&obstack_for_string, '\t');
-  \\v  obstack_1grow (&obstack_for_string, '\v');
+  \\a   obstack_1grow (&obstack_for_string, '\a');
+  \\b   obstack_1grow (&obstack_for_string, '\b');
+  \\f   obstack_1grow (&obstack_for_string, '\f');
+  \\n   obstack_1grow (&obstack_for_string, '\n');
+  \\r   obstack_1grow (&obstack_for_string, '\r');
+  \\t   obstack_1grow (&obstack_for_string, '\t');
+  \\v   obstack_1grow (&obstack_for_string, '\v');

 
   /* \\[\"\'?\\] would be shorter, but it confuses xgettext.  */
   \\("\""|"'"|"?"|"\\")  obstack_1grow (&obstack_for_string, yytext[1]);
 
   \\(u|U[0-9abcdefABCDEF]{4})[0-9abcdefABCDEF]{4} {
     int c = convert_ucn_to_byte (yytext);
 
   /* \\[\"\'?\\] would be shorter, but it confuses xgettext.  */
   \\("\""|"'"|"?"|"\\")  obstack_1grow (&obstack_for_string, yytext[1]);
 
   \\(u|U[0-9abcdefABCDEF]{4})[0-9abcdefABCDEF]{4} {
     int c = convert_ucn_to_byte (yytext);

-    if (c < 0)
-      complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext));
-    else if (! c)
-      complain_at (*loc, _("invalid null character: %s"), quote (yytext));
+    if (c <= 0)
+      complain_at (*loc, _("invalid number after \\-escape: %s"),
+                   yytext+1);

     else
       obstack_1grow (&obstack_for_string, c);
   }
     else
       obstack_1grow (&obstack_for_string, c);
   }

-  \\(.|\n)     {
-    complain_at (*loc, _("unrecognized escape sequence: %s"), quote (yytext));
-    STRING_GROW;
+  \\(.|\n)      {
+    char const *p = yytext + 1;
+    /* Quote only if escaping won't make the character visible.  */
+    if (isspace ((unsigned char) *p) && isprint ((unsigned char) *p))
+      p = quote (p);
+    else
+      p = quotearg_style_mem (escape_quoting_style, p, 1);
+    complain_at (*loc, _("invalid character after \\-escape: %s"), p);

   }
 }
 
   }
 }
 


@@ -453,21 +647,21 @@ splice     (\\[ \f\t\v]*\n)*
 
 <SC_CHARACTER,SC_STRING>
 {
 
 <SC_CHARACTER,SC_STRING>
 {

-  {splice}|\\{splice}[^\n\[\]] STRING_GROW;
+  {splice}|\\{splice}[^\n\[\]]  STRING_GROW;

 }
 
 <SC_CHARACTER>
 {
 }
 
 <SC_CHARACTER>
 {

-  "'"          STRING_GROW; BEGIN context_state;
-  \n           unexpected_newline (token_start, "'"); BEGIN context_state;
-  <<EOF>>      unexpected_eof (token_start, "'"); BEGIN context_state;
+  "'"           STRING_GROW; BEGIN context_state;
+  \n            unexpected_newline (token_start, "'"); BEGIN context_state;
+  <<EOF>>       unexpected_eof (token_start, "'"); BEGIN context_state;

 }
 
 <SC_STRING>
 {
 }
 
 <SC_STRING>
 {

-  "\""         STRING_GROW; BEGIN context_state;
-  \n           unexpected_newline (token_start, "\""); BEGIN context_state;
-  <<EOF>>      unexpected_eof (token_start, "\""); BEGIN context_state;
+  "\""          STRING_GROW; BEGIN context_state;
+  \n            unexpected_newline (token_start, "\""); BEGIN context_state;
+  <<EOF>>       unexpected_eof (token_start, "\""); BEGIN context_state;

 }
 
 
 }
 
 


@@ -475,7 +669,7 @@ splice       (\\[ \f\t\v]*\n)*
   | Strings, comments etc. can be found in user code.  |
   `---------------------------------------------------*/
 
   | Strings, comments etc. can be found in user code.  |
   `---------------------------------------------------*/
 

-<SC_BRACED_CODE,SC_PROLOGUE,SC_EPILOGUE>
+<SC_BRACED_CODE,SC_PROLOGUE,SC_EPILOGUE,SC_PREDICATE>

 {
   "'" {
     STRING_GROW;
 {
   "'" {
     STRING_GROW;


@@ -505,42 +699,63 @@ splice     (\\[ \f\t\v]*\n)*
 
 
   /*-----------------------------------------------------------.
 
 
   /*-----------------------------------------------------------.

-  | Scanning some code in braces (actions). The initial "{" is |
-  | already eaten.                                             |
+  | Scanning some code in braces (actions, predicates). The    |
+  | initial "{" is already eaten.                              |

   `-----------------------------------------------------------*/
 
   `-----------------------------------------------------------*/
 

-<SC_BRACED_CODE>
+<SC_BRACED_CODE,SC_PREDICATE>

 {
 {

-  "{"|"<"{splice}"%"  STRING_GROW; braces_level++;
-  "%"{splice}">"      STRING_GROW; braces_level--;
-  "}" {
-    obstack_1grow (&obstack_for_string, '}');
-
-    --braces_level;
-    if (braces_level < 0)
-      {
-       STRING_FINISH;
-       loc->start = code_start;
-       val->chars = last_string;
-       BEGIN INITIAL;
-       return BRACED_CODE;
-      }
-  }
+  "{"|"<"{splice}"%"  STRING_GROW; nesting++;
+  "%"{splice}">"      STRING_GROW; nesting--;

 
 

-  /* Tokenize `<<%' correctly (as `<<' `%') rather than incorrrectly
-     (as `<' `<%').  */
+  /* Tokenize '<<%' correctly (as '<<' '%') rather than incorrrectly
+     (as '<' '<%').  */

   "<"{splice}"<"  STRING_GROW;
 
   <<EOF>> {
   "<"{splice}"<"  STRING_GROW;
 
   <<EOF>> {

+    int token = (YY_START == SC_BRACED_CODE) ? BRACED_CODE : BRACED_PREDICATE;

     unexpected_eof (code_start, "}");
     STRING_FINISH;
     loc->start = code_start;
     unexpected_eof (code_start, "}");
     STRING_FINISH;
     loc->start = code_start;

-    val->chars = last_string;
+    val->code = last_string;

     BEGIN INITIAL;
     BEGIN INITIAL;

-    return BRACED_CODE;
+    return token;
+  }
+}
+
+<SC_BRACED_CODE>
+{
+  "}" {
+    obstack_1grow (&obstack_for_string, '}');
+
+    --nesting;
+    if (nesting < 0)
+      {
+        STRING_FINISH;
+        loc->start = code_start;
+        val->code = last_string;
+        BEGIN INITIAL;
+        return BRACED_CODE;
+      }

   }
 }
 
   }
 }
 

+<SC_PREDICATE>
+{
+  "}" {
+    --nesting;
+    if (nesting < 0)
+      {
+        STRING_FINISH;
+        loc->start = code_start;
+        val->code = last_string;
+        BEGIN INITIAL;
+        return BRACED_PREDICATE;
+      }
+    else
+      obstack_1grow (&obstack_for_string, '}');
+  }
+}

 
   /*--------------------------------------------------------------.
   | Scanning some prologue: from "%{" (already scanned) to "%}".  |
 
   /*--------------------------------------------------------------.
   | Scanning some prologue: from "%{" (already scanned) to "%}".  |


@@ -588,8 +803,8 @@ splice       (\\[ \f\t\v]*\n)*
   | By default, grow the string obstack with the input.  |
   `-----------------------------------------------------*/
 
   | By default, grow the string obstack with the input.  |
   `-----------------------------------------------------*/
 

-<SC_COMMENT,SC_LINE_COMMENT,SC_BRACED_CODE,SC_PROLOGUE,SC_EPILOGUE,SC_STRING,SC_CHARACTER,SC_ESCAPED_STRING,SC_ESCAPED_CHARACTER>.     |
-<SC_COMMENT,SC_LINE_COMMENT,SC_BRACED_CODE,SC_PROLOGUE,SC_EPILOGUE>\n  STRING_GROW;
+<SC_COMMENT,SC_LINE_COMMENT,SC_BRACED_CODE,SC_PREDICATE,SC_PROLOGUE,SC_EPILOGUE,SC_STRING,SC_CHARACTER,SC_ESCAPED_STRING,SC_ESCAPED_CHARACTER>. |
+  <SC_COMMENT,SC_LINE_COMMENT,SC_BRACED_CODE,SC_PREDICATE,SC_PROLOGUE,SC_EPILOGUE>\n    STRING_GROW;

 
 %%
 
 
 %%
 


@@ -605,35 +820,35 @@ no_cr_read (FILE *fp, char *buf, size_t size)
     {
       char *w = memchr (buf, '\r', bytes_read);
       if (w)
     {
       char *w = memchr (buf, '\r', bytes_read);
       if (w)

-       {
-         char const *r = ++w;
-         char const *lim = buf + bytes_read;
-
-         for (;;)
-           {
-             /* Found an '\r'.  Treat it like '\n', but ignore any
-                '\n' that immediately follows.  */
-             w[-1] = '\n';
-             if (r == lim)
-               {
-                 int ch = getc (fp);
-                 if (ch != '\n' && ungetc (ch, fp) != ch)
-                   break;
-               }
-             else if (*r == '\n')
-               r++;
-
-             /* Copy until the next '\r'.  */
-             do
-               {
-                 if (r == lim)
-                   return w - buf;
-               }
-             while ((*w++ = *r++) != '\r');
-           }
-
-         return w - buf;
-       }
+        {
+          char const *r = ++w;
+          char const *lim = buf + bytes_read;
+
+          for (;;)
+            {
+              /* Found an '\r'.  Treat it like '\n', but ignore any
+                 '\n' that immediately follows.  */
+              w[-1] = '\n';
+              if (r == lim)
+                {
+                  int ch = getc (fp);
+                  if (ch != '\n' && ungetc (ch, fp) != ch)
+                    break;
+                }
+              else if (*r == '\n')
+                r++;
+
+              /* Copy until the next '\r'.  */
+              do
+                {
+                  if (r == lim)
+                    return w - buf;
+                }
+              while ((*w++ = *r++) != '\r');
+            }
+
+          return w - buf;
+        }

     }
 
   return bytes_read;
     }
 
   return bytes_read;


@@ -664,7 +879,7 @@ scan_integer (char const *number, int base, location loc)
 /*------------------------------------------------------------------.
 | Convert universal character name UCN to a single-byte character,  |
 | and return that character.  Return -1 if UCN does not correspond  |
 /*------------------------------------------------------------------.
 | Convert universal character name UCN to a single-byte character,  |
 | and return that character.  Return -1 if UCN does not correspond  |

-| to a single-byte character.                                      |
+| to a single-byte character.                                       |

 `------------------------------------------------------------------*/
 
 static int
 `------------------------------------------------------------------*/
 
 static int


@@ -692,22 +907,22 @@ convert_ucn_to_byte (char const *ucn)
        about.  */
     static signed char const table[] =
       {
        about.  */
     static signed char const table[] =
       {

-       '\0',   -1,   -1,   -1,   -1,   -1,   -1, '\a',
-       '\b', '\t', '\n', '\v', '\f', '\r',   -1,   -1,
-         -1,   -1,   -1,   -1,   -1,   -1,   -1,   -1,
-         -1,   -1,   -1,   -1,   -1,   -1,   -1,   -1,
-        ' ',  '!',  '"',  '#',  '$',  '%',  '&', '\'',
-        '(',  ')',  '*',  '+',  ',',  '-',  '.',  '/',
-        '0',  '1',  '2',  '3',  '4',  '5',  '6',  '7',
-        '8',  '9',  ':',  ';',  '<',  '=',  '>',  '?',
-        '@',  'A',  'B',  'C',  'D',  'E',  'F',  'G',
-        'H',  'I',  'J',  'K',  'L',  'M',  'N',  'O',
-        'P',  'Q',  'R',  'S',  'T',  'U',  'V',  'W',
-        'X',  'Y',  'Z',  '[', '\\',  ']',  '^',  '_',
-        '`',  'a',  'b',  'c',  'd',  'e',  'f',  'g',
-        'h',  'i',  'j',  'k',  'l',  'm',  'n',  'o',
-        'p',  'q',  'r',  's',  't',  'u',  'v',  'w',
-        'x',  'y',  'z',  '{',  '|',  '}',  '~'
+        '\0',   -1,   -1,   -1,   -1,   -1,   -1, '\a',
+        '\b', '\t', '\n', '\v', '\f', '\r',   -1,   -1,
+          -1,   -1,   -1,   -1,   -1,   -1,   -1,   -1,
+          -1,   -1,   -1,   -1,   -1,   -1,   -1,   -1,
+         ' ',  '!',  '"',  '#',  '$',  '%',  '&', '\'',
+         '(',  ')',  '*',  '+',  ',',  '-',  '.',  '/',
+         '0',  '1',  '2',  '3',  '4',  '5',  '6',  '7',
+         '8',  '9',  ':',  ';',  '<',  '=',  '>',  '?',
+         '@',  'A',  'B',  'C',  'D',  'E',  'F',  'G',
+         'H',  'I',  'J',  'K',  'L',  'M',  'N',  'O',
+         'P',  'Q',  'R',  'S',  'T',  'U',  'V',  'W',
+         'X',  'Y',  'Z',  '[', '\\',  ']',  '^',  '_',
+         '`',  'a',  'b',  'c',  'd',  'e',  'f',  'g',
+         'h',  'i',  'j',  'k',  'l',  'm',  'n',  'o',
+         'p',  'q',  'r',  's',  't',  'u',  'v',  'w',
+         'x',  'y',  'z',  '{',  '|',  '}',  '~'

       };
 
     code = code < sizeof table ? table[code] : -1;
       };
 
     code = code < sizeof table ? table[code] : -1;


@@ -719,7 +934,7 @@ convert_ucn_to_byte (char const *ucn)
 
 
 /*----------------------------------------------------------------.
 
 
 /*----------------------------------------------------------------.

-| Handle `#line INT "FILE"'.  ARGS has already skipped `#line '.  |
+| Handle '#line INT "FILE"'.  ARGS has already skipped '#line '.  |

 `----------------------------------------------------------------*/
 
 static void
 `----------------------------------------------------------------*/
 
 static void


@@ -727,8 +942,8 @@ handle_syncline (char *args, location loc)
 {
   char *after_num;
   unsigned long int lineno = strtoul (args, &after_num, 10);
 {
   char *after_num;
   unsigned long int lineno = strtoul (args, &after_num, 10);

-  char *file = strchr (after_num, '"') + 1;
-  *strchr (file, '"') = '\0';
+  char *file = mbschr (after_num, '"') + 1;
+  *mbschr (file, '"') = '\0';

   if (INT_MAX <= lineno)
     {
       warn_at (loc, _("line number overflow"));
   if (INT_MAX <= lineno)
     {
       warn_at (loc, _("line number overflow"));


@@ -741,8 +956,8 @@ handle_syncline (char *args, location loc)
 
 /*----------------------------------------------------------------.
 | For a token or comment starting at START, report message MSGID, |
 
 /*----------------------------------------------------------------.
 | For a token or comment starting at START, report message MSGID, |

-| which should say that an end marker was found before           |
-| the expected TOKEN_END.                                        |
+| which should say that an end marker was found before            |
+| the expected TOKEN_END.                                         |

 `----------------------------------------------------------------*/
 
 static void
 `----------------------------------------------------------------*/
 
 static void


@@ -751,6 +966,10 @@ unexpected_end (boundary start, char const *msgid, char const *token_end)
   location loc;
   loc.start = start;
   loc.end = scanner_cursor;
   location loc;
   loc.start = start;
   loc.end = scanner_cursor;

+  token_end = quote (token_end);
+  // Instead of '\'', display "'".
+  if (STREQ (token_end, "'\\''", '\'', '\\', '\'', '\'', 0,0,0,0,0))
+    token_end = "\"'\"";

   complain_at (loc, _(msgid), token_end);
 }
 
   complain_at (loc, _(msgid), token_end);
 }
 


@@ -763,7 +982,7 @@ unexpected_end (boundary start, char const *msgid, char const *token_end)
 static void
 unexpected_eof (boundary start, char const *token_end)
 {
 static void
 unexpected_eof (boundary start, char const *token_end)
 {

-  unexpected_end (start, N_("missing `%s' at end of file"), token_end);
+  unexpected_end (start, N_("missing %s at end of file"), token_end);

 }
 
 
 }
 
 


@@ -774,7 +993,7 @@ unexpected_eof (boundary start, char const *token_end)
 static void
 unexpected_newline (boundary start, char const *token_end)
 {
 static void
 unexpected_newline (boundary start, char const *token_end)
 {

-  unexpected_end (start, N_("missing `%s' at end of line"), token_end);
+  unexpected_end (start, N_("missing %s at end of line"), token_end);

 }
 
 
 }
 
 


@@ -798,5 +1017,5 @@ gram_scanner_free (void)
 {
   obstack_free (&obstack_for_string, 0);
   /* Reclaim Flex's buffers.  */
 {
   obstack_free (&obstack_for_string, 0);
   /* Reclaim Flex's buffers.  */

-  yy_delete_buffer (YY_CURRENT_BUFFER);
+  yylex_destroy ();

 }
 }