regression.at

GNU的词法/语法分析器bison源码

AT_DATA([input.y],
[[%token	undef_id_tok const_id_tok

%start CONST_DEC_PART

%%
CONST_DEC_PART:
         CONST_DEC_LIST
        ;

CONST_DEC_LIST:
	  CONST_DEC
        | CONST_DEC_LIST CONST_DEC
        ;

CONST_DEC:
	  { } undef_id_tok '=' const_id_tok ';'
        ;
%%
]])

AT_CHECK([bison -v input.y])
AT_CHECK([cat input.output], 0,
[[Grammar

    0 $accept: CONST_DEC_PART $end

    1 CONST_DEC_PART: CONST_DEC_LIST

    2 CONST_DEC_LIST: CONST_DEC
    3               | CONST_DEC_LIST CONST_DEC

    4 @1: /* empty */

    5 CONST_DEC: @1 undef_id_tok '=' const_id_tok ';'


Terminals, with rules where they appear

$end (0) 0
';' (59) 5
'=' (61) 5
error (256)
undef_id_tok (258) 5
const_id_tok (259) 5


Nonterminals, with rules where they appear

$accept (7)
    on left: 0
CONST_DEC_PART (8)
    on left: 1, on right: 0
CONST_DEC_LIST (9)
    on left: 2 3, on right: 1 3
CONST_DEC (10)
    on left: 5, on right: 2 3
@1 (11)
    on left: 4, on right: 5


state 0

    0 $accept: . CONST_DEC_PART $end

    $default  reduce using rule 4 (@1)

    CONST_DEC_PART  go to state 1
    CONST_DEC_LIST  go to state 2
    CONST_DEC       go to state 3
    @1              go to state 4


state 1

    0 $accept: CONST_DEC_PART . $end

    $end  shift, and go to state 5


state 2

    1 CONST_DEC_PART: CONST_DEC_LIST .
    3 CONST_DEC_LIST: CONST_DEC_LIST . CONST_DEC

    undef_id_tok  reduce using rule 4 (@1)
    $default      reduce using rule 1 (CONST_DEC_PART)

    CONST_DEC  go to state 6
    @1         go to state 4


state 3

    2 CONST_DEC_LIST: CONST_DEC .

    $default  reduce using rule 2 (CONST_DEC_LIST)


state 4

    5 CONST_DEC: @1 . undef_id_tok '=' const_id_tok ';'

    undef_id_tok  shift, and go to state 7


state 5

    0 $accept: CONST_DEC_PART $end .

    $default  accept


state 6

    3 CONST_DEC_LIST: CONST_DEC_LIST CONST_DEC .

    $default  reduce using rule 3 (CONST_DEC_LIST)


state 7

    5 CONST_DEC: @1 undef_id_tok . '=' const_id_tok ';'

    '='  shift, and go to state 8


state 8

    5 CONST_DEC: @1 undef_id_tok '=' . const_id_tok ';'

    const_id_tok  shift, and go to state 9


state 9

    5 CONST_DEC: @1 undef_id_tok '=' const_id_tok . ';'

    ';'  shift, and go to state 10


state 10

    5 CONST_DEC: @1 undef_id_tok '=' const_id_tok ';' .

    $default  reduce using rule 5 (CONST_DEC)
]])

AT_CLEANUP


## --------------- ##
## Web2c Actions.  ##
## --------------- ##

# The generation of the mapping `state -> action' was once wrong in
# extremely specific situations.  web2c.y exhibits this situation.
# Below is a stripped version of the grammar.  It looks like one can
# simplify it further, but just don't: it is tuned to exhibit a bug,
# which disapears when applying sane grammar transformations.
#
# It used to be wrong on yydefact only:
#
# static const short int yydefact[] =
#  {
# -       2,     0,     1,     0,     0,     2,     3,     2,     5,     4,
# +       2,     0,     1,     0,     0,     0,     3,     2,     5,     4,
#         0,     0
#  };
#
# but let's check all the tables.


AT_SETUP([Web2c Actions])

AT_KEYWORDS([report])

AT_DATA([input.y],
[[%%
statement:  struct_stat;
struct_stat:  /* empty. */ | if else;
if: "if" "const" "then" statement;
else: "else" statement;
%%
]])

AT_CHECK([bison -v -o input.c input.y])

# Check only the tables.  We don't use --no-parser, because it is
# still to be implemented in the experimental branch of Bison.
[sed -n 's/  *$//;/^static const.*\[\] =/,/^}/p' input.c >tables.c]

AT_CHECK([[cat tables.c]], 0,
[[static const unsigned char yytranslate[] =
{
       0,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     2,     2,     2,     2,
       2,     2,     2,     2,     2,     2,     1,     2,     3,     4,
       5,     6
};
static const unsigned char yyprhs[] =
{
       0,     0,     3,     5,     6,     9,    14
};
static const yysigned_char yyrhs[] =
{
       8,     0,    -1,     9,    -1,    -1,    10,    11,    -1,     3,
       4,     5,     8,    -1,     6,     8,    -1
};
static const unsigned char yyrline[] =
{
       0,     2,     2,     3,     3,     4,     5
};
static const char *const yytname[] =
{
  "$end", "error", "$undefined", "\"if\"", "\"const\"", "\"then\"",
  "\"else\"", "$accept", "statement", "struct_stat", "if", "else", 0
};
static const unsigned short int yytoknum[] =
{
       0,   256,   257,   258,   259,   260,   261
};
static const unsigned char yyr1[] =
{
       0,     7,     8,     9,     9,    10,    11
};
static const unsigned char yyr2[] =
{
       0,     2,     1,     0,     2,     4,     2
};
static const unsigned char yydefact[] =
{
       3,     0,     0,     2,     0,     0,     1,     3,     4,     3,
       6,     5
};
static const yysigned_char yydefgoto[] =
{
      -1,     2,     3,     4,     8
};
static const yysigned_char yypact[] =
{
      -2,    -1,     4,    -8,     0,     2,    -8,    -2,    -8,    -2,
      -8,    -8
};
static const yysigned_char yypgoto[] =
{
      -8,    -7,    -8,    -8,    -8
};
static const unsigned char yytable[] =
{
      10,     1,    11,     5,     6,     0,     7,     9
};
static const yysigned_char yycheck[] =
{
       7,     3,     9,     4,     0,    -1,     6,     5
};
static const unsigned char yystos[] =
{
       0,     3,     8,     9,    10,     4,     0,     6,    11,     5,
       8,     8
};
]])

AT_CLEANUP


## ------------------------- ##
## yycheck Bound Violation.  ##
## ------------------------- ##


# _AT_DATA_DANCER_Y(BISON-OPTIONS)
# --------------------------------
# The following grammar, taken from Andrew Suffield's GPL'd implementation
# of DGMTP, the Dancer Generic Message Transport Protocol, used to violate
# yycheck's bounds where issuing a verbose error message.  Keep this test
# so that possible bound checking compilers could check all the skeletons.
m4_define([_AT_DATA_DANCER_Y],
[AT_DATA_GRAMMAR([dancer.y],
[%{
static int yylex (AT_LALR1_CC_IF([int *], [void]));
AT_LALR1_CC_IF([],
[#include <stdio.h>
static void yyerror (const char *);])
%}
$1
%token ARROW INVALID NUMBER STRING DATA
%defines
%verbose
%error-verbose
/* Grammar follows */
%%
line: header body
   ;

header: '<' from ARROW to '>' type ':'
   | '<' ARROW to '>' type ':'
   | ARROW to type ':'
   | type ':'
   | '<' '>'
   ;

from: DATA
   | STRING
   | INVALID
   ;

to: DATA
   | STRING
   | INVALID
   ;

type: DATA
   | STRING
   | INVALID
   ;

body: /* empty */
   | body member
   ;

member: STRING
   | DATA
   | '+' NUMBER
   | '-' NUMBER
   | NUMBER
   | INVALID
   ;
%%
AT_LALR1_CC_IF(
[/* A C++ error reporting function. */
void
yy::parser::error (const location&, const std::string& m)
{
  std::cerr << m << std::endl;
}

int
yyparse ()
{
  yy::parser parser;
  parser.set_debug_level (!!YYDEBUG);
  return parser.parse ();
}
],
[static void
yyerror (const char *s)
{
  fprintf (stderr, "%s\n", s);
}])

static int
yylex (AT_LALR1_CC_IF([int *lval], [void]))
[{
  static int toknum = 0;
  int tokens[] =
    {
      ':', -1
    };
  ]AT_LALR1_CC_IF([*lval = 0; /* Pacify GCC.  */])[
  return tokens[toknum++];
}]

int
main (void)
{
  return yyparse ();
}
])
])# _AT_DATA_DANCER_Y


# AT_CHECK_DANCER(BISON-OPTIONS)
# ------------------------------
# Generate the grammar, compile it, run it.
m4_define([AT_CHECK_DANCER],
[AT_SETUP([Dancer $1])
AT_BISON_OPTION_PUSHDEFS([$1])
_AT_DATA_DANCER_Y([$1])
AT_CHECK([bison -o dancer.c dancer.y])
AT_LALR1_CC_IF(
  [AT_CHECK([bison -o dancer.cc dancer.y])
   AT_COMPILE_CXX([dancer])],
  [AT_CHECK([bison -o dancer.c dancer.y])
   AT_COMPILE([dancer])])
AT_PARSER_CHECK([./dancer], 1, [],
[syntax error, unexpected ':'
])
AT_BISON_OPTION_POPDEFS
AT_CLEANUP
])

AT_CHECK_DANCER()
AT_CHECK_DANCER([%glr-parser])
AT_CHECK_DANCER([%skeleton "lalr1.cc"])