Future/bison
1
0
Fork 0
mirror of https://git.savannah.gnu.org/git/bison.git synced 2026-03-11 05:13:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

Allow specification of semantic predicates.

Browse Source 
These changes allow users to prefix an action with %? to indicate that it
is a semantic predicate---an expression that is evaluated immediately (not
deferred, even in GLR nondeterministic mode) and causes a syntax error if
false.  In GLR parsers, this has the effect of killing one of a set of
split-off parses, just as would an ordinary syntax error.

Changelog:

    * NEWS: Describe new semantic-predicate feature.
    * data/c.m4 (b4_predicate_case): New definition.
    * data/java.m4 (b4_predicate_case): New definition.
    * data/glr.c (yyimmediate): Add definition.
    (yydoAction): Remove comment, now obsolete.
    Do YY_REDUCE_PRINT here.
    (yyglrReduce): Alter comment to indicate that semantic values
    need not be deferred.
    Remove YY_REDUCE_PRINT from here; done in yydoAction.
    (yyprocessOneStack): Pass immediate flag.
    Delete stacks rejected by predicates in newly split-off parsers.
    Change handling of yyerr so that only current stack gets deleted
    when semantic predicate fails.
    (yyfillin): Don't crash if a semantic value is unresolved (as may
    happen in predicate rules).
    Copy lr state as well in debugging mode.
    Update comment on setting of yysval to include yyloc as well.
    (yy_reduce_print): Add yynormal argument.  Perform fillin properly.
    Report unresolved RHS values.
    (yyimmediate): New table.
    * src/gram.h (struct rule): Add is_predicate field.
    * src/output.c (user_actions_output): Use b4_predicate_case for
    predicates.
    (prepare_symbols): Output yyimmediate.
    * src/scan-gram.l: Add %? token, SC_PREDICATE state.
    * src/scan-code.l (code_props_rule_action_init): Add is_predicate
    argument.
    * src/scan-code.h (struct code_props): Add is_predicate field.
    (code_props_rule_action_init): New interface.
    * src/parse-gram.y (%?{...}): New token.
    (rhs): Add %?{...} rule.
    * src/parse-gram.c: Regenerate.
    * src/parse-gram.h: Regenerate.
    * src/reader.c (grammar_current_rule_action_append): Add
    immediate argument.
    (grammar_midrule_action): Use new interface for
    code_props_rule_action_init.
    (grammar_current_rule_action_append): Ditto.
    (packgram): Transfer is_predicate value.
    * src/reader.h (grammar_current_rule_action_append): New interface.
    * doc/bison.texinfo: Document semantic predicates (%?).

    * data/glr.c (yylhsNonterm, yyisDefaultedState,yyDefaultAction)
    (yygetLRActions,yynewGLRStackItem,yyaddDeferredAction,yyinitStateSet)
    (yyinitGLRStack,yyexpandGLRStack,yyupdateSplit,yymarkStackDeleted)
    (yyundeleteLastStack,yyglrShift,yyglrShiftDefer,yydoAction,yyglrReduce)
    (yyidenticalOptions,yymergeOptionSets,yyresolveStates,yyresolveAction)
    (yyresolveLocations,yyresolveValue,yyreducePrint): Update parameter
    names in comments and mention all parameters.
    (struct yyGLRState): Fix description of yyposn field.
    (yyresolveLocations): Correct comment so as not to imply action when
    yyn1==0.
This commit is contained in:
Paul Hilfinger
2010-07-22 19:08:10 -07:00
parent 804e83b26d
commit ca2a6d1587
16 changed files with 919 additions and 646 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
src/gram.h
View File

@@ -194,6 +194,7 @@ typedef struct
location location;
bool useful;
bool is_predicate;
const char *action;
location action_location;

11
src/output.c
View File

@@ -204,7 +204,7 @@ prepare_symbols (void)
/*----------------------------------------------------------------.
| Prepare the muscles related to the rules: r1, r2, rline, dprec, |
| merger. |
| merger, immediate. |
`----------------------------------------------------------------*/
static void
@@ -215,6 +215,7 @@ prepare_rules (void)
unsigned int *r2 = xnmalloc (nrules, sizeof *r2);
int *dprec = xnmalloc (nrules, sizeof *dprec);
int *merger = xnmalloc (nrules, sizeof *merger);
int *immediate = xnmalloc (nrules, sizeof *immediate);
rule_number r;
for (r = 0; r < nrules; ++r)
@@ -229,6 +230,8 @@ prepare_rules (void)
dprec[r] = rules[r].dprec;
/* Merger-function index (GLR). */
merger[r] = rules[r].merger;
/* Immediate reduction flags (GLR). */
immediate[r] = rules[r].is_predicate;
}
muscle_insert_unsigned_int_table ("rline", rline, 0, 0, nrules);
@@ -236,6 +239,7 @@ prepare_rules (void)
muscle_insert_unsigned_int_table ("r2", r2, 0, 0, nrules);
muscle_insert_int_table ("dprec", dprec, 0, 0, nrules);
muscle_insert_int_table ("merger", merger, 0, 0, nrules);
muscle_insert_int_table ("immediate", immediate, 0, 0, nrules);
MUSCLE_INSERT_INT ("rules_number", nrules);
MUSCLE_INSERT_INT ("max_left_semantic_context", max_left_semantic_context);
@@ -349,8 +353,9 @@ user_actions_output (FILE *out)
for (r = 0; r < nrules; ++r)
if (rules[r].action)
{
fprintf (out, "b4_case(%d, [b4_syncline(%d, ", r + 1,
rules[r].action_location.start.line);
fprintf (out, "b4_%scase(%d, [b4_syncline(%d, ",
rules[r].is_predicate ? "predicate_" : "",
r + 1, rules[r].action_location.start.line);
escaped_output (out, rules[r].action_location.start.file);
fprintf (out, ")\n[ %s]])\n\n", rules[r].action);
}

939
src/parse-gram.c
View File

File diff suppressed because it is too large Load Diff

77
src/parse-gram.h
View File

@@ -1,8 +1,9 @@
/* A Bison parser, made by GNU Bison 2.4.483-4ad39-dirty. */
/* A Bison parser, made by GNU Bison 2.4.516-804e. */
/* Interface for Bison's Yacc-like parsers in C
Copyright (C) 1984, 1989-1990, 2000-2010 Free Software Foundation, Inc.
Copyright (C) 1984, 1989, 1990, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
2007, 2008, 2009, 2010 Free Software Foundation, Inc.
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
@@ -31,7 +32,7 @@
version 2.2 of Bison. */
/* "%code requires" blocks. */
/* Line 1658 of yacc.c */
#line 201 "src/parse-gram.y"
#line 202 "src/parse-gram.y"
# ifndef PARAM_TYPE
# define PARAM_TYPE
@@ -46,7 +47,7 @@
/* Line 1658 of yacc.c */
#line 50 "src/parse-gram.h"
#line 51 "src/parse-gram.h"
/* Tokens. */
#ifndef YYTOKENTYPE
@@ -93,21 +94,22 @@
PERCENT_VERBOSE = 293,
PERCENT_YACC = 294,
BRACED_CODE = 295,
BRACKETED_ID = 296,
CHAR = 297,
EPILOGUE = 298,
EQUAL = 299,
ID = 300,
ID_COLON = 301,
PERCENT_PERCENT = 302,
PIPE = 303,
PROLOGUE = 304,
SEMICOLON = 305,
TAG = 306,
TAG_ANY = 307,
TAG_NONE = 308,
PERCENT_PARAM = 309,
PERCENT_UNION = 310
BRACED_PREDICATE = 296,
BRACKETED_ID = 297,
CHAR = 298,
EPILOGUE = 299,
EQUAL = 300,
ID = 301,
ID_COLON = 302,
PERCENT_PERCENT = 303,
PIPE = 304,
PROLOGUE = 305,
SEMICOLON = 306,
TAG = 307,
TAG_ANY = 308,
TAG_NONE = 309,
PERCENT_PARAM = 310,
PERCENT_UNION = 311
};
#endif
/* Tokens. */
@@ -150,21 +152,22 @@
#define PERCENT_VERBOSE 293
#define PERCENT_YACC 294
#define BRACED_CODE 295
#define BRACKETED_ID 296
#define CHAR 297
#define EPILOGUE 298
#define EQUAL 299
#define ID 300
#define ID_COLON 301
#define PERCENT_PERCENT 302
#define PIPE 303
#define PROLOGUE 304
#define SEMICOLON 305
#define TAG 306
#define TAG_ANY 307
#define TAG_NONE 308
#define PERCENT_PARAM 309
#define PERCENT_UNION 310
#define BRACED_PREDICATE 296
#define BRACKETED_ID 297
#define CHAR 298
#define EPILOGUE 299
#define EQUAL 300
#define ID 301
#define ID_COLON 302
#define PERCENT_PERCENT 303
#define PIPE 304
#define PROLOGUE 305
#define SEMICOLON 306
#define TAG 307
#define TAG_ANY 308
#define TAG_NONE 309
#define PERCENT_PARAM 310
#define PERCENT_UNION 311
@@ -173,7 +176,7 @@
typedef union YYSTYPE
{
/* Line 1658 of yacc.c */
#line 87 "src/parse-gram.y"
#line 88 "src/parse-gram.y"
assoc assoc;
char *code;
@@ -186,13 +189,13 @@ typedef union YYSTYPE
unsigned char character;
/* Line 1658 of yacc.c */
#line 225 "src/parse-gram.y"
#line 226 "src/parse-gram.y"
param_type param;
/* Line 1658 of yacc.c */
#line 196 "src/parse-gram.h"
#line 199 "src/parse-gram.h"
} YYSTYPE;
# define YYSTYPE_IS_TRIVIAL 1
# define yystype YYSTYPE /* obsolescent; will be withdrawn */

9
src/parse-gram.y
View File

@@ -118,7 +118,6 @@ static char const *char_name (char);
%token PERCENT_DPREC "%dprec"
%token PERCENT_MERGE "%merge"
/*----------------------.
| Global Declarations. |
`----------------------*/
@@ -151,6 +150,7 @@ static char const *char_name (char);
;
%token BRACED_CODE "{...}"
%token BRACED_PREDICATE "%?{...}"
%token BRACKETED_ID "[identifier]"
%token CHAR "char"
%token EPILOGUE "epilogue"
@@ -171,7 +171,7 @@ static char const *char_name (char);
/* braceless is not to be used for rule or symbol actions, as it
calls code_props_plain_init. */
%type <chars> STRING "%{...%}" EPILOGUE braceless content.opt
%type <code> "{...}"
%type <code> "{...}" "%?{...}"
%printer { fputs (quotearg_style (c_quoting_style, $$), stderr); }
STRING
%printer { fprintf (stderr, "{\n%s\n}", $$); }
@@ -586,7 +586,9 @@ rhs:
| rhs symbol named_ref.opt
{ grammar_current_rule_symbol_append ($2, @2, $3); }
| rhs "{...}" named_ref.opt
{ grammar_current_rule_action_append ($2, @2, $3); }
{ grammar_current_rule_action_append ($2, @2, $3, false); }
| rhs "%?{...}"
{ grammar_current_rule_action_append ($2, @2, NULL, true); }
| rhs "%prec" symbol
{ grammar_current_rule_prec_set ($3, @3); }
| rhs "%dprec" INT
@@ -601,7 +603,6 @@ named_ref.opt:
BRACKETED_ID { $$ = named_ref_new($1, @1); }
;
/*---------------------------.
| variable and content.opt. |
`---------------------------*/

8
src/reader.c
View File

@@ -377,7 +377,8 @@ grammar_midrule_action (void)
code_props_rule_action_init (&midrule->action_props,
current_rule->action_props.code,
current_rule->action_props.location,
midrule, 0);
midrule, 0,
current_rule->action_props.is_predicate);
code_props_none_init (&current_rule->action_props);
if (previous_rule_end)
@@ -468,14 +469,14 @@ grammar_current_rule_symbol_append (symbol *sym, location loc,
void
grammar_current_rule_action_append (const char *action, location loc,
named_ref *name)
named_ref *name, bool is_predicate)
{
if (current_rule->action_props.code)
grammar_midrule_action ();
/* After all symbol declarations have been parsed, packgram invokes
code_props_translate_code. */
code_props_rule_action_init (&current_rule->action_props, action, loc,
current_rule, name);
current_rule, name, is_predicate);
}
@@ -516,6 +517,7 @@ packgram (void)
rules[ruleno].useful = true;
rules[ruleno].action = p->action_props.code;
rules[ruleno].action_location = p->action_props.location;
rules[ruleno].is_predicate = p->action_props.is_predicate;
/* If the midrule's $$ is set or its $n is used, remove the `$' from the
symbol name so that it's a user-defined symbol so that the default

2
src/reader.h
View File

@@ -53,7 +53,7 @@ void grammar_current_rule_merge_set (uniqstr name, location loc);
void grammar_current_rule_symbol_append (symbol *sym, location loc,
named_ref *named_ref);
void grammar_current_rule_action_append (const char *action, location loc,
named_ref *named_ref);
named_ref *named_ref, bool);
void reader (void);
void free_merger_functions (void);

11
src/scan-code.h
View File

@@ -63,6 +63,13 @@ typedef struct code_props {
*/
bool is_value_used;
/**
* \c true iff this code is an action that is not to be deferred in
* a non-deterministic parser.
*/
bool is_predicate;
/** \c NULL iff \c code_props::kind is not \c CODE_PROPS_RULE_ACTION. */
struct symbol_list *rule;
@@ -80,7 +87,7 @@ void code_props_none_init (code_props *self);
/** Equivalent to \c code_props_none_init. */
#define CODE_PROPS_NONE_INIT \
{CODE_PROPS_NONE, NULL, EMPTY_LOCATION_INIT, false, NULL, NULL}
{CODE_PROPS_NONE, NULL, EMPTY_LOCATION_INIT, false, false, NULL, NULL}
/** Initialized by \c CODE_PROPS_NONE_INIT with no further modification. */
extern code_props const code_props_none;
@@ -134,7 +141,7 @@ void code_props_symbol_action_init (code_props *self, char const *code,
*/
void code_props_rule_action_init (code_props *self, char const *code,
location code_loc, struct symbol_list *rule,
named_ref *name);
named_ref *name, bool is_predicate);
/**
* \pre

3
src/scan-code.l
View File

@@ -902,7 +902,7 @@ code_props_symbol_action_init (code_props *self, char const *code,
void
code_props_rule_action_init (code_props *self, char const *code,
location code_loc, symbol_list *rule,
named_ref *name)
named_ref *name, bool is_predicate)
{
self->kind = CODE_PROPS_RULE_ACTION;
self->code = code;
@@ -910,6 +910,7 @@ code_props_rule_action_init (code_props *self, char const *code,
self->is_value_used = false;
self->rule = rule;
self->named_ref = name;
self->is_predicate = is_predicate;
}
void

73
src/scan-gram.l
View File

@@ -106,11 +106,12 @@ static void unexpected_newline (boundary, char const *);
/* A complex tag, with nested angles brackets. */
%x SC_TAG
/* Three types of user code:
/* Four types of user code:
- prologue (code between `%{' `%}' in the first section, before %%);
- 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. */
@@ -285,6 +286,13 @@ splice (\\[ \f\t\v]*\n)*
BEGIN SC_BRACED_CODE;
}
/* Semantic predicate. */
"%?"[ \f\n\t\v]*"{" {
nesting = 0;
code_start = loc->start;
BEGIN SC_PREDICATE;
}
/* A type. */
"<*>" return TAG_ANY;
"<>" return TAG_NONE;
@@ -661,7 +669,7 @@ splice (\\[ \f\t\v]*\n)*
| 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;
@@ -691,14 +699,32 @@ 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; nesting++;
"%"{splice}">" STRING_GROW; nesting--;
/* Tokenize `<<%' correctly (as `<<' `%') rather than incorrrectly
(as `<' `<%'). */
"<"{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;
val->code = last_string;
BEGIN INITIAL;
return token;
}
}
<SC_BRACED_CODE>
{
"}" {
obstack_1grow (&obstack_for_string, '}');
@@ -712,21 +738,24 @@ splice (\\[ \f\t\v]*\n)*
return BRACED_CODE;
}
}
/* Tokenize `<<%' correctly (as `<<' `%') rather than incorrrectly
(as `<' `<%'). */
"<"{splice}"<" STRING_GROW;
<<EOF>> {
unexpected_eof (code_start, "}");
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 "%}". |
@@ -774,8 +803,8 @@ splice (\\[ \f\t\v]*\n)*
| 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;
%%