/* Bison Grammar Scanner -*- C -*-
- Copyright (C) 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
+ Copyright (C) 2002-2013 Free Software Foundation, Inc.
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
- 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,
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 noyywrap never-interactive
%option prefix="gram_" outfile="lex.yy.c"
%{
#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 "getargs.h" /* yacc_flag */
-#include "gram.h"
-#include "quotearg.h"
-#include "reader.h"
-#include "uniqstr.h"
+#include <src/complain.h>
+#include <src/files.h>
+#include <src/getargs.h>
+#include <src/gram.h>
+#include <quotearg.h>
+#include <src/reader.h>
+#include <src/uniqstr.h>
+#include <c-ctype.h>
#include <mbswidth.h>
#include <quote.h>
-#include "scan-gram.h"
+#include <src/scan-gram.h>
#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. */
-boundary scanner_cursor;
+static boundary scanner_cursor;
#define YY_USER_ACTION location_compute (loc, &scanner_cursor, yytext, yyleng);
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)
+
+#define DEPRECATED(Msg) \
+ do { \
+ size_t i; \
+ deprecated_directive (loc, yytext, Msg); \
+ scanner_cursor.column -= mbsnwidth (Msg, strlen (Msg), 0); \
+ for (i = strlen (Msg); i != 0; --i) \
+ unput (Msg[i - 1]); \
+ } while (0)
+
/* 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)
STRING_FREE;
}
-/* The location of the most recently saved token, if it was a
- BRACED_CODE token; otherwise, this has an unspecified value. */
-location gram_last_braced_code_loc;
-
static void handle_syncline (char *, location);
static unsigned long int scan_integer (char const *p, int base, location loc);
static int convert_ucn_to_byte (char const *hex_text);
/* 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
- /* A keyword that should be followed by some code was read (e.g.
- %printer). */
-%x SC_PRE_CODE
+ /* 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);
- - 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
+ /* 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_]
+notletter [^.abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_]{-}[%\{]
+id {letter}({letter}|[-0-9])*
+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
- 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. */
-splice (\\[ \f\t\v]*\n)*
+splice (\\[ \f\t\v]*\n)*
+
+/* An equal sign, with optional leading whitespaces. This is used in some
+ deprecated constructs. */
+eqopt ([[:space:]]*=)?
%%
%{
- /* 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. */
- int context_state IF_LINT (= 0);
-
- /* Token type to return, when applicable. */
- int token_type IF_LINT (= 0);
+ int context_state PACIFY_CC (= 0);
/* 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 comment or string or character literal started,
when applicable. */
- boundary token_start IF_LINT (= scanner_cursor);
+ boundary token_start PACIFY_CC (= scanner_cursor);
%}
| Scanning white space. |
`-----------------------*/
-<INITIAL,SC_AFTER_IDENTIFIER,SC_PRE_CODE>
+<INITIAL,SC_AFTER_IDENTIFIER,SC_BRACKETED_ID,SC_RETURN_BRACKETED_ID>
{
/* Comments and white space. */
- "," warn_at (*loc, _("stray `,' treated as white space"));
+ "," {
+ complain (loc, Wother, _("stray ',' treated as white space"));
+ }
[ \f\n\t\v] |
- "//".* ;
+ "//".* continue;
"/*" {
token_start = loc->start;
context_state = YY_START;
/* #line directives are not documented, and may be withdrawn or
modified in future versions of Bison. */
- ^"#line "{int}" \"".*"\"\n" {
+ ^"#line "{int}(" \"".*"\"")?"\n" {
handle_syncline (yytext + sizeof "#line " - 1, *loc);
}
}
/*----------------------------.
| 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>
{
- "%binary" return PERCENT_NONASSOC;
- "%debug" return PERCENT_DEBUG;
- "%default"[-_]"prec" return PERCENT_DEFAULT_PREC;
- "%define" return PERCENT_DEFINE;
- "%defines" return PERCENT_DEFINES;
- "%destructor" /* FIXME: Remove once %union handled differently. */ token_type = BRACED_CODE; 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;
- "%initial-action" /* FIXME: Remove once %union handled differently. */ token_type = BRACED_CODE; return PERCENT_INITIAL_ACTION;
- "%glr-parser" return PERCENT_GLR_PARSER;
- "%left" return PERCENT_LEFT;
- "%lex-param" /* FIXME: Remove once %union handled differently. */ token_type = BRACED_CODE; 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" /* FIXME: Remove once %union handled differently. */ token_type = BRACED_CODE; return PERCENT_PARSE_PARAM;
- "%prec" return PERCENT_PREC;
- "%printer" /* FIXME: Remove once %union handled differently. */ token_type = BRACED_CODE; 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;
- "%term" return PERCENT_TOKEN;
- "%token" return PERCENT_TOKEN;
- "%token"[-_]"table" return PERCENT_TOKEN_TABLE;
- "%type" return PERCENT_TYPE;
- "%union" token_type = PERCENT_UNION; BEGIN SC_PRE_CODE;
- "%verbose" return PERCENT_VERBOSE;
- "%yacc" return PERCENT_YACC;
-
- {directive} {
- complain_at (*loc, _("invalid directive: %s"), quote (yytext));
+ "%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;
+ "%empty" return PERCENT_EMPTY;
+ "%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;
+ "%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;
+
+ /* deprecated */
+ "%default"[-_]"prec" DEPRECATED("%default-prec");
+ "%error"[-_]"verbose" DEPRECATED("%define parse.error verbose");
+ "%expect"[-_]"rr" DEPRECATED("%expect-rr");
+ "%file-prefix"{eqopt} DEPRECATED("%file-prefix");
+ "%fixed"[-_]"output"[-_]"files" DEPRECATED("%fixed-output-files");
+ "%name"[-_]"prefix"{eqopt} DEPRECATED("%name-prefix");
+ "%no"[-_]"default"[-_]"prec" DEPRECATED("%no-default-prec");
+ "%no"[-_]"lines" DEPRECATED("%no-lines");
+ "%output"{eqopt} DEPRECATED("%output");
+ "%pure"[-_]"parser" DEPRECATED("%pure-parser");
+ "%token"[-_]"table" DEPRECATED("%token-table");
+
+ "%"{id}|"%"{notletter}([[:graph:]])+ {
+ complain (loc, complaint, _("invalid directive: %s"), quote (yytext));
}
"=" return EQUAL;
";" return SEMICOLON;
{id} {
- val->symbol = symbol_get (yytext, *loc);
+ val->uniqstr = uniqstr_new (yytext);
id_loc = *loc;
+ bracketed_id_str = NULL;
BEGIN SC_AFTER_IDENTIFIER;
}
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 (loc, complaint, _("invalid identifier: %s"), quote (yytext));
+ }
+
+ /* Characters. */
+ "'" token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER;
/* 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;
/* Code in between braces. */
"{" {
- if (current_rule && current_rule->action)
- grammar_midrule_action ();
STRING_GROW;
- token_type = BRACED_CODE;
- braces_level = 0;
+ nesting = 0;
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. */
+ "<*>" 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;
- return TYPE;
+ return TAG;
+ }
+ "<" {
+ nesting = 0;
+ token_start = loc->start;
+ BEGIN SC_TAG;
}
"%%" {
return PERCENT_PERCENT;
}
- . {
- complain_at (*loc, _("invalid character: %s"), quote (yytext));
+ "[" {
+ bracketed_id_str = NULL;
+ bracketed_id_start = loc->start;
+ bracketed_id_context_state = YY_START;
+ BEGIN SC_BRACKETED_ID;
+ }
+
+ [^\[%A-Za-z0-9_<>{}\"\'*;|=/, \f\n\t\v]+|. {
+ complain (loc, complaint, "%s: %s",
+ ngettext ("invalid character", "invalid characters", yyleng),
+ quote_mem (yytext, yyleng));
}
<<EOF>> {
}
+ /*--------------------------------------------------------------.
+ | Supporting \0 complexifies our implementation for no expected |
+ | added value. |
+ `--------------------------------------------------------------*/
+
+<SC_ESCAPED_CHARACTER,SC_ESCAPED_STRING,SC_TAG>
+{
+ \0 complain (loc, complaint, _("invalid null character"));
+}
+
+
/*-----------------------------------------------------------------.
| 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;
- BEGIN INITIAL;
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;
- BEGIN INITIAL;
return ID;
}
<<EOF>> {
+ BEGIN (bracketed_id_str ? SC_RETURN_BRACKETED_ID : INITIAL);
*loc = id_loc;
- BEGIN INITIAL;
return ID;
}
}
+ /*--------------------------------.
+ | Scanning bracketed identifiers. |
+ `--------------------------------*/
+
+<SC_BRACKETED_ID>
+{
+ {id} {
+ if (bracketed_id_str)
+ {
+ complain (loc, complaint,
+ _("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 (loc, complaint, _("an identifier expected"));
+ }
+
+ [^\].A-Za-z0-9_/ \f\n\t\v]+|. {
+ complain (loc, complaint, "%s: %s",
+ ngettext ("invalid character in bracketed name",
+ "invalid characters in bracketed name", yyleng),
+ quote_mem (yytext, yyleng));
+ }
+
+ <<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;
- .|\n ;
+ .|\n continue;
<<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;
- <<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>
{
- "\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;
}
BEGIN INITIAL;
return STRING;
}
- \n unexpected_newline (token_start, "\""); BEGIN INITIAL;
- <<EOF>> unexpected_eof (token_start, "\""); BEGIN INITIAL;
+ <<EOF>> unexpected_eof (token_start, "\"");
+ "\n" unexpected_newline (token_start, "\"");
}
/*----------------------------------------------------------.
| Scanning a Bison character literal, decoding its escapes. |
- | The initial quote is already eaten. |
+ | The initial quote is already eaten. |
`----------------------------------------------------------*/
<SC_ESCAPED_CHARACTER>
{
"'" {
- unsigned char last_string_1;
- STRING_GROW;
STRING_FINISH;
loc->start = token_start;
- val->symbol = symbol_get (quotearg_style (escape_quoting_style,
- last_string),
- *loc);
- symbol_class_set (val->symbol, token_sym, *loc, false);
- last_string_1 = last_string[1];
- symbol_user_token_number_set (val->symbol, last_string_1, *loc);
+ val->character = last_string[0];
+
+ /* FIXME: Eventually, make these errors. */
+ if (last_string[0] == '\0')
+ {
+ complain (loc, Wother, _("empty character literal"));
+ /* '\0' seems dangerous even if we are about to complain. */
+ val->character = '\'';
+ }
+ else if (last_string[1] != '\0')
+ complain (loc, Wother,
+ _("extra characters in character literal"));
STRING_FREE;
BEGIN INITIAL;
- return ID;
+ return CHAR;
}
- \n unexpected_newline (token_start, "'"); BEGIN INITIAL;
- <<EOF>> unexpected_eof (token_start, "'"); BEGIN INITIAL;
+ "\n" unexpected_newline (token_start, "'");
+ <<EOF>> unexpected_eof (token_start, "'");
}
-<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, ">");
+}
/*----------------------------.
| Decode escaped characters. |
{
\\[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 (loc, complaint, _("invalid number after \\-escape: %s"),
+ yytext+1);
else
obstack_1grow (&obstack_for_string, c);
}
\\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 (loc, complaint, _("invalid number after \\-escape: %s"),
+ yytext+1);
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);
- 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 (loc, complaint, _("invalid number after \\-escape: %s"),
+ yytext+1);
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 (c_isspace ((unsigned char) *p) && c_isprint ((unsigned char) *p))
+ p = quote (p);
+ else
+ p = quotearg_style_mem (escape_quoting_style, p, 1);
+ complain (loc, complaint, _("invalid character after \\-escape: %s"),
+ p);
}
}
<SC_CHARACTER,SC_STRING>
{
- {splice}|\\{splice}[^\n\[\]] STRING_GROW;
+ {splice}|\\{splice}[^\n\[\]] STRING_GROW;
}
<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, "'");
+ <<EOF>> unexpected_eof (token_start, "'");
}
<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, "\"");
+ <<EOF>> unexpected_eof (token_start, "\"");
}
| 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;
}
- /*---------------------------------------------------------------.
- | Scanning after %union etc., possibly followed by white space. |
- | For %union only, allow arbitrary C code to appear before the |
- | following brace, as an extension to POSIX. |
- `---------------------------------------------------------------*/
-<SC_PRE_CODE>
+ /*-----------------------------------------------------------.
+ | Scanning some code in braces (actions, predicates). The |
+ | initial "{" is already eaten. |
+ `-----------------------------------------------------------*/
+
+<SC_BRACED_CODE,SC_PREDICATE>
{
- . {
- bool valid = yytext[0] == '{' || token_type == PERCENT_UNION;
- scanner_cursor.column -= mbsnwidth (yytext, yyleng, 0);
- yyless (0);
+ "{"|"<"{splice}"%" STRING_GROW; nesting++;
+ "%"{splice}">" STRING_GROW; nesting--;
- if (valid)
- {
- braces_level = -1;
- code_start = loc->start;
- BEGIN SC_BRACED_CODE;
- }
- else
- {
- complain_at (*loc, _("missing `{' in %s"),
- token_name (token_type));
- obstack_sgrow (&obstack_for_string, "{}");
- STRING_FINISH;
- val->chars = last_string;
- BEGIN INITIAL;
- return token_type;
- }
- }
+ /* Tokenize '<<%' correctly (as '<<' '%') rather than incorrrectly
+ (as '<' '<%'). */
+ "<"{splice}"<" STRING_GROW;
- <<EOF>> unexpected_eof (scanner_cursor, "{}"); BEGIN INITIAL;
+ <<EOF>> unexpected_eof (code_start, "}");
}
-
- /*---------------------------------------------------------------.
- | Scanning some code in braces (%union and actions). The initial |
- | "{" is already eaten. |
- `---------------------------------------------------------------*/
-
<SC_BRACED_CODE>
{
- "{"|"<"{splice}"%" STRING_GROW; braces_level++;
- "%"{splice}">" STRING_GROW; braces_level--;
"}" {
- bool outer_brace = --braces_level < 0;
-
- /* As an undocumented Bison extension, append `;' before the last
- brace in braced code, so that the user code can omit trailing
- `;'. But do not append `;' if emulating Yacc, since Yacc does
- not append one.
-
- FIXME: Bison should warn if a semicolon seems to be necessary
- here, and should omit the semicolon if it seems unnecessary
- (e.g., after ';', '{', or '}', each followed by comments or
- white space). Such a warning shouldn't depend on --yacc; it
- should depend on a new --pedantic option, which would cause
- Bison to warn if it detects an extension to POSIX. --pedantic
- should also diagnose other Bison extensions like %yacc.
- Perhaps there should also be a GCC-style --pedantic-errors
- option, so that such warnings are diagnosed as errors. */
- if (outer_brace && token_type == BRACED_CODE && ! yacc_flag)
- obstack_1grow (&obstack_for_string, ';');
-
obstack_1grow (&obstack_for_string, '}');
- if (outer_brace)
+ --nesting;
+ if (nesting < 0)
{
- STRING_FINISH;
- loc->start = code_start;
- val->chars = last_string;
- gram_last_braced_code_loc = *loc;
- BEGIN INITIAL;
- return token_type;
+ STRING_FINISH;
+ loc->start = code_start;
+ val->code = last_string;
+ BEGIN INITIAL;
+ return BRACED_CODE;
}
}
-
- /* Tokenize `<<%' correctly (as `<<' `%') rather than incorrrectly
- (as `<' `<%'). */
- "<"{splice}"<" STRING_GROW;
-
- <<EOF>> unexpected_eof (code_start, "}"); BEGIN INITIAL;
}
+<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 "%}". |
return PROLOGUE;
}
- <<EOF>> unexpected_eof (code_start, "%}"); BEGIN INITIAL;
+ <<EOF>> unexpected_eof (code_start, "%}");
}
| 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;
%%
{
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;
if (INT_MAX < num)
{
- complain_at (loc, _("integer out of range: %s"), quote (number));
+ complain (&loc, complaint, _("integer out of range: %s"),
+ quote (number));
num = INT_MAX;
}
/*------------------------------------------------------------------.
| 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
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;
}
-/*----------------------------------------------------------------.
-| Handle `#line INT "FILE"'. ARGS has already skipped `#line '. |
-`----------------------------------------------------------------*/
+/*---------------------------------------------------------------------.
+| Handle '#line INT( "FILE")?\n'. ARGS has already skipped '#line '. |
+`---------------------------------------------------------------------*/
static void
handle_syncline (char *args, location loc)
{
- char *after_num;
- unsigned long int lineno = strtoul (args, &after_num, 10);
- char *file = strchr (after_num, '"') + 1;
- *strchr (file, '"') = '\0';
+ char *file;
+ unsigned long int lineno = strtoul (args, &file, 10);
if (INT_MAX <= lineno)
{
- warn_at (loc, _("line number overflow"));
+ complain (&loc, Wother, _("line number overflow"));
lineno = INT_MAX;
}
- current_file = uniqstr_new (file);
- boundary_set, (&scanner_cursor, current_file, lineno, 1);
+
+ file = strchr (file, '"');
+ if (file)
+ {
+ *strchr (file + 1, '"') = '\0';
+ current_file = uniqstr_new (file + 1);
+ }
+ boundary_set (&scanner_cursor, current_file, lineno, 1);
}
/*----------------------------------------------------------------.
| 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. Then, pretend that TOKEN_END was found. |
`----------------------------------------------------------------*/
static void
location loc;
loc.start = start;
loc.end = scanner_cursor;
- complain_at (loc, _(msgid), token_end);
+ size_t i = strlen (token_end);
+
+/* Adjust scanner cursor so that any later message does not count
+ the characters about to be inserted. */
+ scanner_cursor.column -= i;
+
+ while (i != 0)
+ unput (token_end[--i]);
+
+ token_end = quote (token_end);
+ /* Instead of '\'', display "'". */
+ if (STREQ (token_end, "'\\''"))
+ token_end = "\"'\"";
+ complain (&loc, complaint, _(msgid), token_end);
}
/*------------------------------------------------------------------------.
| Report an unexpected EOF in a token or comment starting at START. |
| An end of file was encountered and the expected TOKEN_END was missing. |
+| After reporting the problem, pretend that TOKEN_END was found. |
`------------------------------------------------------------------------*/
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);
}
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);
}
{
obstack_free (&obstack_for_string, 0);
/* Reclaim Flex's buffers. */
- yy_delete_buffer (YY_CURRENT_BUFFER);
+ yylex_destroy ();
}
projects / bison.git / blobdiff