/* Bison Grammar Scanner -*- C -*-
- Copyright (C) 2002-2011 Free Software Foundation, Inc.
+ Copyright (C) 2002-2012 Free Software Foundation, Inc.
This file is part of Bison, the GNU Compiler Compiler.
#include <src/reader.h>
#include <src/uniqstr.h>
-#include <ctype.h>
+#include <c-ctype.h>
#include <mbswidth.h>
#include <quote.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. */
static boundary scanner_cursor;
#define ROLLBACK_CURRENT_TOKEN \
do { \
- scanner_cursor.column -= mbsnwidth (yytext, yyleng, 0); \
+ scanner_cursor.column -= mbsnwidth (yytext, yyleng, 0); \
yyless (0); \
} while (0)
/* 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 complex tag, with nested angles brackets. */
%x SC_TAG
/* Four types of user code:
- - prologue (code between `%{' `%}' in the first section, before %%);
+ - prologue (code between '%{' '%}' in the first section, before %%);
- actions, printers, union, etc, (between braced in the middle section);
- - epilogue (everything after the second %%).
- - predicate (code between `%?{' and `{' in middle section); */
+ - 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
/* Bracketed identifiers support. */
%x SC_BRACKETED_ID SC_RETURN_BRACKETED_ID
-letter [-.abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_]
-id {letter}({letter}|[0-9])*
+letter [.abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_]
+id {letter}({letter}|[-0-9])*
directive %{id}
-int [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, 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<>]+
+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)*
%%
%{
/* Nesting level. Either for nested braces, or nested angle brackets
(but not mixed). */
- int nesting IF_LINT (= 0);
+ int nesting PACIFY_CC (= 0);
/* 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 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);
%}
<INITIAL,SC_AFTER_IDENTIFIER,SC_BRACKETED_ID,SC_RETURN_BRACKETED_ID>
{
/* Comments and white space. */
- "," warn_at (*loc, _("stray `,' treated as white space"));
+ "," {
+ complain_at (*loc, Wother, _("stray ',' treated as white space"));
+ }
[ \f\n\t\v] |
"//".* ;
"/*" {
`----------------------------*/
/* 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. */
"%yacc" return PERCENT_YACC;
{directive} {
- complain_at (*loc, _("invalid directive: %s"), quote (yytext));
+ complain_at (*loc, complaint, _("invalid directive: %s"), quote (yytext));
}
"=" return EQUAL;
/* 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));
+ complain_at (*loc, complaint, _("invalid identifier: %s"), quote (yytext));
}
/* Characters. */
- "'" token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER;
+ "'" 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;
}
. {
- complain_at (*loc, _("invalid character: %s"), quote (yytext));
+ complain_at (*loc, complaint, _("invalid character: %s"), quote (yytext));
}
<<EOF>> {
<SC_ESCAPED_CHARACTER,SC_ESCAPED_STRING,SC_TAG>
{
- \0 complain_at (*loc, _("invalid null character"));
+ \0 complain_at (*loc, complaint, _("invalid null character"));
}
"[" {
if (bracketed_id_str)
{
- ROLLBACK_CURRENT_TOKEN;
- BEGIN SC_RETURN_BRACKETED_ID;
- *loc = id_loc;
- return ID;
+ 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;
+ bracketed_id_start = loc->start;
+ bracketed_id_context_state = YY_START;
+ BEGIN SC_BRACKETED_ID;
}
}
":" {
{id} {
if (bracketed_id_str)
{
- complain_at (*loc, _("unexpected identifier in bracketed name: %s"),
- quote (yytext));
+ complain_at (*loc, complaint,
+ _("unexpected identifier in bracketed name: %s"),
+ quote (yytext));
}
else
{
- bracketed_id_str = uniqstr_new (yytext);
- bracketed_id_loc = *loc;
+ 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;
- }
+ 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, complaint, _("an identifier expected"));
}
. {
- complain_at (*loc, _("invalid character in bracketed name: %s"),
- quote (yytext));
+ complain_at (*loc, complaint, _("invalid character in bracketed name: %s"),
+ quote (yytext));
}
<<EOF>> {
BEGIN bracketed_id_context_state;
/*---------------------------------------------------------------.
- | 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 ;
<<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;
}
/*----------------------------------------------------------.
| Scanning a Bison character literal, decoding its escapes. |
- | The initial quote is already eaten. |
+ | The initial quote is already eaten. |
`----------------------------------------------------------*/
<SC_ESCAPED_CHARACTER>
/* FIXME: Eventually, make these errors. */
if (last_string[0] == '\0')
{
- warn_at (*loc, _("empty character literal"));
+ complain_at (*loc, Wother, _("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"));
+ complain_at (*loc, Wother,
+ _("extra characters in character literal"));
}
if (yytext[0] == '\n')
unexpected_newline (token_start, "'");
/* FIXME: Eventually, make these errors. */
if (last_string[0] == '\0')
{
- warn_at (*loc, _("empty character literal"));
+ complain_at (*loc, Wother, _("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"));
+ complain_at (*loc, Wother,
+ _("extra characters in character literal"));
}
unexpected_eof (token_start, "'");
STRING_FREE;
\\[0-7]{1,3} {
unsigned long int c = strtoul (yytext + 1, NULL, 8);
if (!c || UCHAR_MAX < c)
- complain_at (*loc, _("invalid number after \\-escape: %s"),
+ complain_at (*loc, complaint, _("invalid number after \\-escape: %s"),
yytext+1);
else
obstack_1grow (&obstack_for_string, c);
verify (UCHAR_MAX < ULONG_MAX);
unsigned long int c = strtoul (yytext + 2, NULL, 16);
if (!c || UCHAR_MAX < c)
- complain_at (*loc, _("invalid number after \\-escape: %s"),
+ complain_at (*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 number after \\-escape: %s"),
+ complain_at (*loc, complaint, _("invalid number after \\-escape: %s"),
yytext+1);
else
obstack_1grow (&obstack_for_string, c);
}
- \\(.|\n) {
+ \\(.|\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))
+ 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_at (*loc, _("invalid character after \\-escape: %s"), p);
+ complain_at (*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, "'"); BEGIN context_state;
+ <<EOF>> unexpected_eof (token_start, "'"); BEGIN context_state;
}
<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;
}
"{"|"<"{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>> {
--nesting;
if (nesting < 0)
{
- STRING_FINISH;
- loc->start = code_start;
- val->code = last_string;
- BEGIN INITIAL;
- return BRACED_CODE;
+ STRING_FINISH;
+ loc->start = code_start;
+ val->code = last_string;
+ BEGIN INITIAL;
+ return BRACED_CODE;
}
}
}
--nesting;
if (nesting < 0)
{
- STRING_FINISH;
- loc->start = code_start;
- val->code = last_string;
- BEGIN INITIAL;
- return BRACED_PREDICATE;
+ STRING_FINISH;
+ loc->start = code_start;
+ val->code = last_string;
+ BEGIN INITIAL;
+ return BRACED_PREDICATE;
}
else
obstack_1grow (&obstack_for_string, '}');
| By default, grow the string obstack with the input. |
`-----------------------------------------------------*/
-<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;
+<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_at (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"'. ARGS has already skipped '#line '. |
`----------------------------------------------------------------*/
static void
*strchr (file, '"') = '\0';
if (INT_MAX <= lineno)
{
- warn_at (loc, _("line number overflow"));
+ complain_at (loc, Wother, _("line number overflow"));
lineno = INT_MAX;
}
current_file = uniqstr_new (file);
/*----------------------------------------------------------------.
| 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
location loc;
loc.start = start;
loc.end = scanner_cursor;
- complain_at (loc, _(msgid), token_end);
+ token_end = quote (token_end);
+ // Instead of '\'', display "'".
+ if (STREQ (token_end, "'\\''"))
+ token_end = "\"'\"";
+ complain_at (loc, complaint, _(msgid), 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);
}
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);
}
projects / bison.git / blobdiff