X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/379f0ac840fa8b04eacbfdd7d19394f86574031e..276f48df88293dc0a67237d1124a26e0f9ee82fb:/src/scan-gram.l diff --git a/src/scan-gram.l b/src/scan-gram.l index 77fe5beb..cf704c77 100644 --- a/src/scan-gram.l +++ b/src/scan-gram.l @@ -1,6 +1,6 @@ /* Bison Grammar Scanner -*- C -*- - Copyright (C) 2002, 2003 Free Software Foundation, Inc. + Copyright (C) 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc. This file is part of Bison, the GNU Compiler Compiler. @@ -16,24 +16,29 @@ 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., 59 Temple Place, Suite 330, Boston, MA - 02111-1307 USA + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + 02110-1301 USA */ -%option debug nodefault noyywrap never-interactive +%option debug nodefault nounput noyywrap never-interactive %option prefix="gram_" outfile="lex.yy.c" %{ +/* Work around a bug in flex 2.5.31. See Debian bug 333231 + . */ +#undef gram_wrap +#define gram_wrap() 1 + #include "system.h" #include -#include #include #include "complain.h" #include "files.h" #include "getargs.h" #include "gram.h" +#include "quotearg.h" #include "reader.h" #include "uniqstr.h" @@ -47,6 +52,19 @@ } \ while (0) +/* Pacify "gcc -Wmissing-prototypes" when flex 2.5.31 is used. */ +int gram_get_lineno (void); +FILE *gram_get_in (void); +FILE *gram_get_out (void); +int gram_get_leng (void); +char *gram_get_text (void); +void gram_set_lineno (int); +void gram_set_in (FILE *); +void gram_set_out (FILE *); +int gram_get_debug (void); +void gram_set_debug (int); +int gram_lex_destroy (void); + /* Location of scanner cursor. */ boundary scanner_cursor; @@ -69,8 +87,11 @@ static size_t no_cr_read (FILE *, char *, size_t); static struct obstack obstack_for_string; /* A string representing the most recently saved token. */ -static char *last_string; +char *last_string; +/* The location of the most recently saved token, if it was a + BRACED_CODE token; otherwise, this has an unspecified value. */ +location last_braced_code_loc; #define STRING_GROW \ obstack_grow (&obstack_for_string, yytext, yyleng) @@ -98,11 +119,27 @@ scanner_last_string_free (void) Outside of well-formed rules, RULE_LENGTH has an undefined value. */ static int rule_length; +static void rule_length_overflow (location) __attribute__ ((__noreturn__)); + +/* Increment the rule length by one, checking for overflow. */ +static inline void +increment_rule_length (location loc) +{ + rule_length++; + + /* Don't allow rule_length == INT_MAX, since that might cause + confusion with strtol if INT_MAX == LONG_MAX. */ + if (rule_length == INT_MAX) + rule_length_overflow (loc); +} + static void handle_dollar (int token_type, char *cp, location loc); static void handle_at (int token_type, char *cp, location loc); -static void handle_syncline (char *args); +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); -static void unexpected_end_of_file (boundary, char const *); +static void unexpected_eof (boundary, char const *); +static void unexpected_newline (boundary, char const *); %} %x SC_COMMENT SC_LINE_COMMENT SC_YACC_COMMENT @@ -156,10 +193,9 @@ splice (\\[ \f\t\v]*\n)* { - [ \f\n\t\v] ; + /* Comments and white space. */ "," warn_at (*loc, _("stray `,' treated as white space")); - - /* Comments. */ + [ \f\n\t\v] | "//".* ; "/*" { token_start = loc->start; @@ -170,7 +206,7 @@ splice (\\[ \f\t\v]*\n)* /* #line directives are not documented, and may be withdrawn or modified in future versions of Bison. */ ^"#line "{int}" \"".*"\"\n" { - handle_syncline (yytext + sizeof "#line " - 1); + handle_syncline (yytext + sizeof "#line " - 1, *loc); } } @@ -182,28 +218,34 @@ splice (\\[ \f\t\v]*\n)* { "%binary" return PERCENT_NONASSOC; "%debug" return PERCENT_DEBUG; + "%default"[-_]"prec" return PERCENT_DEFAULT_PREC; "%define" return PERCENT_DEFINE; "%defines" return PERCENT_DEFINES; "%destructor" token_type = PERCENT_DESTRUCTOR; BEGIN SC_PRE_CODE; "%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" token_type = PERCENT_INITIAL_ACTION; BEGIN SC_PRE_CODE; "%glr-parser" return PERCENT_GLR_PARSER; "%left" return PERCENT_LEFT; "%lex-param" token_type = PERCENT_LEX_PARAM; BEGIN SC_PRE_CODE; "%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" token_type = PERCENT_PARSE_PARAM; BEGIN SC_PRE_CODE; "%prec" rule_length--; return PERCENT_PREC; "%printer" token_type = PERCENT_PRINTER; BEGIN SC_PRE_CODE; "%pure"[-_]"parser" return PERCENT_PURE_PARSER; + "%require" return PERCENT_REQUIRE; "%right" return PERCENT_RIGHT; "%skeleton" return PERCENT_SKELETON; "%start" return PERCENT_START; @@ -226,20 +268,16 @@ splice (\\[ \f\t\v]*\n)* {id} { val->symbol = symbol_get (yytext, *loc); id_loc = *loc; - rule_length++; + increment_rule_length (*loc); BEGIN SC_AFTER_IDENTIFIER; } {int} { - unsigned long num; - set_errno (0); - num = strtoul (yytext, 0, 10); - if (INT_MAX < num || get_errno ()) - { - complain_at (*loc, _("integer out of range: %s"), quote (yytext)); - num = INT_MAX; - } - val->integer = num; + val->integer = scan_integer (yytext, 10, *loc); + return INT; + } + 0[xX][0-9abcdefABCDEF]+ { + val->integer = scan_integer (yytext, 16, *loc); return INT; } @@ -247,13 +285,15 @@ splice (\\[ \f\t\v]*\n)* "'" STRING_GROW; token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER; /* Strings. */ - "\"" STRING_GROW; 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; @@ -323,7 +363,7 @@ splice (\\[ \f\t\v]*\n)* { "*/" BEGIN context_state; .|\n ; - <> unexpected_end_of_file (token_start, "*/"); + <> unexpected_eof (token_start, "*/"); BEGIN context_state; } @@ -334,7 +374,7 @@ splice (\\[ \f\t\v]*\n)* { "*"{splice}"/" STRING_GROW; BEGIN context_state; - <> unexpected_end_of_file (token_start, "*/"); + <> unexpected_eof (token_start, "*/"); BEGIN context_state; } @@ -350,31 +390,29 @@ splice (\\[ \f\t\v]*\n)* } - /*----------------------------------------------------------------. - | Scanning a C string, including its escapes. The initial `"' is | - | already eaten. | - `----------------------------------------------------------------*/ + /*------------------------------------------------. + | Scanning a Bison string, including its escapes. | + | The initial quote is already eaten. | + `------------------------------------------------*/ { "\"" { - STRING_GROW; STRING_FINISH; loc->start = token_start; val->chars = last_string; - rule_length++; + increment_rule_length (*loc); BEGIN INITIAL; return STRING; } - - .|\n STRING_GROW; - <> unexpected_end_of_file (token_start, "\""); + \n unexpected_newline (token_start, "\""); BEGIN INITIAL; + <> unexpected_eof (token_start, "\""); BEGIN INITIAL; } - /*---------------------------------------------------------------. - | Scanning a C character, decoding its escapes. The initial "'" | - | is already eaten. | - `---------------------------------------------------------------*/ + /*----------------------------------------------------------. + | Scanning a Bison character literal, decoding its escapes. | + | The initial quote is already eaten. | + `----------------------------------------------------------*/ { @@ -383,18 +421,24 @@ splice (\\[ \f\t\v]*\n)* STRING_GROW; STRING_FINISH; loc->start = token_start; - val->symbol = symbol_get (last_string, *loc); - symbol_class_set (val->symbol, token_sym, *loc); + 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); STRING_FREE; - rule_length++; + increment_rule_length (*loc); BEGIN INITIAL; return ID; } + \n unexpected_newline (token_start, "'"); BEGIN INITIAL; + <> unexpected_eof (token_start, "'"); BEGIN INITIAL; +} - .|\n STRING_GROW; - <> unexpected_end_of_file (token_start, "'"); + +{ + \0 complain_at (*loc, _("invalid null character")); } @@ -405,19 +449,22 @@ splice (\\[ \f\t\v]*\n)* { \\[0-7]{1,3} { - unsigned long c = strtoul (yytext + 1, 0, 8); + 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)); else obstack_1grow (&obstack_for_string, c); } \\x[0-9abcdefABCDEF]+ { - unsigned long c; - set_errno (0); - c = strtoul (yytext + 2, 0, 16); - if (UCHAR_MAX < c || get_errno ()) + 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)); else obstack_1grow (&obstack_for_string, c); } @@ -437,6 +484,8 @@ splice (\\[ \f\t\v]*\n)* 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)); else obstack_1grow (&obstack_for_string, c); } @@ -446,30 +495,27 @@ splice (\\[ \f\t\v]*\n)* } } + /*--------------------------------------------. + | Scanning user-code characters and strings. | + `--------------------------------------------*/ - /*----------------------------------------------------------. - | Scanning a C character without decoding its escapes. The | - | initial "'" is already eaten. | - `----------------------------------------------------------*/ + +{ + {splice}|\\{splice}[^\n$@\[\]] STRING_GROW; +} { - "'" STRING_GROW; BEGIN context_state; - \\{splice}[^$@\[\]] STRING_GROW; - <> unexpected_end_of_file (token_start, "'"); + "'" STRING_GROW; BEGIN context_state; + \n unexpected_newline (token_start, "'"); BEGIN context_state; + <> unexpected_eof (token_start, "'"); BEGIN context_state; } - - /*----------------------------------------------------------------. - | Scanning a C string, without decoding its escapes. The initial | - | `"' is already eaten. | - `----------------------------------------------------------------*/ - { - "\"" STRING_GROW; BEGIN context_state; - \\{splice}[^$@\[\]] STRING_GROW; - <> unexpected_end_of_file (token_start, "\""); + "\"" STRING_GROW; BEGIN context_state; + \n unexpected_newline (token_start, "\""); BEGIN context_state; + <> unexpected_eof (token_start, "\""); BEGIN context_state; } @@ -526,7 +572,7 @@ splice (\\[ \f\t\v]*\n)* } else { - complain_at (*loc, _("missing `{' in `%s'"), + complain_at (*loc, _("missing `{' in %s"), token_name (token_type)); obstack_sgrow (&obstack_for_string, "{}"); STRING_FINISH; @@ -536,7 +582,7 @@ splice (\\[ \f\t\v]*\n)* } } - <> unexpected_end_of_file (scanner_cursor, "{}"); + <> unexpected_eof (scanner_cursor, "{}"); BEGIN INITIAL; } @@ -574,9 +620,10 @@ splice (\\[ \f\t\v]*\n)* if (outer_brace) { STRING_FINISH; - rule_length++; loc->start = code_start; val->chars = last_string; + increment_rule_length (*loc); + last_braced_code_loc = *loc; BEGIN INITIAL; return token_type; } @@ -589,7 +636,16 @@ splice (\\[ \f\t\v]*\n)* "$"("<"{tag}">")?(-?[0-9]+|"$") handle_dollar (token_type, yytext, *loc); "@"(-?[0-9]+|"$") handle_at (token_type, yytext, *loc); - <> unexpected_end_of_file (code_start, "}"); + "$" { + warn_at (*loc, _("stray `$'")); + obstack_sgrow (&obstack_for_string, "$]["); + } + "@" { + warn_at (*loc, _("stray `@'")); + obstack_sgrow (&obstack_for_string, "@@"); + } + + <> unexpected_eof (code_start, "}"); BEGIN INITIAL; } @@ -607,7 +663,7 @@ splice (\\[ \f\t\v]*\n)* return PROLOGUE; } - <> unexpected_end_of_file (code_start, "%}"); + <> unexpected_eof (code_start, "%}"); BEGIN INITIAL; } @@ -628,10 +684,9 @@ splice (\\[ \f\t\v]*\n)* } - /*----------------------------------------------------------------. - | By default, grow the string obstack with the input, escaping M4 | - | quoting characters. | - `----------------------------------------------------------------*/ + /*-----------------------------------------. + | Escape M4 quoting characters in C code. | + `-----------------------------------------*/ { @@ -639,12 +694,46 @@ splice (\\[ \f\t\v]*\n)* \@ obstack_sgrow (&obstack_for_string, "@@"); \[ obstack_sgrow (&obstack_for_string, "@{"); \] obstack_sgrow (&obstack_for_string, "@}"); - .|\n STRING_GROW; } + /*-----------------------------------------------------. + | By default, grow the string obstack with the input. | + `-----------------------------------------------------*/ + +. | +\n STRING_GROW; + %% +/* Keeps track of the maximum number of semantic values to the left of + a handle (those referenced by $0, $-1, etc.) are required by the + semantic actions of this grammar. */ +int max_left_semantic_context = 0; + +/* If BUF is null, add BUFSIZE (which in this case must be less than + INT_MAX) to COLUMN; otherwise, add mbsnwidth (BUF, BUFSIZE, 0) to + COLUMN. If an overflow occurs, or might occur but is undetectable, + return INT_MAX. Assume COLUMN is nonnegative. */ + +static inline int +add_column_width (int column, char const *buf, size_t bufsize) +{ + size_t width; + unsigned int remaining_columns = INT_MAX - column; + + if (buf) + { + if (INT_MAX / 2 <= bufsize) + return INT_MAX; + width = mbsnwidth (buf, bufsize, 0); + } + else + width = bufsize; + + return width <= remaining_columns ? column + width : INT_MAX; +} + /* Set *LOC and adjust scanner cursor to account for token TOKEN of size SIZE. */ @@ -663,22 +752,30 @@ adjust_location (location *loc, char const *token, size_t size) switch (*p) { case '\n': - line++; + line += line < INT_MAX; column = 1; p0 = p + 1; break; case '\t': - column += mbsnwidth (p0, p - p0, 0); - column += 8 - ((column - 1) & 7); + column = add_column_width (column, p0, p - p0); + column = add_column_width (column, NULL, 8 - ((column - 1) & 7)); p0 = p + 1; break; + + default: + break; } scanner_cursor.line = line; - scanner_cursor.column = column + mbsnwidth (p0, p - p0, 0); + scanner_cursor.column = column = add_column_width (column, p0, p - p0); loc->end = scanner_cursor; + + if (line == INT_MAX && loc->start.line != INT_MAX) + warn_at (*loc, _("line number overflow")); + if (column == INT_MAX && loc->start.column != INT_MAX) + warn_at (*loc, _("column number overflow")); } @@ -767,17 +864,18 @@ handle_action_dollar (char *text, location loc) type_name = ""; obstack_fgrow1 (&obstack_for_string, "]b4_lhs_value([%s])[", type_name); + current_rule->used = true; } else { - long num; - set_errno (0); - num = strtol (cp, 0, 10); + long int num = strtol (cp, NULL, 10); - if (INT_MIN <= num && num <= rule_length && ! get_errno ()) + if (1 - INT_MAX + rule_length <= num && num <= rule_length) { int n = num; - if (!type_name && n > 0) + if (max_left_semantic_context < 1 - n) + max_left_semantic_context = 1 - n; + if (!type_name && 0 < n) type_name = symbol_list_n_type_name_get (current_rule, loc, n); if (!type_name && typed) complain_at (loc, _("$%d of `%s' has no declared type"), @@ -785,8 +883,9 @@ handle_action_dollar (char *text, location loc) if (!type_name) type_name = ""; obstack_fgrow3 (&obstack_for_string, - "]b4_rhs_value([%d], [%d], [%s])[", + "]b4_rhs_value(%d, %d, [%s])[", rule_length, n, type_name); + symbol_list_n_used_set (current_rule, n, true); } else complain_at (loc, _("integer out of range: %s"), quote (text)); @@ -796,10 +895,10 @@ handle_action_dollar (char *text, location loc) } -/*-----------------------------------------------------------------. -| Dispatch onto handle_action_dollar, or handle_destructor_dollar, | -| depending upon TOKEN_TYPE. | -`-----------------------------------------------------------------*/ +/*----------------------------------------------------------------. +| Map `$?' onto the proper M4 symbol, depending on its TOKEN_TYPE | +| (are we in an action?). | +`----------------------------------------------------------------*/ static void handle_dollar (int token_type, char *text, location loc) @@ -812,6 +911,7 @@ handle_dollar (int token_type, char *text, location loc) break; case PERCENT_DESTRUCTOR: + case PERCENT_INITIAL_ACTION: case PERCENT_PRINTER: if (text[1] == '$') { @@ -837,7 +937,7 @@ static inline bool handle_action_at (char *text, location loc) { char *cp = text + 1; - locations_flag = 1; + locations_flag = true; if (! current_rule) return false; @@ -846,14 +946,12 @@ handle_action_at (char *text, location loc) obstack_sgrow (&obstack_for_string, "]b4_lhs_location["); else { - long num; - set_errno (0); - num = strtol (cp, 0, 10); + long int num = strtol (cp, NULL, 10); - if (INT_MIN <= num && num <= rule_length && ! get_errno ()) + if (1 - INT_MAX + rule_length <= num && num <= rule_length) { int n = num; - obstack_fgrow2 (&obstack_for_string, "]b4_rhs_location([%d], [%d])[", + obstack_fgrow2 (&obstack_for_string, "]b4_rhs_location(%d, %d)[", rule_length, n); } else @@ -864,10 +962,10 @@ handle_action_at (char *text, location loc) } -/*-------------------------------------------------------------------. -| Dispatch onto handle_action_at, or handle_destructor_at, depending | -| upon CODE_KIND. | -`-------------------------------------------------------------------*/ +/*----------------------------------------------------------------. +| Map `@?' onto the proper M4 symbol, depending on its TOKEN_TYPE | +| (are we in an action?). | +`----------------------------------------------------------------*/ static void handle_at (int token_type, char *text, location loc) @@ -878,6 +976,7 @@ handle_at (int token_type, char *text, location loc) handle_action_at (text, loc); return; + case PERCENT_INITIAL_ACTION: case PERCENT_DESTRUCTOR: case PERCENT_PRINTER: if (text[1] == '$') @@ -895,6 +994,26 @@ handle_at (int token_type, char *text, location loc) } +/*------------------------------------------------------. +| Scan NUMBER for a base-BASE integer at location LOC. | +`------------------------------------------------------*/ + +static unsigned long int +scan_integer (char const *number, int base, location loc) +{ + verify (INT_MAX < ULONG_MAX); + unsigned long int num = strtoul (number, NULL, base); + + if (INT_MAX < num) + { + complain_at (loc, _("integer out of range: %s"), quote (number)); + num = INT_MAX; + } + + return num; +} + + /*------------------------------------------------------------------. | Convert universal character name UCN to a single-byte character, | | and return that character. Return -1 if UCN does not correspond | @@ -904,7 +1023,8 @@ handle_at (int token_type, char *text, location loc) static int convert_ucn_to_byte (char const *ucn) { - unsigned long code = strtoul (ucn + 2, 0, 16); + verify (UCHAR_MAX <= INT_MAX); + unsigned long int code = strtoul (ucn + 2, NULL, 16); /* FIXME: Currently we assume Unicode-compatible unibyte characters on ASCII hosts (i.e., Latin-1 on hosts with 8-bit bytes). On @@ -956,40 +1076,70 @@ convert_ucn_to_byte (char const *ucn) `----------------------------------------------------------------*/ static void -handle_syncline (char *args) +handle_syncline (char *args, location loc) { - int lineno = strtol (args, &args, 10); - const char *file = NULL; - file = strchr (args, '"') + 1; - *strchr (file, '"') = 0; - scanner_cursor.file = current_file = xstrdup (file); + char *after_num; + unsigned long int lineno = strtoul (args, &after_num, 10); + char *file = strchr (after_num, '"') + 1; + *strchr (file, '"') = '\0'; + if (INT_MAX <= lineno) + { + warn_at (loc, _("line number overflow")); + lineno = INT_MAX; + } + scanner_cursor.file = current_file = uniqstr_new (file); scanner_cursor.line = lineno; scanner_cursor.column = 1; } +/*---------------------------------. +| Report a rule that is too long. | +`---------------------------------*/ + +static void +rule_length_overflow (location loc) +{ + fatal_at (loc, _("rule is too long")); +} + + +/*----------------------------------------------------------------. +| 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. | +`----------------------------------------------------------------*/ + +static void +unexpected_end (boundary start, char const *msgid, char const *token_end) +{ + location loc; + loc.start = start; + loc.end = scanner_cursor; + complain_at (loc, _(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_end_of_file (boundary start, char const *token_end) +unexpected_eof (boundary start, char const *token_end) { - size_t i = strlen (token_end); + unexpected_end (start, N_("missing `%s' at end of file"), token_end); +} - location loc; - loc.start = start; - loc.end = scanner_cursor; - complain_at (loc, _("missing `%s' at end of file"), token_end); - /* Adjust scanner cursor so that any later message does not count - the characters about to be inserted. */ - scanner_cursor.column -= i; +/*----------------------------------------. +| Likewise, but for unexpected newlines. | +`----------------------------------------*/ - while (i != 0) - unput (token_end[--i]); +static void +unexpected_newline (boundary start, char const *token_end) +{ + unexpected_end (start, N_("missing `%s' at end of line"), token_end); }