X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/624a35e20b675af53c8e23805ce79b3b6e3b11ed..b203fc2c07f45cdd77a59f38795fed0109c75716:/src/scan-gram.l diff --git a/src/scan-gram.l b/src/scan-gram.l index da041a4f..7541bf9f 100644 --- a/src/scan-gram.l +++ b/src/scan-gram.l @@ -1,6 +1,6 @@ /* Bison Grammar Scanner -*- C -*- - Copyright (C) 2002 Free Software Foundation, Inc. + Copyright (C) 2002, 2003, 2004 Free Software Foundation, Inc. This file is part of Bison, the GNU Compiler Compiler. @@ -20,7 +20,7 @@ 02111-1307 USA */ -%option debug nodefault noyywrap never-interactive +%option debug nodefault nounput noyywrap never-interactive %option prefix="gram_" outfile="lex.yy.c" %{ @@ -43,6 +43,7 @@ scanner_cursor.file = current_file; \ scanner_cursor.line = 1; \ scanner_cursor.column = 1; \ + code_start = scanner_cursor; \ } \ while (0) @@ -100,8 +101,10 @@ static int rule_length; 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 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 @@ -136,14 +139,16 @@ splice (\\[ \f\t\v]*\n)* int token_type IF_LINT (= 0); /* Location of most recent identifier, when applicable. */ - location id_loc IF_LINT (= *loc); + location id_loc IF_LINT (= empty_location); - /* Where containing code started, when applicable. */ - boundary code_start IF_LINT (= loc->start); + /* Where containing code started, when applicable. Its initial + value is relevant only when yylex is invoked in the SC_EPILOGUE + start condition. */ + boundary code_start = scanner_cursor; /* Where containing comment or string or character literal started, when applicable. */ - boundary token_start IF_LINT (= loc->start); + boundary token_start IF_LINT (= scanner_cursor); %} @@ -153,11 +158,15 @@ splice (\\[ \f\t\v]*\n)* { - [ \f\n\t\v] ; - - /* Comments. */ - "/*" token_start = loc->start; context_state = YY_START; BEGIN SC_YACC_COMMENT; + /* Comments and white space. */ + "," warn_at (*loc, _("stray `,' treated as white space")); + [ \f\n\t\v] | "//".* ; + "/*" { + token_start = loc->start; + context_state = YY_START; + BEGIN SC_YACC_COMMENT; + } /* #line directives are not documented, and may be withdrawn or modified in future versions of Bison. */ @@ -174,22 +183,27 @@ 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; @@ -215,10 +229,6 @@ splice (\\[ \f\t\v]*\n)* "|" rule_length = 0; return PIPE; ";" return SEMICOLON; - "," { - warn_at (*loc, _("stray `,' treated as white space")); - } - {id} { val->symbol = symbol_get (yytext, *loc); id_loc = *loc; @@ -227,15 +237,11 @@ splice (\\[ \f\t\v]*\n)* } {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; } @@ -269,16 +275,18 @@ splice (\\[ \f\t\v]*\n)* "%%" { static int percent_percent_count; if (++percent_percent_count == 2) - { - code_start = loc->start; - BEGIN SC_EPILOGUE; - } + BEGIN SC_EPILOGUE; return PERCENT_PERCENT; } . { complain_at (*loc, _("invalid character: %s"), quote (yytext)); } + + <> { + loc->start = loc->end = scanner_cursor; + yyterminate (); + } } @@ -317,7 +325,7 @@ splice (\\[ \f\t\v]*\n)* { "*/" BEGIN context_state; .|\n ; - <> unexpected_end_of_file (token_start, "*/"); + <> unexpected_eof (token_start, "*/"); BEGIN context_state; } @@ -328,7 +336,7 @@ splice (\\[ \f\t\v]*\n)* { "*"{splice}"/" STRING_GROW; BEGIN context_state; - <> unexpected_end_of_file (token_start, "*/"); + <> unexpected_eof (token_start, "*/"); BEGIN context_state; } @@ -344,10 +352,10 @@ 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. | + `------------------------------------------------*/ { @@ -360,15 +368,14 @@ splice (\\[ \f\t\v]*\n)* 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. | + `----------------------------------------------------------*/ { @@ -386,9 +393,13 @@ splice (\\[ \f\t\v]*\n)* 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")); } @@ -399,19 +410,23 @@ splice (\\[ \f\t\v]*\n)* { \\[0-7]{1,3} { - unsigned long c = strtoul (yytext + 1, 0, 8); + unsigned long int c = strtoul (yytext + 1, 0, 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; + unsigned long int c; set_errno (0); c = strtoul (yytext + 2, 0, 16); if (UCHAR_MAX < c || get_errno ()) 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); } @@ -431,6 +446,8 @@ 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); } @@ -440,30 +457,27 @@ 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; } @@ -529,6 +543,8 @@ splice (\\[ \f\t\v]*\n)* return token_type; } } + + <> unexpected_eof (scanner_cursor, "{}"); BEGIN INITIAL; } @@ -542,9 +558,28 @@ splice (\\[ \f\t\v]*\n)* "{"|"<"{splice}"%" STRING_GROW; braces_level++; "%"{splice}">" STRING_GROW; braces_level--; "}" { - STRING_GROW; - braces_level--; - if (braces_level < 0) + 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) { STRING_FINISH; rule_length++; @@ -562,7 +597,7 @@ 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, "}"); + <> unexpected_eof (code_start, "}"); BEGIN INITIAL; } @@ -580,7 +615,7 @@ splice (\\[ \f\t\v]*\n)* return PROLOGUE; } - <> unexpected_end_of_file (code_start, "%}"); + <> unexpected_eof (code_start, "%}"); BEGIN INITIAL; } @@ -601,10 +636,9 @@ 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. | + `-----------------------------------------*/ { @@ -612,12 +646,23 @@ 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, $-1, etc.) are required by the + semantic actions of this grammar. */ +int max_left_semantic_context = 0; + /* Set *LOC and adjust scanner cursor to account for token TOKEN of size SIZE. */ @@ -662,14 +707,14 @@ adjust_location (location *loc, char const *token, size_t size) static size_t no_cr_read (FILE *fp, char *buf, size_t size) { - size_t s = fread (buf, 1, size, fp); - if (s) + size_t bytes_read = fread (buf, 1, size, fp); + if (bytes_read) { - char *w = memchr (buf, '\r', s); + char *w = memchr (buf, '\r', bytes_read); if (w) { char const *r = ++w; - char const *lim = buf + s; + char const *lim = buf + bytes_read; for (;;) { @@ -698,7 +743,7 @@ no_cr_read (FILE *fp, char *buf, size_t size) } } - return s; + return bytes_read; } @@ -743,13 +788,15 @@ handle_action_dollar (char *text, location loc) } else { - long num; + long int num; set_errno (0); num = strtol (cp, 0, 10); if (INT_MIN <= num && num <= rule_length && ! get_errno ()) { int n = num; + if (1-n > max_left_semantic_context) + max_left_semantic_context = 1-n; if (!type_name && n > 0) type_name = symbol_list_n_type_name_get (current_rule, loc, n); if (!type_name && typed) @@ -769,10 +816,10 @@ handle_action_dollar (char *text, location loc) } -/*-----------------------------------------------------------------. -| Dispatch onto handle_action_dollar, or handle_destructor_dollar, | -| depending upon TOKEN_TYPE. | -`-----------------------------------------------------------------*/ +/*----------------------------------------------------------------. +| Map `$?' 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) @@ -785,6 +832,7 @@ handle_dollar (int token_type, char *text, location loc) break; case PERCENT_DESTRUCTOR: + case PERCENT_INITIAL_ACTION: case PERCENT_PRINTER: if (text[1] == '$') { @@ -810,7 +858,7 @@ static inline bool handle_action_at (char *text, location loc) { char *cp = text + 1; - locations_flag = 1; + locations_flag = true; if (! current_rule) return false; @@ -819,7 +867,7 @@ handle_action_at (char *text, location loc) obstack_sgrow (&obstack_for_string, "]b4_lhs_location["); else { - long num; + long int num; set_errno (0); num = strtol (cp, 0, 10); @@ -837,10 +885,10 @@ handle_action_at (char *text, location loc) } -/*-------------------------------------------------------------------. -| Dispatch onto handle_action_at, or handle_destructor_at, depending | -| upon CODE_KIND. | -`-------------------------------------------------------------------*/ +/*----------------------------------------------------------------. +| Map `@?' onto the proper M4 symbol, depending on its TOKEN_TYPE | +| (are we in an action?). | +`----------------------------------------------------------------*/ static void handle_at (int token_type, char *text, location loc) @@ -851,6 +899,7 @@ handle_at (int token_type, char *text, location loc) handle_action_at (text, loc); return; + case PERCENT_INITIAL_ACTION: case PERCENT_DESTRUCTOR: case PERCENT_PRINTER: if (text[1] == '$') @@ -868,6 +917,25 @@ 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) +{ + unsigned long int num; + set_errno (0); + num = strtoul (number, 0, base); + if (INT_MAX < num || get_errno ()) + { + 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 | @@ -877,7 +945,7 @@ 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); + unsigned long int code = strtoul (ucn + 2, 0, 16); /* FIXME: Currently we assume Unicode-compatible unibyte characters on ASCII hosts (i.e., Latin-1 on hosts with 8-bit bytes). On @@ -935,34 +1003,48 @@ handle_syncline (char *args) const char *file = NULL; file = strchr (args, '"') + 1; *strchr (file, '"') = 0; - scanner_cursor.file = current_file = xstrdup (file); + scanner_cursor.file = current_file = uniqstr_new (file); scanner_cursor.line = lineno; scanner_cursor.column = 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. | +`----------------------------------------------------------------*/ + +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); }