X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/02650b7f0e3ef470a949768bdb907183da4b3beb..e34e97bcd736ab3193bfd3abd4162db3b159836d:/src/scan-gram.l?ds=inline diff --git a/src/scan-gram.l b/src/scan-gram.l index 5cf5a28c..75023f4f 100644 --- a/src/scan-gram.l +++ b/src/scan-gram.l @@ -1,12 +1,12 @@ /* Bison Grammar Scanner -*- C -*- - Copyright (C) 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc. + Copyright (C) 2002-2012 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, @@ -15,12 +15,9 @@ 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 . */ -%option debug nodefault nounput noyywrap never-interactive +%option debug nodefault noinput nounput noyywrap never-interactive %option prefix="gram_" outfile="lex.yy.c" %{ @@ -29,109 +26,56 @@ #undef gram_wrap #define gram_wrap() 1 -#include "system.h" - -#include -#include +#define FLEX_PREFIX(Id) gram_ ## Id +#include "flex-scanner.h" #include "complain.h" #include "files.h" -#include "getargs.h" #include "gram.h" #include "quotearg.h" #include "reader.h" #include "uniqstr.h" +#include +#include +#include + +#include "scan-gram.h" + +#define YY_DECL GRAM_LEX_DECL + #define YY_USER_INIT \ - do \ - { \ - scanner_cursor.file = current_file; \ - scanner_cursor.line = 1; \ - scanner_cursor.column = 1; \ - code_start = scanner_cursor; \ - } \ - 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); + code_start = scanner_cursor = loc->start; \ /* Location of scanner cursor. */ -boundary scanner_cursor; +static boundary scanner_cursor; -static void adjust_location (location *, char const *, size_t); -#define YY_USER_ACTION adjust_location (loc, yytext, yyleng); +#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)) - -/* OBSTACK_FOR_STRING -- Used to store all the characters that we need to - keep (to construct ID, STRINGS etc.). Use the following macros to - use it. - - Use STRING_GROW to append what has just been matched, and - STRING_FINISH to end the string (it puts the ending 0). - STRING_FINISH also stores this string in LAST_STRING, which can be - used, and which is used by STRING_FREE to free the last string. */ - -static struct obstack obstack_for_string; +#define ROLLBACK_CURRENT_TOKEN \ + do { \ + scanner_cursor.column -= mbsnwidth (yytext, yyleng, 0); \ + yyless (0); \ + } while (0) /* A string representing the most recently saved token. */ static char *last_string; - -#define STRING_GROW \ - obstack_grow (&obstack_for_string, yytext, yyleng) - -#define STRING_FINISH \ - do { \ - obstack_1grow (&obstack_for_string, '\0'); \ - last_string = obstack_finish (&obstack_for_string); \ - } while (0) - -#define STRING_FREE \ - obstack_free (&obstack_for_string, 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 -scanner_last_string_free (void) +gram_scanner_last_string_free (void) { STRING_FREE; } -/* Within well-formed rules, RULE_LENGTH is the number of values in - the current rule so far, which says where to find `$0' with respect - to the top of the stack. It is not the same as the rule->length in - the case of mid rule actions. - - 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 *, location); static unsigned long int scan_integer (char const *p, int base, location loc); static int convert_ucn_to_byte (char const *hex_text); @@ -139,15 +83,29 @@ static void unexpected_eof (boundary, char const *); static void unexpected_newline (boundary, char const *); %} -%x SC_COMMENT SC_LINE_COMMENT SC_YACC_COMMENT -%x SC_STRING SC_CHARACTER -%x SC_AFTER_IDENTIFIER + /* A C-like comment in directives/rules. */ +%x SC_YACC_COMMENT + /* Strings and characters in directives/rules. */ %x SC_ESCAPED_STRING SC_ESCAPED_CHARACTER -%x SC_PRE_CODE SC_BRACED_CODE SC_PROLOGUE SC_EPILOGUE + /* A identifier was just read in directives/rules. Special state + to capture the sequence 'identifier :'. */ +%x SC_AFTER_IDENTIFIER + + /* Three 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 + /* 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]|-)* +id {letter}({letter}|[-0-9])* +directive %{id} int [0-9]+ /* POSIX says that a tag must be both an id and a C union member, but @@ -162,16 +120,13 @@ splice (\\[ \f\t\v]*\n)* %% %{ /* Nesting level of the current code in braces. */ - int braces_level IF_LINT (= 0); + int braces_level 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 @@ -180,7 +135,7 @@ splice (\\[ \f\t\v]*\n)* /* 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); %} @@ -188,10 +143,10 @@ splice (\\[ \f\t\v]*\n)* | Scanning white space. | `-----------------------*/ - + { /* Comments and white space. */ - "," warn_at (*loc, _("stray `,' treated as white space")); + "," warn_at (*loc, _("stray ',' treated as white space")); [ \f\n\t\v] | "//".* ; "/*" { @@ -211,61 +166,72 @@ splice (\\[ \f\t\v]*\n)* /*----------------------------. | 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. */ + { - "%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; + "%binary" return PERCENT_NONASSOC; + "%code" return PERCENT_CODE; + "%debug" return PERCENT_DEBUG; + "%default"[-_]"prec" return PERCENT_DEFAULT_PREC; + "%define" return PERCENT_DEFINE; + "%defines" return PERCENT_DEFINES; + "%destructor" 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" 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; - "%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; + "%initial-action" return PERCENT_INITIAL_ACTION; + "%glr-parser" return PERCENT_GLR_PARSER; + "%language" return PERCENT_LANGUAGE; + "%left" return PERCENT_LEFT; + "%lex-param" 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" return PERCENT_PARSE_PARAM; + "%prec" return PERCENT_PREC; + "%printer" 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" return PERCENT_UNION; + "%verbose" return PERCENT_VERBOSE; + "%yacc" return PERCENT_YACC; {directive} { complain_at (*loc, _("invalid directive: %s"), quote (yytext)); } "=" return EQUAL; - "|" rule_length = 0; return PIPE; + "|" return PIPE; ";" return SEMICOLON; + "<*>" return TYPE_TAG_ANY; + "<>" return TYPE_TAG_NONE; {id} { - val->symbol = symbol_get (yytext, *loc); + val->uniqstr = uniqstr_new (yytext); id_loc = *loc; - increment_rule_length (*loc); + bracketed_id_str = NULL; BEGIN SC_AFTER_IDENTIFIER; } @@ -278,8 +244,14 @@ splice (\\[ \f\t\v]*\n)* 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_at (*loc, _("invalid identifier: %s"), quote (yytext)); + } + + /* Characters. */ + "'" token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER; /* Strings. */ "\"" token_start = loc->start; BEGIN SC_ESCAPED_STRING; @@ -290,7 +262,6 @@ splice (\\[ \f\t\v]*\n)* /* Code in between braces. */ "{" { STRING_GROW; - token_type = BRACED_CODE; braces_level = 0; code_start = loc->start; BEGIN SC_BRACED_CODE; @@ -312,6 +283,13 @@ splice (\\[ \f\t\v]*\n)* return PERCENT_PERCENT; } + "[" { + bracketed_id_str = NULL; + bracketed_id_start = loc->start; + bracketed_id_context_state = YY_START; + BEGIN SC_BRACKETED_ID; + } + . { complain_at (*loc, _("invalid character: %s"), quote (yytext)); } @@ -329,29 +307,97 @@ splice (\\[ \f\t\v]*\n)* { + "[" { + 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; + } + } ":" { - rule_length = 0; + 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; } <> { + BEGIN (bracketed_id_str ? SC_RETURN_BRACKETED_ID : INITIAL); *loc = id_loc; - BEGIN INITIAL; return ID; } } + /*--------------------------------. + | Scanning bracketed identifiers. | + `--------------------------------*/ + + +{ + {id} { + if (bracketed_id_str) + { + complain_at (*loc, _("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_at (*loc, _("an identifier expected")); + } + . { + complain_at (*loc, _("invalid character in bracketed name: %s"), + quote (yytext)); + } + <> { + BEGIN bracketed_id_context_state; + unexpected_eof (bracketed_id_start, "]"); + } +} + + +{ + . { + 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. | `---------------------------------------------------------------*/ @@ -363,7 +409,7 @@ splice (\\[ \f\t\v]*\n)* /*------------------------------------------------------------. - | Scanning a C comment. The initial `/ *' is already eaten. | + | Scanning a C comment. The initial '/ *' is already eaten. | `------------------------------------------------------------*/ @@ -374,7 +420,7 @@ splice (\\[ \f\t\v]*\n)* /*--------------------------------------------------------------. - | Scanning a line comment. The initial `//' is already eaten. | + | Scanning a line comment. The initial '//' is already eaten. | `--------------------------------------------------------------*/ @@ -392,16 +438,23 @@ splice (\\[ \f\t\v]*\n)* { - "\"" { + "\""|"\n" { + if (yytext[0] == '\n') + unexpected_newline (token_start, "\""); + STRING_FINISH; + loc->start = token_start; + val->chars = last_string; + BEGIN INITIAL; + return STRING; + } + <> { + unexpected_eof (token_start, "\""); STRING_FINISH; loc->start = token_start; val->chars = last_string; - increment_rule_length (*loc); BEGIN INITIAL; return STRING; } - \n unexpected_newline (token_start, "\""); BEGIN INITIAL; - <> unexpected_eof (token_start, "\""); BEGIN INITIAL; } /*----------------------------------------------------------. @@ -411,24 +464,47 @@ splice (\\[ \f\t\v]*\n)* { - "'" { - unsigned char last_string_1; - STRING_GROW; + "'"|"\n" { 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') + { + warn_at (*loc, _("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")); + } + if (yytext[0] == '\n') + unexpected_newline (token_start, "'"); STRING_FREE; - increment_rule_length (*loc); BEGIN INITIAL; - return ID; + return CHAR; + } + <> { + STRING_FINISH; + loc->start = token_start; + val->character = last_string[0]; + { + /* FIXME: Eventually, make these errors. */ + if (last_string[0] == '\0') + { + warn_at (*loc, _("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")); + } + unexpected_eof (token_start, "'"); + STRING_FREE; + BEGIN INITIAL; + return CHAR; } - \n unexpected_newline (token_start, "'"); BEGIN INITIAL; - <> unexpected_eof (token_start, "'"); BEGIN INITIAL; } @@ -445,10 +521,9 @@ splice (\\[ \f\t\v]*\n)* { \\[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_at (*loc, _("invalid number after \\-escape: %s"), + yytext+1); else obstack_1grow (&obstack_for_string, c); } @@ -456,10 +531,9 @@ splice (\\[ \f\t\v]*\n)* \\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_at (*loc, _("invalid number after \\-escape: %s"), + yytext+1); else obstack_1grow (&obstack_for_string, c); } @@ -477,16 +551,20 @@ splice (\\[ \f\t\v]*\n)* \\(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_at (*loc, _("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; + char const *p = yytext + 1; + /* Quote only if escaping won't make the character visible. */ + if (isspace ((unsigned char) *p) && 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); } } @@ -496,7 +574,7 @@ splice (\\[ \f\t\v]*\n)* { - {splice}|\\{splice}[^\n$@\[\]] STRING_GROW; + {splice}|\\{splice}[^\n\[\]] STRING_GROW; } @@ -546,100 +624,42 @@ splice (\\[ \f\t\v]*\n)* } - /*---------------------------------------------------------------. - | 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. | - `---------------------------------------------------------------*/ - - -{ - . { - bool valid = yytext[0] == '{' || token_type == PERCENT_UNION; - scanner_cursor.column -= mbsnwidth (yytext, yyleng, 0); - yyless (0); - - 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; - } - } - - <> unexpected_eof (scanner_cursor, "{}"); BEGIN INITIAL; -} - - /*---------------------------------------------------------------. - | Scanning some code in braces (%union and actions). The initial | - | "{" is already eaten. | - `---------------------------------------------------------------*/ + /*-----------------------------------------------------------. + | Scanning some code in braces (actions). The initial "{" is | + | already eaten. | + `-----------------------------------------------------------*/ { "{"|"<"{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) + --braces_level; + if (braces_level < 0) { STRING_FINISH; loc->start = code_start; - val->chars = last_string; - increment_rule_length (*loc); + val->code = last_string; BEGIN INITIAL; - return token_type; + return BRACED_CODE; } } - /* Tokenize `<<%' correctly (as `<<' `%') rather than incorrrectly - (as `<' `<%'). */ + /* Tokenize '<<%' correctly (as '<<' '%') rather than incorrrectly + (as '<' '<%'). */ "<"{splice}"<" STRING_GROW; - "$"("<"{tag}">")?(-?[0-9]+|"$") handle_dollar (token_type, yytext, *loc); - "@"(-?[0-9]+|"$") handle_at (token_type, yytext, *loc); - - "$" { - warn_at (*loc, _("stray `$'")); - obstack_sgrow (&obstack_for_string, "$]["); - } - "@" { - warn_at (*loc, _("stray `@'")); - obstack_sgrow (&obstack_for_string, "@@"); + <> { + unexpected_eof (code_start, "}"); + STRING_FINISH; + loc->start = code_start; + val->code = last_string; + BEGIN INITIAL; + return BRACED_CODE; } - - <> unexpected_eof (code_start, "}"); BEGIN INITIAL; } @@ -657,7 +677,14 @@ splice (\\[ \f\t\v]*\n)* return PROLOGUE; } - <> unexpected_eof (code_start, "%}"); BEGIN INITIAL; + <> { + unexpected_eof (code_start, "%}"); + STRING_FINISH; + loc->start = code_start; + val->chars = last_string; + BEGIN INITIAL; + return PROLOGUE; + } } @@ -678,19 +705,6 @@ splice (\\[ \f\t\v]*\n)* } - /*-----------------------------------------. - | Escape M4 quoting characters in C code. | - `-----------------------------------------*/ - - -{ - \$ obstack_sgrow (&obstack_for_string, "$]["); - \@ obstack_sgrow (&obstack_for_string, "@@"); - \[ obstack_sgrow (&obstack_for_string, "@{"); - \] obstack_sgrow (&obstack_for_string, "@}"); -} - - /*-----------------------------------------------------. | By default, grow the string obstack with the input. | `-----------------------------------------------------*/ @@ -700,78 +714,6 @@ splice (\\[ \f\t\v]*\n)* %% -/* 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. */ - -static void -adjust_location (location *loc, char const *token, size_t size) -{ - int line = scanner_cursor.line; - int column = scanner_cursor.column; - char const *p0 = token; - char const *p = token; - char const *lim = token + size; - - loc->start = scanner_cursor; - - for (p = token; p < lim; p++) - switch (*p) - { - case '\n': - line += line < INT_MAX; - column = 1; - p0 = p + 1; - break; - - case '\t': - { - column = add_column_width (column, p0, p - p0); - column = add_column_width (column, NULL, 8 - ((column - 1) & 7)); - p0 = p + 1; - break; - } - } - - scanner_cursor.line = line; - 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")); -} - - /* Read bytes from FP into buffer BUF of size SIZE. Return the number of bytes read. Remove '\r' from input, treating \r\n and isolated \r as \n. */ @@ -819,173 +761,6 @@ no_cr_read (FILE *fp, char *buf, size_t size) } -/*------------------------------------------------------------------. -| TEXT is pointing to a wannabee semantic value (i.e., a `$'). | -| | -| Possible inputs: $[]($|integer) | -| | -| Output to OBSTACK_FOR_STRING a reference to this semantic value. | -`------------------------------------------------------------------*/ - -static inline bool -handle_action_dollar (char *text, location loc) -{ - const char *type_name = NULL; - char *cp = text + 1; - - if (! current_rule) - return false; - - /* Get the type name if explicit. */ - if (*cp == '<') - { - type_name = ++cp; - while (*cp != '>') - ++cp; - *cp = '\0'; - ++cp; - } - - if (*cp == '$') - { - if (!type_name) - type_name = symbol_list_n_type_name_get (current_rule, loc, 0); - if (!type_name && typed) - complain_at (loc, _("$$ of `%s' has no declared type"), - current_rule->sym->tag); - if (!type_name) - type_name = ""; - obstack_fgrow1 (&obstack_for_string, - "]b4_lhs_value([%s])[", type_name); - current_rule->used = true; - } - else - { - long int num = strtol (cp, NULL, 10); - - if (1 - INT_MAX + rule_length <= num && num <= rule_length) - { - int n = num; - 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, _("$%d of `%s' has no declared type"), - n, current_rule->sym->tag); - if (!type_name) - type_name = ""; - obstack_fgrow3 (&obstack_for_string, - "]b4_rhs_value(%d, %d, [%s])[", - rule_length, n, type_name); - symbol_list_n_used_set (current_rule, n, true); - } - else - complain_at (loc, _("integer out of range: %s"), quote (text)); - } - - return true; -} - - -/*----------------------------------------------------------------. -| Map `$?' 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) -{ - switch (token_type) - { - case BRACED_CODE: - if (handle_action_dollar (text, loc)) - return; - break; - - case PERCENT_DESTRUCTOR: - case PERCENT_INITIAL_ACTION: - case PERCENT_PRINTER: - if (text[1] == '$') - { - obstack_sgrow (&obstack_for_string, "]b4_dollar_dollar["); - return; - } - break; - - default: - break; - } - - complain_at (loc, _("invalid value: %s"), quote (text)); -} - - -/*------------------------------------------------------. -| TEXT is a location token (i.e., a `@...'). Output to | -| OBSTACK_FOR_STRING a reference to this location. | -`------------------------------------------------------*/ - -static inline bool -handle_action_at (char *text, location loc) -{ - char *cp = text + 1; - locations_flag = true; - - if (! current_rule) - return false; - - if (*cp == '$') - obstack_sgrow (&obstack_for_string, "]b4_lhs_location["); - else - { - long int num = strtol (cp, NULL, 10); - - if (1 - INT_MAX + rule_length <= num && num <= rule_length) - { - int n = num; - obstack_fgrow2 (&obstack_for_string, "]b4_rhs_location(%d, %d)[", - rule_length, n); - } - else - complain_at (loc, _("integer out of range: %s"), quote (text)); - } - - return true; -} - - -/*----------------------------------------------------------------. -| 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) -{ - switch (token_type) - { - case BRACED_CODE: - handle_action_at (text, loc); - return; - - case PERCENT_INITIAL_ACTION: - case PERCENT_DESTRUCTOR: - case PERCENT_PRINTER: - if (text[1] == '$') - { - obstack_sgrow (&obstack_for_string, "]b4_at_dollar["); - return; - } - break; - - default: - break; - } - - complain_at (loc, _("invalid value: %s"), quote (text)); -} - /*------------------------------------------------------. | Scan NUMBER for a base-BASE integer at location LOC. | @@ -1065,7 +840,7 @@ convert_ucn_to_byte (char const *ucn) /*----------------------------------------------------------------. -| Handle `#line INT "FILE"'. ARGS has already skipped `#line '. | +| Handle '#line INT "FILE"'. ARGS has already skipped '#line '. | `----------------------------------------------------------------*/ static void @@ -1073,27 +848,15 @@ 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 = mbschr (after_num, '"') + 1; + *mbschr (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")); + current_file = uniqstr_new (file); + boundary_set (&scanner_cursor, current_file, lineno, 1); } @@ -1109,6 +872,10 @@ unexpected_end (boundary start, char const *msgid, char const *token_end) location loc; loc.start = start; loc.end = scanner_cursor; + token_end = quote (token_end); + // Instead of '\'', display "'". + if (!strcmp (token_end, "'\\''")) + token_end = "\"'\""; complain_at (loc, _(msgid), token_end); } @@ -1121,7 +888,7 @@ unexpected_end (boundary start, char const *msgid, char const *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); } @@ -1132,7 +899,7 @@ unexpected_eof (boundary start, char const *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); } @@ -1141,7 +908,7 @@ unexpected_newline (boundary start, char const *token_end) `-------------------------*/ void -scanner_initialize (void) +gram_scanner_initialize (void) { obstack_init (&obstack_for_string); } @@ -1152,9 +919,9 @@ scanner_initialize (void) `-----------------------------------------------*/ void -scanner_free (void) +gram_scanner_free (void) { obstack_free (&obstack_for_string, 0); /* Reclaim Flex's buffers. */ - yy_delete_buffer (YY_CURRENT_BUFFER); + yylex_destroy (); }