X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/dafdc66ff0a0344c390d0f5612e9b469d6855e00..ae7453f2ba115eba9d24aad81b0f17b10c6b839a:/src/scan-gram.l diff --git a/src/scan-gram.l b/src/scan-gram.l index 2c80c65b..235272f1 100644 --- a/src/scan-gram.l +++ b/src/scan-gram.l @@ -34,14 +34,14 @@ #define YY_USER_INIT \ do { \ LOCATION_RESET (*yylloc); \ + yylloc->file = infile; \ /* This is only to avoid GCC warnings. */ \ if (yycontrol) {;}; \ } while (0) -#define YY_USER_ACTION LOCATION_COLUMNS (*yylloc, yyleng) -#define YY_LINES LOCATION_LINES (*yylloc, yyleng); lineno += yyleng; -#define YY_STEP LOCATION_STEP (*yylloc) - +#define YY_USER_ACTION LOCATION_COLUMNS (*yylloc, yyleng); +#define YY_LINES LOCATION_LINES (*yylloc, yyleng); +#define YY_STEP LOCATION_STEP (*yylloc); /* STRING_OBSTACK -- Used to store all the characters that we need to keep (to construct ID, STRINGS etc.). Use the following macros to @@ -76,12 +76,21 @@ scanner_last_string_free (void) } - static int braces_level = 0; static int percent_percent_count = 0; -static void handle_dollar PARAMS ((char *cp, location_t location)); -static void handle_at PARAMS ((char *cp)); +/* 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 handle_dollar PARAMS ((braced_code_t code_kind, + char *cp, location_t location)); +static void handle_at PARAMS ((braced_code_t code_kind, + char *cp, location_t location)); %} %x SC_COMMENT @@ -89,7 +98,7 @@ static void handle_at PARAMS ((char *cp)); %x SC_ESCAPED_STRING SC_ESCAPED_CHARACTER %x SC_BRACED_CODE SC_PROLOGUE SC_EPILOGUE -id [.a-zA-Z][.a-zA-Z_0-9]* +id [.a-zA-Z_][.a-zA-Z_0-9]* int [0-9]+ eols (\n|\r|\n\r|\r\n)+ blanks [ \t\f]+ @@ -98,18 +107,7 @@ blanks [ \t\f]+ %{ /* At each yylex invocation, mark the current position as the start of the next token. */ -#define TR_POS 0 -#if TR_POS - fprintf (stderr, "FOO1: %p: ", yylloc); - LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, "\n"); -#endif YY_STEP; -#if TR_POS - fprintf (stderr, "BAR1: "); - LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, "\n"); -#endif %} @@ -122,20 +120,27 @@ blanks [ \t\f]+ "%debug" return PERCENT_DEBUG; "%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; "%file-prefix" return PERCENT_FILE_PREFIX; "%fixed"[-_]"output"[-_]"files" return PERCENT_YACC; + "%glr-parser" return PERCENT_GLR_PARSER; "%left" return PERCENT_LEFT; "%locations" return PERCENT_LOCATIONS; + "%merge" return PERCENT_MERGE; "%name"[-_]"prefix" return PERCENT_NAME_PREFIX; "%no"[-_]"lines" return PERCENT_NO_LINES; "%nonassoc" return PERCENT_NONASSOC; "%nterm" return PERCENT_NTERM; "%output" return PERCENT_OUTPUT; - "%prec" return PERCENT_PREC; + "%parse-param" return PERCENT_PARSE_PARAM; + "%prec" { rule_length--; return PERCENT_PREC; } + "%printer" return PERCENT_PRINTER; "%pure"[-_]"parser" return PERCENT_PURE_PARSER; "%right" return PERCENT_RIGHT; + "%lex-param" return PERCENT_LEX_PARAM; "%skeleton" return PERCENT_SKELETON; "%start" return PERCENT_START; "%term" return PERCENT_TOKEN; @@ -147,24 +152,26 @@ blanks [ \t\f]+ "%yacc" return PERCENT_YACC; "=" return EQUAL; - ":" return COLON; - "|" return PIPE; + ":" { rule_length = 0; return COLON; } + "|" { rule_length = 0; return PIPE; } + "," return COMMA; ";" return SEMICOLON; {eols} YY_LINES; YY_STEP; {blanks} YY_STEP; {id} { - yylval->symbol = getsym (yytext, *yylloc); + yylval->symbol = symbol_get (yytext, *yylloc); + rule_length++; return ID; } {int} yylval->integer = strtol (yytext, 0, 10); return INT; /* Characters. We don't check there is only one. */ - \' YY_OBS_GROW; yy_push_state (SC_ESCAPED_CHARACTER); + "'" YY_OBS_GROW; yy_push_state (SC_ESCAPED_CHARACTER); /* Strings. */ - \" YY_OBS_GROW; yy_push_state (SC_ESCAPED_STRING); + "\"" YY_OBS_GROW; yy_push_state (SC_ESCAPED_STRING); /* Comments. */ "/*" yy_push_state (SC_COMMENT); @@ -193,7 +200,7 @@ blanks [ \t\f]+ . { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": invalid character: `%c'\n", *yytext); + fprintf (stderr, _(": invalid character: `%c'\n"), *yytext); YY_STEP; } } @@ -237,7 +244,7 @@ blanks [ \t\f]+ <> { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": unexpected end of file in a comment\n"); + fprintf (stderr, _(": unexpected end of file in a comment\n")); yy_pop_state (); } } @@ -250,12 +257,13 @@ blanks [ \t\f]+ { - \" { + "\"" { assert (yy_top_state () == INITIAL); YY_OBS_GROW; YY_OBS_FINISH; yylval->string = last_string; yy_pop_state (); + rule_length++; return STRING; } @@ -265,7 +273,7 @@ blanks [ \t\f]+ <> { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": unexpected end of file in a string\n"); + fprintf (stderr, _(": unexpected end of file in a string\n")); assert (yy_top_state () == INITIAL); YY_OBS_FINISH; yylval->string = last_string; @@ -281,27 +289,29 @@ blanks [ \t\f]+ { - \' { + "'" { YY_OBS_GROW; assert (yy_top_state () == INITIAL); { YY_OBS_FINISH; - yylval->symbol = getsym (last_string, *yylloc); - symbol_class_set (yylval->symbol, token_sym); - symbol_user_token_number_set (yylval->symbol, last_string[1]); + yylval->symbol = symbol_get (last_string, *yylloc); + symbol_class_set (yylval->symbol, token_sym, *yylloc); + symbol_user_token_number_set (yylval->symbol, + (unsigned char) last_string[1], *yylloc); YY_OBS_FREE; yy_pop_state (); + rule_length++; return ID; } } - [^\'\n\r\\] YY_OBS_GROW; + [^\n\r\\] YY_OBS_GROW; {eols} obstack_1grow (&string_obstack, '\n'); YY_LINES; <> { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": unexpected end of file in a character\n"); + fprintf (stderr, _(": unexpected end of file in a character\n")); assert (yy_top_state () == INITIAL); YY_OBS_FINISH; yylval->string = last_string; @@ -322,7 +332,7 @@ blanks [ \t\f]+ if (c > 255) { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": invalid escape: %s\n", yytext); + fprintf (stderr, _(": invalid escape: %s\n"), quote (yytext)); YY_STEP; } else @@ -340,12 +350,14 @@ blanks [ \t\f]+ \\r obstack_1grow (&string_obstack, '\r'); \\t obstack_1grow (&string_obstack, '\t'); \\v obstack_1grow (&string_obstack, '\v'); - \\[\\""] obstack_1grow (&string_obstack, yytext[1]); - \\. { + \\[\\""''] obstack_1grow (&string_obstack, yytext[1]); + \\(.|\n) { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": unrecognized escape: %s\n", yytext); + fprintf (stderr, _(": unrecognized escape: %s\n"), quote (yytext)); YY_OBS_GROW; } + /* FLex wants this rule, in case of a `\<>'. */ + \\ YY_OBS_GROW; } @@ -356,20 +368,22 @@ blanks [ \t\f]+ { - \' { + "'" { YY_OBS_GROW; assert (yy_top_state () != INITIAL); yy_pop_state (); } - [^\[\]\'\n\r\\] YY_OBS_GROW; - \\. YY_OBS_GROW; + [^\[\]\'\n\r\\]+ YY_OBS_GROW; + \\(.|\n) YY_OBS_GROW; + /* FLex wants this rule, in case of a `\<>'. */ + \\ YY_OBS_GROW; {eols} YY_OBS_GROW; YY_LINES; <> { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": unexpected end of file in a character\n"); + fprintf (stderr, _(": unexpected end of file in a character\n")); assert (yy_top_state () != INITIAL); yy_pop_state (); } @@ -383,20 +397,22 @@ blanks [ \t\f]+ { - \" { + "\"" { assert (yy_top_state () != INITIAL); YY_OBS_GROW; yy_pop_state (); } [^\[\]\"\n\r\\]+ YY_OBS_GROW; - \\. YY_OBS_GROW; + \\(.|\n) YY_OBS_GROW; + /* FLex wants this rule, in case of a `\<>'. */ + \\ YY_OBS_GROW; {eols} YY_OBS_GROW; YY_LINES; <> { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": unexpected end of file in a string\n"); + fprintf (stderr, _(": unexpected end of file in a string\n")); assert (yy_top_state () != INITIAL); yy_pop_state (); } @@ -410,14 +426,17 @@ blanks [ \t\f]+ { /* Characters. We don't check there is only one. */ - \' YY_OBS_GROW; yy_push_state (SC_CHARACTER); + "'" YY_OBS_GROW; yy_push_state (SC_CHARACTER); /* Strings. */ - \" YY_OBS_GROW; yy_push_state (SC_STRING); + "\"" YY_OBS_GROW; yy_push_state (SC_STRING); /* Comments. */ "/*" YY_OBS_GROW; yy_push_state (SC_COMMENT); "//".* YY_OBS_GROW; + + /* Not comments. */ + "/" YY_OBS_GROW; } @@ -435,14 +454,17 @@ blanks [ \t\f]+ yy_pop_state (); YY_OBS_FINISH; yylval->string = last_string; + rule_length++; return BRACED_CODE; } } "{" YY_OBS_GROW; braces_level++; - "$"("<"[^>]+">")?(-?[0-9]+|"$") { handle_dollar (yytext, *yylloc); } - "@"(-?[0-9]+|"$") { handle_at (yytext); } + "$"("<"[^>]+">")?(-?[0-9]+|"$") { handle_dollar (current_braced_code, + yytext, *yylloc); } + "@"(-?[0-9]+|"$") { handle_at (current_braced_code, + yytext, *yylloc); } [^$@\[\]/\'\"\{\}\n\r]+ YY_OBS_GROW; {eols} YY_OBS_GROW; YY_LINES; @@ -452,7 +474,7 @@ blanks [ \t\f]+ <> { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": unexpected end of file in a braced code\n"); + fprintf (stderr, _(": unexpected end of file in a braced code\n")); yy_pop_state (); YY_OBS_FINISH; yylval->string = last_string; @@ -475,19 +497,18 @@ blanks [ \t\f]+ return PROLOGUE; } - [^%\[\]/\'\"\n\r]+ YY_OBS_GROW; - "%"+[^%\}\n\r]+ YY_OBS_GROW; - {eols} YY_OBS_GROW; YY_LINES; + [^%\[\]/\'\"\n\r]+ YY_OBS_GROW; + "%" YY_OBS_GROW; + {eols} YY_OBS_GROW; YY_LINES; <> { LOCATION_PRINT (stderr, *yylloc); - fprintf (stderr, ": unexpected end of file in a prologue\n"); + fprintf (stderr, _(": unexpected end of file in a prologue\n")); yy_pop_state (); YY_OBS_FINISH; yylval->string = last_string; return PROLOGUE; } - } @@ -512,19 +533,18 @@ blanks [ \t\f]+ %% /*------------------------------------------------------------------. -| CP is pointing to a wannabee semantic value (i.e., a `$'). | +| TEXT is pointing to a wannabee semantic value (i.e., a `$'). | | | | Possible inputs: $[]($|integer) | | | | Output to the STRING_OBSTACK a reference to this semantic value. | `------------------------------------------------------------------*/ -static void -handle_dollar (char *cp, location_t location) +static inline void +handle_action_dollar (char *text, location_t location) { const char *type_name = NULL; - - ++cp; + char *cp = text + 1; /* Get the type name if explicit. */ if (*cp == '<') @@ -548,13 +568,8 @@ handle_dollar (char *cp, location_t location) obstack_fgrow1 (&string_obstack, "]b4_lhs_value([%s])[", type_name); } - else if (isdigit (*cp) || *cp == '-') + else if (('0' <= *cp && *cp <= '9') || *cp == '-') { - /* RULE_LENGTH is the number of values in the current rule so - far, which says where to find `$0' with respect to the top of - the stack. It is not the same as the rule->length in the - case of mid rule actions. */ - int rule_length = symbol_list_length (current_rule->next); int n = strtol (cp, &cp, 10); if (n > rule_length) @@ -576,50 +591,125 @@ handle_dollar (char *cp, location_t location) } else { - char buf[] = "$c"; - buf[1] = *cp; - complain (_("%s is invalid"), quote (buf)); + complain_at (location, _("%s is invalid"), quote (text)); } } -/*-------------------------------------------------------. -| CP is pointing to a location (i.e., a `@'). Output to | -| STRING_OBSTACK a reference to this location. | -`-------------------------------------------------------*/ + +/*---------------------------------------------------------------. +| TEXT is expexted tp be $$ in some code associated to a symbol: | +| destructor or printer. | +`---------------------------------------------------------------*/ + +static inline void +handle_symbol_code_dollar (char *text, location_t location) +{ + char *cp = text + 1; + if (*cp == '$') + obstack_sgrow (&string_obstack, "]b4_dollar_dollar["); + else + complain_at (location, _("%s is invalid"), quote (text)); +} + + +/*-----------------------------------------------------------------. +| Dispatch onto handle_action_dollar, or handle_destructor_dollar, | +| depending upon CODE_KIND. | +`-----------------------------------------------------------------*/ static void -handle_at (char *cp) +handle_dollar (braced_code_t braced_code_kind, + char *text, location_t location) { + switch (braced_code_kind) + { + case action_braced_code: + handle_action_dollar (text, location); + break; + + case destructor_braced_code: + case printer_braced_code: + handle_symbol_code_dollar (text, location); + break; + } +} + + +/*------------------------------------------------------. +| TEXT is a location token (i.e., a `@...'). Output to | +| STRING_OBSTACK a reference to this location. | +`------------------------------------------------------*/ + +static inline void +handle_action_at (char *text, location_t location) +{ + char *cp = text + 1; locations_flag = 1; - ++cp; if (*cp == '$') { obstack_sgrow (&string_obstack, "]b4_lhs_location["); } - else if (isdigit (*cp) || *cp == '-') + else if (('0' <= *cp && *cp <= '9') || *cp == '-') { - /* RULE_LENGTH is the number of values in the current rule so - far, which says where to find `$0' with respect to the top of - the stack. It is not the same as the rule->length in the - case of mid rule actions. */ - int rule_length = symbol_list_length (current_rule->next); int n = strtol (cp, &cp, 10); if (n > rule_length) - complain (_("invalid value: %s%d"), "@", n); + complain_at (location, _("invalid value: %s%d"), "@", n); else obstack_fgrow2 (&string_obstack, "]b4_rhs_location([%d], [%d])[", rule_length, n); } else { - char buf[] = "@c"; - buf[1] = *cp; - complain (_("%s is invalid"), quote (buf)); + complain_at (location, _("%s is invalid"), quote (text)); + } +} + + +/*---------------------------------------------------------------. +| TEXT is expexted tp be @$ in some code associated to a symbol: | +| destructor or printer. | +`---------------------------------------------------------------*/ + +static inline void +handle_symbol_code_at (char *text, location_t location) +{ + char *cp = text + 1; + if (*cp == '$') + obstack_sgrow (&string_obstack, "]b4_at_dollar["); + else + complain_at (location, _("%s is invalid"), quote (text)); +} + + +/*-------------------------------------------------------------------. +| Dispatch onto handle_action_at, or handle_destructor_at, depending | +| upon CODE_KIND. | +`-------------------------------------------------------------------*/ + +static void +handle_at (braced_code_t braced_code_kind, + char *text, location_t location) +{ + switch (braced_code_kind) + { + case action_braced_code: + handle_action_at (text, location); + break; + + case destructor_braced_code: + case printer_braced_code: + handle_symbol_code_at (text, location); + break; } } + +/*-------------------------. +| Initialize the scanner. | +`-------------------------*/ + void scanner_initialize (void) { @@ -627,8 +717,14 @@ scanner_initialize (void) } +/*-----------------------------------------------. +| Free all the memory allocated to the scanner. | +`-----------------------------------------------*/ + void scanner_free (void) { obstack_free (&string_obstack, 0); + /* Reclaim Flex's buffers. */ + yy_delete_buffer (YY_CURRENT_BUFFER); }