1 /*
2 * This file is part of cparser.
3 * Copyright (C) 2007-2009 Matthias Braun <matze@braunis.de>
4 *
5 * This program is free software; you can redistribute it and/or
6 * modify it under the terms of the GNU General Public License
7 * as published by the Free Software Foundation; either version 2
8 * of the License, or (at your option) any later version.
9 *
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
14 *
15 * You should have received a copy of the GNU General Public License
16 * along with this program; if not, write to the Free Software
17 * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
18 * 02111-1307, USA.
19 */
20 #include <config.h>
21
22 #include "token_t.h"
23 #include "symbol_t.h"
24
25 #include <assert.h>
26 #include <stdio.h>
27
28 #include "symbol.h"
29 #include "lang_features.h"
30 #include "adt/array.h"
31 #include "adt/error.h"
32 #include "adt/util.h"
33
34 symbol_t *token_symbols[T_LAST_TOKEN];
35
36 const source_position_t builtin_source_position = { "<built-in>", 0, 0, true };
37
38 static token_kind_t last_id;
39
intern_register_token(token_kind_t id,const char * string)40 static symbol_t *intern_register_token(token_kind_t id, const char *string)
41 {
42 assert(id < T_LAST_TOKEN);
43 symbol_t *symbol = symbol_table_insert(string);
44 if (token_symbols[id] == NULL)
45 token_symbols[id] = symbol;
46 return symbol;
47 }
48
register_token(unsigned mode,token_kind_t id,const char * string)49 static void register_token(unsigned mode, token_kind_t id, const char *string)
50 {
51 if (id > 255) {
52 assert(id >= last_id);
53 last_id = id;
54 }
55 if (c_mode & mode) {
56 symbol_t *symbol = intern_register_token(id, string);
57 symbol->ID = id;
58 }
59 }
60
register_pp_token(pp_token_kind_t const id,char const * const string)61 static void register_pp_token(pp_token_kind_t const id, char const *const string)
62 {
63 assert(id < TP_LAST_TOKEN);
64 symbol_t *const symbol = symbol_table_insert(string);
65 symbol->pp_ID = id;
66 }
67
init_tokens(void)68 void init_tokens(void)
69 {
70 memset(token_symbols, 0, T_LAST_TOKEN * sizeof(token_symbols[0]));
71
72 #define T(mode,x,str,val) register_token(mode, x, str);
73 #include "tokens.inc"
74 #undef T
75
76 #define T(token) register_pp_token(TP_##token, #token);
77 #include "tokens_preprocessor.inc"
78 #undef T
79 }
80
exit_tokens(void)81 void exit_tokens(void)
82 {
83 }
84
print_token_kind(FILE * f,token_kind_t token_kind)85 void print_token_kind(FILE *f, token_kind_t token_kind)
86 {
87 if (token_kind >= lengthof(token_symbols)) {
88 fputs("invalid token", f);
89 return;
90 }
91
92 fputs(token_symbols[token_kind]->string, f);
93 }
94
get_string_encoding_prefix(string_encoding_t const enc)95 char const *get_string_encoding_prefix(string_encoding_t const enc)
96 {
97 switch (enc) {
98 case STRING_ENCODING_CHAR: return "";
99 case STRING_ENCODING_CHAR16: return "u";
100 case STRING_ENCODING_CHAR32: return "U";
101 case STRING_ENCODING_UTF8: return "u8";
102 case STRING_ENCODING_WIDE: return "L";
103 }
104 panic("invalid string encoding");
105 }
106
print_token(FILE * f,const token_t * token)107 void print_token(FILE *f, const token_t *token)
108 {
109 char delim = '\'';
110 char const *enc = "";
111 char const *val;
112 switch (token->kind) {
113 case T_IDENTIFIER:
114 case T_MACRO_PARAMETER:
115 case T_UNKNOWN_CHAR:
116 val = token->base.symbol->string;
117 break;
118
119 case T_STRING_LITERAL:
120 delim = '"';
121 /* FALLTHROUGH */
122 case T_CHARACTER_CONSTANT:
123 enc = get_string_encoding_prefix(token->literal.string.encoding);
124 /* FALLTHROUGH */
125 case T_NUMBER:
126 val = token->literal.string.begin;
127 break;
128
129 default: {
130 char const *kind = (token->base.symbol ? token->base.symbol : token_symbols[token->kind])->string;
131 fprintf(f, "'%s'", kind);
132 return;
133 }
134 }
135 fprintf(f, "%s %s%c%s%c", token_symbols[token->kind]->string, enc, delim, val, delim);
136 }
137
tokens_would_paste(token_kind_t token1,token_kind_t token2)138 bool tokens_would_paste(token_kind_t token1, token_kind_t token2)
139 {
140 char const c = token_symbols[token2]->string[0];
141
142 switch (token1) {
143 case '>': return c == '>' || c == '=';
144 case '<': return c == '<' || c == '=' || c == '%' || c == ':';
145 case '+': return c == '+' || c == '=';
146 case '-': return c == '-' || c == '>';
147 case '/': return c == '/' || c == '=' || c == '*';
148 case '%': return c == ':' || c == '=' || c == '>';
149 case '&': return c == '&' || c == '=';
150 case '|': return c == '|' || c == '=';
151 case ':': return c == ':' || c == '>';
152 case '*': return c == '*' || c == '=';
153 case '.': return c == '.' || c == '%' || token2 == T_NUMBER;
154 case '#': return c == '#' || c == '%';
155 case T_GREATERGREATER: return c == '=';
156 case T_LESSLESS: return c == '=';
157 case '^': return c == '=';
158 case '!': return c == '=';
159
160 case T_IDENTIFIER:
161 return token2 == T_CHARACTER_CONSTANT ||
162 token2 == T_IDENTIFIER ||
163 token2 == T_NUMBER ||
164 token2 == T_STRING_LITERAL; /* L */
165
166 case T_NUMBER:
167 return token2 == T_IDENTIFIER || token2 == T_NUMBER ||
168 token2 == '.' || token2 == '+' || token2 == '-';
169
170 default:
171 return false;
172 }
173 }
174