-
Notifications
You must be signed in to change notification settings - Fork 0
/
lex.cpp
335 lines (294 loc) · 12.3 KB
/
lex.cpp
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
#include <assert.h>
#include "lex.hpp"
#include "error.hpp"
#include <queue>
FILE *fp = nullptr;
bool is_space( int c ) { return c == '\n' || c == '\t' || c == ' ' || c == '\r'; }
bool is_letter( int c ) { return isalpha( c ) || c == '_'; }
std::queue<Tokinfo> tokens;
std::queue<Tokinfo> semicolons; // HACK: unlex semicolons with lower priority???
void unlex( Tokinfo t ) { tokens.push( t ); }
Tokinfo lex() {
// static to preserve value across calls.
// this allows me to inspect result's token before I set it in a given call
// to check what the previous token was
static Tokinfo result;
// prioritize tokens in the queue
if ( !tokens.empty() ) {
result = tokens.front();
tokens.pop();
return result;
}
if ( !semicolons.empty() ) {
result = semicolons.front();
semicolons.pop();
return result;
}
// CITATION-ish: the following code draws heavily from the FSM example given to us
// by Dr. Aycock in class -
// linked here -https://d2l.ucalgary.ca/d2l/le/content/500748/viewContent/5744437/View
spin:
int ch;
// spin until we find something interesting
while ( ( ch = getc( fp ) ) != EOF && is_space( ch ) )
if ( ch == '\n' ) result.linenum++;
// TODO: semicolon insertion is triggered on a newline or EOF, so just do it here
result.lexeme = ch; // first (and possibly only) character of the lexeme
// FSM based on our character that chooses what token to return
// note that after the statement nothing happens except returning `result`
// this means that calling `break` inside the statement is effectively a
// return
switch ( ch ) {
// NULL character isn't allowed in input
case NULL:
warning( result.linenum , "skipping NULL character");
goto spin;
case EOF:
// insert semicolon according to spec
if ( result.token == TOKEN_ID
|| result.token == TOKEN_INT
|| result.token == TOKEN_STRING
|| result.token == TOKEN_BREAK
|| result.token == TOKEN_RETURN
|| result.token == TOKEN_RBRACE
|| result.token == TOKEN_RPAREN
) {
result.token = TOKEN_SEMICOLON;
ungetc( ch, fp );
}
else result.token = TOKEN_EOF;
result.lexeme = "";
break;
case '+':
result.token = TOKEN_PLUS;
break;
case '-':
result.token = TOKEN_MINUS;
break;
case '*':
result.token = TOKEN_STAR;
break;
case '%':
result.token = TOKEN_PERCENT;
break;
case '(':
result.token = TOKEN_LPAREN;
break;
case ')':
result.token = TOKEN_RPAREN;
// peek next char to see if its a newline. if so, insert a semicolon
if ( ( ch = getc( fp ) ) == '\n' || ch == '\r' )
semicolons.push ( Tokinfo { TOKEN_SEMICOLON, result.linenum, "", } );
// put back to mimick "peeking"
ungetc( ch, fp );
break;
case '{':
result.token = TOKEN_LBRACE;
break;
case '}':
// if last token wasn't a semicolon, insert a semicolon into the token stream
if ( result.token != TOKEN_SEMICOLON && result.token != TOKEN_UNSET ) {
result.token = TOKEN_SEMICOLON;
result.lexeme = "";
ungetc( ch, fp );
break;
}
result.token = TOKEN_RBRACE;
// peek next char to see if its a newline. if so, insert a semicolon
if ( ( ch = getc( fp ) ) == '\n' || ch == '\r' )
semicolons.push( Tokinfo { TOKEN_SEMICOLON, result.linenum, "", } );
// put back to mimick "peeking"
ungetc( ch, fp );
break;
case ',':
result.token = TOKEN_COMMA;
break;
case ';':
result.token = TOKEN_SEMICOLON;
break;
case '&':
// no unary & operator
if ( ( ch = getc( fp ) ) != '&' ) error( result.linenum, "bitwise AND not supported in GoLF" );
result.token = TOKEN_LOGIC_AND;
result.lexeme.push_back( ch );
break;
case '|':
// no unary | operator
if ( ( ch = getc( fp ) ) != '|' ) error( result.linenum, "bitwise OR not supported in GoLF");
result.token = TOKEN_LOGIC_OR;
result.lexeme.push_back( ch );
break;
case '=':
// if '=' is next, then it's == comparison, else =
if ( ( ch = getc( fp ) ) == '=' ) {
result.token = TOKEN_EQ;
result.lexeme.push_back( ch );
} else {
result.token = TOKEN_ASSIGN;
ungetc( ch, fp );
}
break;
case '<':
// if '=' is next, then it's <= comparison, else =
if ( ( ch = getc( fp ) ) == '=' ) {
result.token = TOKEN_LEQ;
result.lexeme.push_back( ch );
} else {
result.token = TOKEN_LT;
ungetc( ch, fp );
}
break;
case '>':
// if '=' is next, then it's >= comparison, else =
if ( ( ch = getc( fp ) ) == '=' ) {
result.token = TOKEN_GEQ;
result.lexeme.push_back( ch );
} else {
result.token = TOKEN_GT;
ungetc( ch, fp );
}
break;
case '!':
// if '=' is next, then it's >= comparison, else =
if ( ( ch = getc( fp ) ) == '=' ) {
result.token = TOKEN_NEQ;
result.lexeme.push_back( ch );
} else {
result.token = TOKEN_BANG;
ungetc( ch, fp );
}
break;
case '/':
// single slash
if ( ( ch = getc( fp ) ) != '/' ) {
result.token = TOKEN_SLASH;
ungetc( ch, fp );
break;
}
// comment - spin until we hit newline or EOF then lex at that newline/EOF
while ( ( ch = getc( fp ) ) != EOF && ch != '\n' )
;
ungetc( ch, fp );
// insert semicolon according to spec
if ( result.token == TOKEN_ID
|| result.token == TOKEN_INT
|| result.token == TOKEN_STRING
|| result.token == TOKEN_BREAK
|| result.token == TOKEN_RETURN
|| result.token == TOKEN_RBRACE
|| result.token == TOKEN_RPAREN
) {
result.token = TOKEN_SEMICOLON;
result.lexeme = "";
break;
}
goto spin;
case '"': // beginning of string literal
result.lexeme.clear(); // don't include the quote in the lexeme
// spin until we hit a terminating quote
while ( ( ch = getc( fp ) ) && ch != '"' ) {
// error if newline or EOF encountered within string
if ( ch == '\n' ) error( result.linenum, "string contains newline" );
if ( ch == EOF ) error( result.linenum, "string terminated by EOF" );
// on backslash: check for valid escape sequences
if ( ch == '\\' ) {
result.lexeme += ch;
ch = getc( fp );
// error if newline or EOF encountered within string
if ( ch == '\n' ) error( result.linenum, "string contains newline" );
if ( ch == EOF ) error( result.linenum, "string terminated by EOF" );
if ( ch != 'b' && ch != 'f' &&ch != 'n' &&ch != 'r' &&ch != 't' &&ch != '\\' &&ch != '"' )
error( result.linenum, "bad string escape '%s'", ascii_char_to_string( ch ).data() );
}
result.lexeme += ch;
}
// add semicolon if this thing is the last thing on the line or in file
if ( ( ch = getc( fp ) ) == '\n' || ch == '\r' )
semicolons.push( Tokinfo{ TOKEN_SEMICOLON, result.linenum, "", } );
result.token = TOKEN_STRING;
ungetc( ch, fp );
break;
default:
// spin on whitespace
if ( is_space( ch ) ) goto spin;
// number means integer literal
if ( isdigit( ch ) ) {
// spin until we hit a non-num
while ( isdigit( ( ch = getc( fp ) ) ) )
result.lexeme += ch;
// add semicolon if this thing is the last thing on the line or in file
if ( ch == '\n' || ch == '\r' ) semicolons.push( Tokinfo { TOKEN_SEMICOLON, result.linenum, "", } );
result.token = TOKEN_INT;
ungetc( ch, fp );
break;
}
// letter means identifier or reserved word
if ( is_letter( ch ) ) {
// spin until we hit a non-letter
while ( ( ch = getc( fp ) ) != EOF && ( is_letter( ch ) || isdigit( ch ) ) )
result.lexeme += ch;
// check if token is a reserved work and act accordingly
if ( result.lexeme == "break" ) result.token = TOKEN_BREAK;
else if ( result.lexeme == "else" ) result.token = TOKEN_ELSE;
else if ( result.lexeme == "for" ) result.token = TOKEN_FOR;
else if ( result.lexeme == "func" ) result.token = TOKEN_FUNC;
else if ( result.lexeme == "if" ) result.token = TOKEN_IF;
else if ( result.lexeme == "return" ) result.token = TOKEN_RETURN;
else if ( result.lexeme == "var" ) result.token = TOKEN_VAR;
//otherwise, result is a generic identifier
else result.token = TOKEN_ID;
// if end of line and token is break, return, or an identifier, insert a semicolon
if ( ( ch == '\n' || ch == '\r' ) && ( result.token == TOKEN_BREAK || result.token == TOKEN_RETURN || result.token == TOKEN_ID ) )
semicolons.push( Tokinfo { TOKEN_SEMICOLON, result.linenum, "", } );
ungetc( ch, fp );
break;
}
// skip non-ascii with a warning
if ( ch > 127 ) {
warning( result.linenum, "skipping non-ASCII input character");
goto spin;
}
// skip unknown character with warning
warning( result.linenum, "skipping unknown character '%s'", ascii_char_to_string( ch ).data() );
goto spin;
break;
}
return result;
}
const char *token_to_string( TokenID token ) {
switch ( token ) {
case TOKEN_EOF: return "EOF";
case TOKEN_BREAK: return "break";
case TOKEN_ELSE: return "else";
case TOKEN_FOR: return "for";
case TOKEN_FUNC: return "func";
case TOKEN_IF: return "if";
case TOKEN_RETURN: return "return";
case TOKEN_VAR: return "var";
case TOKEN_PLUS: return "+";
case TOKEN_MINUS: return "-";
case TOKEN_STAR: return "*";
case TOKEN_SLASH: return "/";
case TOKEN_PERCENT: return "%";
case TOKEN_LOGIC_AND: return "&&";
case TOKEN_LOGIC_OR: return "||";
case TOKEN_EQ: return "==";
case TOKEN_LT: return "<";
case TOKEN_GT: return ">";
case TOKEN_ASSIGN: return "=";
case TOKEN_BANG: return "!";
case TOKEN_NEQ: return "!=";
case TOKEN_LEQ: return "<=";
case TOKEN_GEQ: return ">=";
case TOKEN_LPAREN: return "(";
case TOKEN_RPAREN: return ")";
case TOKEN_LBRACE: return "{";
case TOKEN_RBRACE: return "}";
case TOKEN_COMMA: return ",";
case TOKEN_SEMICOLON: return ";";
case TOKEN_ID: return "id";
case TOKEN_INT: return "int";
case TOKEN_STRING: return "string";
case TOKEN_UNSET: return "DEBUG";
}
}