Context Navigation

lex.ll @ daf9671

ADTaaron-thesisarm-ehast-experimentalcleanup-dtorsdeferred_resndemanglerenumforall-pointer-decayjacob/cs343-translationjenkins-sandboxnew-astnew-ast-unique-exprnew-envno_listpersistent-indexerpthread-emulationqualifiedEnumresolv-newwith_gc

Last change on this file since daf9671 was daf9671, checked in by Peter A. Buhr <pabuhr@…>, 7 years ago
change lexing of multiply operator identifier
Property mode set to `100644`
File size: 16.3 KB

Line
1	/*
2	* Cforall Version 1.0.0 Copyright (C) 2015 University of Waterloo
3	*
4	* The contents of this file are covered under the licence agreement in the
5	* file "LICENCE" distributed with Cforall.
6	*
7	* lex.l --
8	*
9	* Author : Peter A. Buhr
10	* Created On : Sat Sep 22 08:58:10 2001
11	* Last Modified By : Peter A. Buhr
12	* Last Modified On : Mon May 22 07:46:30 2017
13	* Update Count : 525
14	*/
15
16	%option yylineno
17	%option nounput
18
19	%{
20	// This lexer assumes the program has been preprocessed by cpp. Hence, all user level preprocessor directive have been
21	// performed and removed from the source. The only exceptions are preprocessor directives passed to the compiler (e.g.,
22	// line-number directives) and C/C++ style comments, which are ignored.
23
24	//************************** Includes and Defines **************************
25
26	#include <string>
27	#include <cstdio> // FILENAME_MAX
28
29	#include "lex.h"
30	#include "parser.h" // YACC generated definitions based on C++ grammar
31	#include "ParseNode.h"
32	#include "TypedefTable.h"
33
34	char *yyfilename;
35	std::string *strtext; // accumulate parts of character and string constant value
36
37	#define RETURN_LOCN(x) yylval.tok.loc.file = yyfilename; yylval.tok.loc.line = yylineno; return( x )
38	#define RETURN_VAL(x) yylval.tok.str = new std::string( yytext ); RETURN_LOCN( x )
39	#define RETURN_CHAR(x) yylval.tok.str = nullptr; RETURN_LOCN( x )
40	#define RETURN_STR(x) yylval.tok.str = strtext; RETURN_LOCN( x )
41
42	#define WHITE_RETURN(x) // do nothing
43	#define NEWLINE_RETURN() WHITE_RETURN( '\n' )
44	#define ASCIIOP_RETURN() RETURN_CHAR( (int)yytext[0] ) // single character operator
45	#define NAMEDOP_RETURN(x) RETURN_CHAR( x ) // multichar operator, with a name
46	#define NUMERIC_RETURN(x) rm_underscore(); RETURN_VAL( x ) // numeric constant
47	#define KEYWORD_RETURN(x) RETURN_CHAR( x ) // keyword
48	#define IDENTIFIER_RETURN() RETURN_VAL( typedefTable.isKind( yytext ) )
49	#define ATTRIBUTE_RETURN() RETURN_VAL( ATTR_IDENTIFIER )
50
51	void rm_underscore() {
52	// Remove underscores in numeric constant by copying the non-underscore characters to the front of the string.
53	yyleng = 0;
54	for ( int i = 0; yytext[i] != '\0'; i += 1 ) {
55	if ( yytext[i] != '_' ) {
56	yytext[yyleng] = yytext[i];
57	yyleng += 1;
58	} // if
59	} // for
60	yytext[yyleng] = '\0';
61	}
62
63	%}
64
65	octal [0-7]
66	nonzero [1-9]
67	decimal [0-9]
68	hex [0-9a-fA-F]
69	universal_char "\\"((u"_"?{hex_quad})\|(U"_"?{hex_quad}{2}))
70
71	// identifier, GCC: $ in identifier
72	identifier ([a-zA-Z_$]\|{universal_char})([0-9a-zA-Z_$]\|{universal_char})*
73
74	// attribute identifier, GCC: $ in identifier
75	attr_identifier "@"{identifier}
76
77	// numeric constants, CFA: '_' in constant
78	hex_quad {hex}("_"?{hex}){3}
79	integer_suffix "_"?(([uU](("ll"\|"LL"\|[lL])[iI]\|[iI]?("ll"\|"LL"\|[lL])?))\|([iI](("ll"\|"LL"\|[lL])[uU]\|[uU]?("ll"\|"LL"\|[lL])?))\|(("ll"\|"LL"\|[lL])([iI][uU]\|[uU]?[iI]?)))
80
81	octal_digits ({octal})\|({octal}({octal}\|"_")*{octal})
82	octal_prefix "0""_"?
83	octal_constant (("0")\|({octal_prefix}{octal_digits})){integer_suffix}?
84
85	nonzero_digits ({nonzero})\|({nonzero}({decimal}\|"_")*{decimal})
86	decimal_constant {nonzero_digits}{integer_suffix}?
87
88	hex_digits ({hex})\|({hex}({hex}\|"_")*{hex})
89	hex_prefix "0"[xX]"_"?
90	hex_constant {hex_prefix}{hex_digits}{integer_suffix}?
91
92	decimal_digits ({decimal})\|({decimal}({decimal}\|"_")*{decimal})
93	real_decimal {decimal_digits}"."{exponent}?{floating_suffix}?
94	real_fraction "."{decimal_digits}{exponent}?{floating_suffix}?
95	real_constant {decimal_digits}{real_fraction}
96	exponent "_"?[eE]"_"?[+-]?{decimal_digits}
97	// GCC: D (double) and iI (imaginary) suffixes, and DL (long double)
98	floating_suffix "_"?([fFdDlL][iI]?\|[iI][lLfFdD]?\|"DL")
99	floating_constant (({real_constant}{exponent}?)\|({decimal_digits}{exponent})){floating_suffix}?
100
101	binary_exponent "_"?[pP]"_"?[+-]?{decimal_digits}
102	hex_fractional_constant ({hex_digits}?"."{hex_digits})\|({hex_digits}".")
103	hex_floating_constant {hex_prefix}(({hex_fractional_constant}{binary_exponent})\|({hex_digits}{binary_exponent})){floating_suffix}?
104
105	// character escape sequence, GCC: \e => esc character
106	simple_escape "\\"[abefnrtv'"?\\]
107	// ' stop highlighting
108	octal_escape "\\"{octal}("_"?{octal}){0,2}
109	hex_escape "\\""x""_"?{hex_digits}
110	escape_seq {simple_escape}\|{octal_escape}\|{hex_escape}\|{universal_char}
111	cwide_prefix "L"\|"U"\|"u"
112	swide_prefix {cwide_prefix}\|"u8"
113
114	// display/white-space characters
115	h_tab [\011]
116	form_feed [\014]
117	v_tab [\013]
118	c_return [\015]
119	h_white [ ]\|{h_tab}
120
121	// overloadable operators
122	op_unary_only "~"\|"!"
123	op_unary_binary "+"\|"-"\|"*"
124	op_unary_pre_post "++"\|"--"
125	op_unary {op_unary_only}\|{op_unary_binary}\|{op_unary_pre_post}
126
127	op_binary_only "/"\|"%"\|"^"\|"&"\|"\|"\|"<"\|">"\|"="\|"=="\|"!="\|"<<"\|">>"\|"<="\|">="\|"+="\|"-="\|"*="\|"/="\|"%="\|"&="\|"\|="\|"^="\|"<<="\|">>="
128	op_binary_over {op_unary_binary}\|{op_binary_only}
129	// op_binary_not_over "?"\|"->"\|"."\|"&&"\|"\|\|"\|"@="
130	// operator {op_unary_pre_post}\|{op_binary_over}\|{op_binary_not_over}
131
132	%x COMMENT
133	%x BKQUOTE
134	%x QUOTE
135	%x STRING
136
137	%%
138	/* line directives */
139	^{h_white}"#"{h_white}[0-9]+{h_white}["][^"\n]+["]."\n" {
140	/* " stop highlighting */
141	static char filename[FILENAME_MAX]; // temporarily store current source-file name
142	char *end_num;
143	char begin_string, end_string;
144	long lineno, length;
145	lineno = strtol( yytext + 1, &end_num, 0 );
146	begin_string = strchr( end_num, '"' );
147	if ( begin_string ) { // file name ?
148	end_string = strchr( begin_string + 1, '"' ); // look for ending delimiter
149	assert( end_string ); // closing quote ?
150	length = end_string - begin_string - 1; // file-name length without quotes or sentinel
151	assert( length < FILENAME_MAX ); // room for sentinel ?
152	memcpy( &filename, begin_string + 1, length ); // copy file name from yytext
153	filename[ length ] = '\0'; // terminate string with sentinel
154	//std::cout << "file " << filename << " line " << lineno << std::endl;
155	yylineno = lineno;
156	yyfilename = filename;
157	} // if
158	}
159
160	/* ignore preprocessor directives (for now) */
161	^{h_white}"#"[^\n]"\n" ;
162
163	/* ignore C style comments (ALSO HANDLED BY CPP) */
164	"/*" { BEGIN COMMENT; }
165	<COMMENT>.\|\n ;
166	<COMMENT>"*/" { BEGIN 0; }
167
168	/* ignore C++ style comments (ALSO HANDLED BY CPP) */
169	"//"[^\n]*"\n" ;
170
171	/* ignore whitespace */
172	{h_white}+ { WHITE_RETURN(' '); }
173	({v_tab}\|{c_return}\|{form_feed})+ { WHITE_RETURN(' '); }
174	({h_white}\|{v_tab}\|{c_return}\|{form_feed})*"\n" { NEWLINE_RETURN(); }
175
176	/* keywords */
177	_Alignas { KEYWORD_RETURN(ALIGNAS); } // C11
178	_Alignof { KEYWORD_RETURN(ALIGNOF); } // C11
179	__alignof { KEYWORD_RETURN(ALIGNOF); } // GCC
180	__alignof__ { KEYWORD_RETURN(ALIGNOF); } // GCC
181	asm { KEYWORD_RETURN(ASM); }
182	__asm { KEYWORD_RETURN(ASM); } // GCC
183	__asm__ { KEYWORD_RETURN(ASM); } // GCC
184	_At { KEYWORD_RETURN(AT); } // CFA
185	_Atomic { KEYWORD_RETURN(ATOMIC); } // C11
186	__attribute { KEYWORD_RETURN(ATTRIBUTE); } // GCC
187	__attribute__ { KEYWORD_RETURN(ATTRIBUTE); } // GCC
188	auto { KEYWORD_RETURN(AUTO); }
189	_Bool { KEYWORD_RETURN(BOOL); } // C99
190	break { KEYWORD_RETURN(BREAK); }
191	case { KEYWORD_RETURN(CASE); }
192	catch { KEYWORD_RETURN(CATCH); } // CFA
193	catchResume { KEYWORD_RETURN(CATCHRESUME); } // CFA
194	char { KEYWORD_RETURN(CHAR); }
195	choose { KEYWORD_RETURN(CHOOSE); } // CFA
196	_Complex { KEYWORD_RETURN(COMPLEX); } // C99
197	__complex { KEYWORD_RETURN(COMPLEX); } // GCC
198	__complex__ { KEYWORD_RETURN(COMPLEX); } // GCC
199	const { KEYWORD_RETURN(CONST); }
200	__const { KEYWORD_RETURN(CONST); } // GCC
201	__const__ { KEYWORD_RETURN(CONST); } // GCC
202	continue { KEYWORD_RETURN(CONTINUE); }
203	coroutine { KEYWORD_RETURN(COROUTINE); } // CFA
204	default { KEYWORD_RETURN(DEFAULT); }
205	disable { KEYWORD_RETURN(DISABLE); } // CFA
206	do { KEYWORD_RETURN(DO); }
207	double { KEYWORD_RETURN(DOUBLE); }
208	dtype { KEYWORD_RETURN(DTYPE); } // CFA
209	else { KEYWORD_RETURN(ELSE); }
210	enable { KEYWORD_RETURN(ENABLE); } // CFA
211	enum { KEYWORD_RETURN(ENUM); }
212	__extension__ { KEYWORD_RETURN(EXTENSION); } // GCC
213	extern { KEYWORD_RETURN(EXTERN); }
214	fallthrough { KEYWORD_RETURN(FALLTHRU); } // CFA
215	fallthru { KEYWORD_RETURN(FALLTHRU); } // CFA
216	finally { KEYWORD_RETURN(FINALLY); } // CFA
217	float { KEYWORD_RETURN(FLOAT); }
218	__float128 { KEYWORD_RETURN(FLOAT); } // GCC
219	for { KEYWORD_RETURN(FOR); }
220	forall { KEYWORD_RETURN(FORALL); } // CFA
221	fortran { KEYWORD_RETURN(FORTRAN); }
222	ftype { KEYWORD_RETURN(FTYPE); } // CFA
223	_Generic { KEYWORD_RETURN(GENERIC); } // C11
224	goto { KEYWORD_RETURN(GOTO); }
225	if { KEYWORD_RETURN(IF); }
226	_Imaginary { KEYWORD_RETURN(IMAGINARY); } // C99
227	__imag { KEYWORD_RETURN(IMAGINARY); } // GCC
228	__imag__ { KEYWORD_RETURN(IMAGINARY); } // GCC
229	inline { KEYWORD_RETURN(INLINE); } // C99
230	__inline { KEYWORD_RETURN(INLINE); } // GCC
231	__inline__ { KEYWORD_RETURN(INLINE); } // GCC
232	int { KEYWORD_RETURN(INT); }
233	__int128 { KEYWORD_RETURN(INT); } // GCC
234	__label__ { KEYWORD_RETURN(LABEL); } // GCC
235	long { KEYWORD_RETURN(LONG); }
236	lvalue { KEYWORD_RETURN(LVALUE); } // CFA
237	monitor { KEYWORD_RETURN(MONITOR); } // CFA
238	mutex { KEYWORD_RETURN(MUTEX); } // CFA
239	_Noreturn { KEYWORD_RETURN(NORETURN); } // C11
240	__builtin_offsetof { KEYWORD_RETURN(OFFSETOF); } // GCC
241	one_t { NUMERIC_RETURN(ONE_T); } // CFA
242	otype { KEYWORD_RETURN(OTYPE); } // CFA
243	register { KEYWORD_RETURN(REGISTER); }
244	restrict { KEYWORD_RETURN(RESTRICT); } // C99
245	__restrict { KEYWORD_RETURN(RESTRICT); } // GCC
246	__restrict__ { KEYWORD_RETURN(RESTRICT); } // GCC
247	return { KEYWORD_RETURN(RETURN); }
248	short { KEYWORD_RETURN(SHORT); }
249	signed { KEYWORD_RETURN(SIGNED); }
250	__signed { KEYWORD_RETURN(SIGNED); } // GCC
251	__signed__ { KEYWORD_RETURN(SIGNED); } // GCC
252	sizeof { KEYWORD_RETURN(SIZEOF); }
253	static { KEYWORD_RETURN(STATIC); }
254	_Static_assert { KEYWORD_RETURN(STATICASSERT); } // C11
255	struct { KEYWORD_RETURN(STRUCT); }
256	switch { KEYWORD_RETURN(SWITCH); }
257	thread { KEYWORD_RETURN(THREAD); } // C11
258	_Thread_local { KEYWORD_RETURN(THREADLOCAL); } // C11
259	throw { KEYWORD_RETURN(THROW); } // CFA
260	throwResume { KEYWORD_RETURN(THROWRESUME); } // CFA
261	trait { KEYWORD_RETURN(TRAIT); } // CFA
262	try { KEYWORD_RETURN(TRY); } // CFA
263	ttype { KEYWORD_RETURN(TTYPE); } // CFA
264	typedef { KEYWORD_RETURN(TYPEDEF); }
265	typeof { KEYWORD_RETURN(TYPEOF); } // GCC
266	__typeof { KEYWORD_RETURN(TYPEOF); } // GCC
267	__typeof__ { KEYWORD_RETURN(TYPEOF); } // GCC
268	union { KEYWORD_RETURN(UNION); }
269	unsigned { KEYWORD_RETURN(UNSIGNED); }
270	__builtin_va_list { KEYWORD_RETURN(VALIST); } // GCC
271	void { KEYWORD_RETURN(VOID); }
272	volatile { KEYWORD_RETURN(VOLATILE); }
273	__volatile { KEYWORD_RETURN(VOLATILE); } // GCC
274	__volatile__ { KEYWORD_RETURN(VOLATILE); } // GCC
275	while { KEYWORD_RETURN(WHILE); }
276	zero_t { NUMERIC_RETURN(ZERO_T); } // CFA
277
278	/* identifier */
279	{identifier} { IDENTIFIER_RETURN(); }
280	{attr_identifier} { ATTRIBUTE_RETURN(); }
281	"`" { BEGIN BKQUOTE; }
282	<BKQUOTE>{identifier} { IDENTIFIER_RETURN(); }
283	<BKQUOTE>"`" { BEGIN 0; }
284
285	/* numeric constants */
286	"0" { NUMERIC_RETURN(ZERO); } // CFA
287	"1" { NUMERIC_RETURN(ONE); } // CFA
288	{decimal_constant} { NUMERIC_RETURN(INTEGERconstant); }
289	{octal_constant} { NUMERIC_RETURN(INTEGERconstant); }
290	{hex_constant} { NUMERIC_RETURN(INTEGERconstant); }
291	{real_decimal} { NUMERIC_RETURN(REALDECIMALconstant); } // must appear before floating_constant
292	{real_fraction} { NUMERIC_RETURN(REALFRACTIONconstant); } // must appear before floating_constant
293	{floating_constant} { NUMERIC_RETURN(FLOATINGconstant); }
294	{hex_floating_constant} { NUMERIC_RETURN(FLOATINGconstant); }
295
296	/* character constant, allows empty value */
297	({cwide_prefix}[_]?)?['] { BEGIN QUOTE; rm_underscore(); strtext = new std::string( yytext, yyleng ); }
298	<QUOTE>[^'\\\n]* { strtext->append( yytext, yyleng ); }
299	<QUOTE>['\n] { BEGIN 0; strtext->append( yytext, yyleng ); RETURN_STR(CHARACTERconstant); }
300	/* ' stop highlighting */
301
302	/* string constant */
303	({swide_prefix}[_]?)?["] { BEGIN STRING; rm_underscore(); strtext = new std::string( yytext, yyleng ); }
304	<STRING>[^"\\\n]* { strtext->append( yytext, yyleng ); }
305	<STRING>["\n] { BEGIN 0; strtext->append( yytext, yyleng ); RETURN_STR(STRINGliteral); }
306	/* " stop highlighting */
307
308	/* common character/string constant */
309	<QUOTE,STRING>{escape_seq} { rm_underscore(); strtext->append( yytext, yyleng ); }
310	<QUOTE,STRING>"\\"{h_white}*"\n" {} // continuation (ALSO HANDLED BY CPP)
311	<QUOTE,STRING>"\\" { strtext->append( yytext, yyleng ); } // unknown escape character
312
313	/* punctuation */
314	"@" { ASCIIOP_RETURN(); }
315	"[" { ASCIIOP_RETURN(); }
316	"]" { ASCIIOP_RETURN(); }
317	"(" { ASCIIOP_RETURN(); }
318	")" { ASCIIOP_RETURN(); }
319	"{" { ASCIIOP_RETURN(); }
320	"}" { ASCIIOP_RETURN(); }
321	"," { ASCIIOP_RETURN(); } // also operator
322	":" { ASCIIOP_RETURN(); }
323	";" { ASCIIOP_RETURN(); }
324	"." { ASCIIOP_RETURN(); } // also operator
325	"..." { NAMEDOP_RETURN(ELLIPSIS); }
326
327	/* alternative C99 brackets, "<:" & "<:<:" handled by preprocessor */
328	"<:" { RETURN_VAL('['); }
329	":>" { RETURN_VAL(']'); }
330	"<%" { RETURN_VAL('{'); }
331	"%>" { RETURN_VAL('}'); }
332
333	/* operators */
334	"!" { ASCIIOP_RETURN(); }
335	"+" { ASCIIOP_RETURN(); }
336	"-" { ASCIIOP_RETURN(); }
337	"*" { ASCIIOP_RETURN(); }
338	"/" { ASCIIOP_RETURN(); }
339	"%" { ASCIIOP_RETURN(); }
340	"^" { ASCIIOP_RETURN(); }
341	"~" { ASCIIOP_RETURN(); }
342	"&" { ASCIIOP_RETURN(); }
343	"\|" { ASCIIOP_RETURN(); }
344	"<" { ASCIIOP_RETURN(); }
345	">" { ASCIIOP_RETURN(); }
346	"=" { ASCIIOP_RETURN(); }
347	"?" { ASCIIOP_RETURN(); }
348
349	"++" { NAMEDOP_RETURN(ICR); }
350	"--" { NAMEDOP_RETURN(DECR); }
351	"==" { NAMEDOP_RETURN(EQ); }
352	"!=" { NAMEDOP_RETURN(NE); }
353	"<<" { NAMEDOP_RETURN(LS); }
354	">>" { NAMEDOP_RETURN(RS); }
355	"<=" { NAMEDOP_RETURN(LE); }
356	">=" { NAMEDOP_RETURN(GE); }
357	"&&" { NAMEDOP_RETURN(ANDAND); }
358	"\|\|" { NAMEDOP_RETURN(OROR); }
359	"->" { NAMEDOP_RETURN(ARROW); }
360	"+=" { NAMEDOP_RETURN(PLUSassign); }
361	"-=" { NAMEDOP_RETURN(MINUSassign); }
362	"*=" { NAMEDOP_RETURN(MULTassign); }
363	"/=" { NAMEDOP_RETURN(DIVassign); }
364	"%=" { NAMEDOP_RETURN(MODassign); }
365	"&=" { NAMEDOP_RETURN(ANDassign); }
366	"\|=" { NAMEDOP_RETURN(ORassign); }
367	"^=" { NAMEDOP_RETURN(ERassign); }
368	"<<=" { NAMEDOP_RETURN(LSassign); }
369	">>=" { NAMEDOP_RETURN(RSassign); }
370
371	"@=" { NAMEDOP_RETURN(ATassign); } // CFA
372
373	/* CFA, operator identifier */
374	{op_unary}"?" { IDENTIFIER_RETURN(); } // unary
375	"?"({op_unary_pre_post}\|"()"\|"[?]"\|"{}") { IDENTIFIER_RETURN(); }
376	"^?{}" { IDENTIFIER_RETURN(); }
377	"?"{op_binary_over}"?" { IDENTIFIER_RETURN(); } // binary
378	/*
379	This rule handles ambiguous cases with operator identifiers, e.g., "int ??()", where the string "??" can be
380	lexed as "?"/"?" or ""/"??". Since it is common practise to put a unary operator juxtaposed to an identifier,
381	e.g., "*i", users will be annoyed if they cannot do this with respect to operator identifiers. Therefore, there is
382	a lexical look-ahead for the second case, with backtracking to return the leading unary operator and then
383	reparsing the trailing operator identifier. Otherwise a space is needed between the unary operator and operator
384	identifier to disambiguate this common case.
385
386	A similar issue occurs with the dereference, *?(...), and routine-call, ?()(...) identifiers. The ambiguity
387	occurs when the deference operator has no parameters, ?() and ?()(...), requiring arbitrary whitespace
388	look-ahead for the routine-call parameter-list to disambiguate. However, the dereference operator must have a
389	parameter/argument to dereference ?(...). Hence, always interpreting the string ?() as * ?() does not preclude
390	any meaningful program.
391
392	The remaining cases are with the increment/decrement operators and conditional expression:
393
394	i++? ...(...);
395	i?++ ...(...);
396
397	requiring arbitrary whitespace look-ahead for the operator parameter-list, even though that interpretation is an
398	incorrect expression (juxtaposed identifiers). Therefore, it is necessary to disambiguate these cases with a
399	space:
400
401	i++ ? i : 0;
402	i? ++i : 0;
403	*/
404	{op_unary}"?"({op_unary_pre_post}\|"()"\|"[?]"\|{op_binary_over}"?") {
405	// 1 or 2 character unary operator ?
406	int i = yytext[1] == '?' ? 1 : 2;
407	yyless( i ); // put back characters up to first '?'
408	if ( i > 1 ) {
409	NAMEDOP_RETURN( yytext[0] == '+' ? ICR : DECR );
410	} else {
411	ASCIIOP_RETURN();
412	} // if
413	}
414
415	/* unknown characters */
416	. { printf("unknown character(s):\"%s\" on line %d\n", yytext, yylineno); }
417
418	%%
419
420	// Local Variables: //
421	// mode: c++ //
422	// tab-width: 4 //
423	// compile-command: "make install" //
424	// End: //

Note: See TracBrowser for help on using the repository browser.

Download in other formats:

Original Format