src/mdParser/MDParser.g

header
{

#include "antlr/CharScanner.hpp"
#include "utils/StringUtils.hpp"
#include "mdParser/FilenameObserver.hpp"
}

options
  {
  language = "Cpp";
  }
                     
class MDParser extends Parser;

options
{
        k = 3;
        exportVocab = MD;
        buildAST = true;
        codeGenMakeSwitchThreshold = 2;
        codeGenBitsetTestThreshold = 3;
        
}

tokens
{
  COMPONENT   = "component";
  MOLECULE    = "molecule";
  ZCONSTRAINT = "zconstraint";
  RESTRAINT   = "restraint";
  ATOM        = "atom";
  BOND        = "bond";
  BEND        = "bend";
  TORSION     = "torsion";
  INVERSION   = "inversion";
  RIGIDBODY   = "rigidBody";
  CUTOFFGROUP = "cutoffGroup";
  CONSTRAINT  = "constraint";
  DISTANCE    = "distance";
  FRAGMENT    = "fragment";
  MEMBERS     = "members";
  CENTER      = "center";
  SATELLITES  = "satellites";
  POSITION    = "position";
  ORIENTATION = "orientation";
  FLUCQ       = "flucQ";
  RNEMD       = "RNEMD";
  MINIMIZER   = "minimizer";
  ENDBLOCK;
}


mdfile  : (statement)*
        ;

statement : assignment
    | componentblock
    | moleculeblock
    | zconstraintblock
    | restraintblock
    | flucqblock
    | rnemdblock
    | minimizerblock
    ;

assignment  : ID ASSIGNEQUAL^ constant SEMICOLON!
            ;
            
constant    : intConst
                        | floatConst
            | vectorConst
            | ID
            | StringLiteral
            ;

componentblock  : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
                ;
    
zconstraintblock  : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
                  ;

restraintblock  : RESTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
                  ;

flucqblock  : FLUCQ^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
    ;

rnemdblock  : RNEMD^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
    ;

minimizerblock  : MINIMIZER^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
    ;
  
moleculeblock : MOLECULE^ LCURLY! (moleculestatement)*  RCURLY {#RCURLY->setType(ENDBLOCK);}
              ;

moleculestatement : assignment
                  | atomblock
                  | bondblock
                  | bendblock
                  | torsionblock
                  | inversionblock
                  | rigidbodyblock
                  | cutoffgroupblock
                  | fragmentblock
                  | constraintblock
                  ;

atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

atomstatement : assignment
              | POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
              | ORIENTATION^  LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
              ;

                      
bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)?  LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

bondstatement : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!  (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

bendstatement : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

torsionblock  : TORSION^ (LBRACKET! intConst! RBRACKET!)?  LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

torsionstatement  : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

inversionblock  : INVERSION^ (LBRACKET! intConst! RBRACKET!)?  LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

inversionstatement  : assignment
              | CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
              | SATELLITES^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

rigidbodyblock  : RIGIDBODY^  LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
                ;

rigidbodystatement  : assignment
              | MEMBERS^ LPAREN!  inttuple  RPAREN! SEMICOLON!
              ;

cutoffgroupblock  : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
                  ;

cutoffgroupstatement  : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
              ;

fragmentstatement : assignment
              ;

constraintblock : CONSTRAINT^ (LBRACKET! intConst! RBRACKET!)?  LCURLY!(constraintstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

constraintstatement : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

              
doubleNumberTuple   : doubleNumber (COMMA! doubleNumber)* 
              ;
                          
inttuple      : intConst (COMMA! intConst)*
              ;
              
protected
intConst
        :  NUM_INT | NUM_LONG
        ;

protected
doubleNumber  :  
                (intConst | floatConst)
              ;
              
protected
floatConst
        : 
          NUM_FLOAT | NUM_DOUBLE
        ;


/* 
protected
vectorConst
    : 
        LPAREN^ doubleNumber COMMA doubleNumber COMMA doubleNumber RPAREN
    ;
*/

protected
vectorConst
    :
        LPAREN^ doubleNumberTuple RPAREN
    ;


class MDLexer extends Lexer;

options
  {
  k = 3;
  exportVocab = MD;
  testLiterals = false;
  }

tokens {
    DOT;
}

{


        int deferredLineCount;
        FilenameObserver* observer;
        
  public:
        void setObserver(FilenameObserver* osv) {observer = osv;}
  void initDeferredLineCount() { deferredLineCount = 0;}
  void deferredNewline() { 
        deferredLineCount++;
  }

        
  virtual void newline() { 
                for (;deferredLineCount>0;deferredLineCount--) {
                        CharScanner::newline();
                }
    CharScanner::newline();
  }
    
}


// Operators:

ASSIGNEQUAL     : '=' ;
COLON           : ':' ;
COMMA           : ',' ;
QUESTIONMARK    : '?' ;
SEMICOLON       : ';' ;

LPAREN          : '(' ;
RPAREN          : ')' ;
LBRACKET        : '[' ;
RBRACKET        : ']' ;
LCURLY          : '{' ;
RCURLY          : '}' ;

Whitespace  
  : 
    ( // whitespace ignored
      (' ' |'\t' | '\f')
    | // handle newlines
      ( '\r' '\n' // MS
      | '\r'    // Mac
      | '\n'    // Unix 
      ) { newline(); }
    | // handle continuation lines
      ( '\\' '\r' '\n'  // MS
      | '\\' '\r'   // Mac
      | '\\' '\n'   // Unix 
      ) {printf("CPP_parser.g continuation line detected\n");
        deferredNewline();}
    ) 
    {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
  ;

Comment  
  : 
    "/*"   
    ( {LA(2) != '/'}? '*'
    | EndOfLine {deferredNewline();}
    | ~('*'| '\r' | '\n')
    )*
    "*/" {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
  ;

CPPComment
  : 
    "//" (~('\n' | '\r'))* EndOfLine
    {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}                     
  ;

PREPROC_DIRECTIVE
  options{paraphrase = "a line directive";}
  : 
    '#' LineDirective
    {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();} 
  ;

protected 
LineDirective
  :
  {
   deferredLineCount = 0;
  }
    ("line")?  // this would be for if the directive started "#line"
    (Space)+
    n:Decimal { setLine(OpenMD::lexi_cast<int>(n->getText()) - 1); } 
    (Space)+
    (sl:StringLiteral) {std::string filename = sl->getText().substr(1,sl->getText().length()-2); observer->notify(filename);} 
    ((Space)+ Decimal)* // To support cpp flags (GNU)
    EndOfLine
  ;

protected  
Space
  : 
    (' '|'\t'|'\f')
  ;


// Literals:

/*
 * Note that we do NOT handle tri-graphs nor multi-byte sequences.
 */

/*
 * Note that we can't have empty character constants (even though we
 * can have empty strings :-).
 */
CharLiteral
  : 
    '\'' (Escape | ~('\'')) '\''
  ;

/*
 * Can't have raw imbedded newlines in string constants.  Strict reading of
 * the standard gives odd dichotomy between newlines & carriage returns.
 * Go figure.
 */
StringLiteral
  : 
    '"'
    ( Escape
    | 
      ( "\\\r\n"   // MS 
      | "\\\r"     // MAC
      | "\\\n"     // Unix
      ) {deferredNewline();}
    | 
      ~('"'|'\r'|'\n'|'\\')
    )*
    '"'
  ;

protected
EndOfLine
  : 
    ( options{generateAmbigWarnings = false;}:
      "\r\n"  // MS
    | '\r'    // Mac
    | '\n'    // Unix
    )
  ;

/*
 * Handle the various escape sequences.
 *
 * Note carefully that these numeric escape *sequences* are *not* of the
 * same form as the C language numeric *constants*.
 *
 * There is no such thing as a binary numeric escape sequence.
 *
 * Octal escape sequences are either 1, 2, or 3 octal digits exactly.
 *
 * There is no such thing as a decimal escape sequence.
 *
 * Hexadecimal escape sequences are begun with a leading \x and continue
 * until a non-hexadecimal character is found.
 *
 * No real handling of tri-graph sequences, yet.
 */

protected
Escape  
  : 
    '\\'
    ( options{warnWhenFollowAmbig=false;}:
      'a'
    | 'b'
    | 'f'
    | 'n'
    | 'r'
    | 't'
    | 'v'
    | '"'
    | '\''
    | '\\'
    | '?'
    | ('0'..'3') (options{warnWhenFollowAmbig=false;}: Digit (options{warnWhenFollowAmbig=false;}: Digit)? )?
    | ('4'..'7') (options{warnWhenFollowAmbig=false;}: Digit)?
    | 'x' (options{warnWhenFollowAmbig=false;}: Digit | 'a'..'f' | 'A'..'F')+
    )
  ;


protected
Vocabulary
  : 
    '\3'..'\377'
  ;


ID
  options {testLiterals = true;}
  : 
    ('a'..'z'|'A'..'Z'|'_')
    ('a'..'z'|'A'..'Z'|'_'|'0'..'9')*
  ;


protected
Digit
  : 
    '0'..'9'
  ;

protected
Decimal
  : 
    ('0'..'9')+
  ;

// hexadecimal digit (again, note it's protected!)
protected
HEX_DIGIT
        :       ('0'..'9'|'A'..'F'|'a'..'f')
        ;


// a numeric literal
NUM_INT
        {
                bool isDecimal = false;
                ANTLR_USE_NAMESPACE(antlr)RefToken t = ANTLR_USE_NAMESPACE(antlr)nullToken;
        }
    : ('+' | '-')?
    (
      '.' {_ttype = DOT;}
            (   ('0'..'9')+ (EXPONENT)? (f1:FLOAT_SUFFIX {t=f1;})?
            {
                                        if ( t &&
                                                  (t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos ||
                                                        t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
                                                _ttype = NUM_FLOAT;
                                        }
                                        else {
                                                _ttype = NUM_DOUBLE; // assume double
                                        }
                                }
            )?

        |       (       '0' {isDecimal = true;} // special case for just '0'
                        (       ('x'|'X')
                                (                                                                                       // hex
                                        // the 'e'|'E' and float suffix stuff look
                                        // like hex digits, hence the (...)+ doesn't
                                        // know when to stop: ambig.  ANTLR resolves
                                        // it correctly by matching immediately.  It
                                        // is therefor ok to hush warning.
                                        options {
                                                warnWhenFollowAmbig=false;
                                        }
                                :       HEX_DIGIT
                                )+
                        |       //float or double with leading zero
                                (('0'..'9')+ ('.'|EXPONENT|FLOAT_SUFFIX)) => ('0'..'9')+
                        |       ('0'..'7')+                                                                     // octal
                        )?
                |       ('1'..'9') ('0'..'9')*  {isDecimal=true;}               // non-zero decimal
                )
                (       ('l'|'L') { _ttype = NUM_LONG; }

                // only check to see if it's a float if looks like decimal so far
                |       {isDecimal}?
            (   '.' ('0'..'9')* (EXPONENT)? (f2:FLOAT_SUFFIX {t=f2;})?
            |   EXPONENT (f3:FLOAT_SUFFIX {t=f3;})?
            |   f4:FLOAT_SUFFIX {t=f4;}
            )
            {
                                        if ( t &&
                                                  (t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos ||
                                                        t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
                                                _ttype = NUM_FLOAT;
                                        }
                                        else {
                                                _ttype = NUM_DOUBLE; // assume double
                                        }
                                }
        )?
  )
        ;

// a couple protected methods to assist in matching floating point numbers
protected
EXPONENT
        :       ('e'|'E'|'d'|'D') ('+'|'-')? ('0'..'9')+
        ;

protected
FLOAT_SUFFIX
        :       'f'|'F'|'d'|'D'
        ;
Revision:	2026
Committed:	Wed Oct 22 12:23:59 2014 UTC (10 years, 9 months ago) by gezelter
File size:	11769 byte(s)
Log Message:	Starting to add support for UniformGradient. Changed Vector3d input type to a more general std::vector<RealType> input. This change alters RNEMD and UniformField inputs.
#	Content
1	header
2	{
3
4	#include "antlr/CharScanner.hpp"
5	#include "utils/StringUtils.hpp"
6	#include "mdParser/FilenameObserver.hpp"
7	}
8
9	options
10	{
11	language = "Cpp";
12	}
13
14	class MDParser extends Parser;
15
16	options
17	{
18	k = 3;
19	exportVocab = MD;
20	buildAST = true;
21	codeGenMakeSwitchThreshold = 2;
22	codeGenBitsetTestThreshold = 3;
23
24	}
25
26	tokens
27	{
28	COMPONENT = "component";
29	MOLECULE = "molecule";
30	ZCONSTRAINT = "zconstraint";
31	RESTRAINT = "restraint";
32	ATOM = "atom";
33	BOND = "bond";
34	BEND = "bend";
35	TORSION = "torsion";
36	INVERSION = "inversion";
37	RIGIDBODY = "rigidBody";
38	CUTOFFGROUP = "cutoffGroup";
39	CONSTRAINT = "constraint";
40	DISTANCE = "distance";
41	FRAGMENT = "fragment";
42	MEMBERS = "members";
43	CENTER = "center";
44	SATELLITES = "satellites";
45	POSITION = "position";
46	ORIENTATION = "orientation";
47	FLUCQ = "flucQ";
48	RNEMD = "RNEMD";
49	MINIMIZER = "minimizer";
50	ENDBLOCK;
51	}
52
53
54	mdfile : (statement)*
55	;
56
57	statement : assignment
58	\| componentblock
59	\| moleculeblock
60	\| zconstraintblock
61	\| restraintblock
62	\| flucqblock
63	\| rnemdblock
64	\| minimizerblock
65	;
66
67	assignment : ID ASSIGNEQUAL^ constant SEMICOLON!
68	;
69
70	constant : intConst
71	\| floatConst
72	\| vectorConst
73	\| ID
74	\| StringLiteral
75	;
76
77	componentblock : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
78	;
79
80	zconstraintblock : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
81	;
82
83	restraintblock : RESTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
84	;
85
86	flucqblock : FLUCQ^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
87	;
88
89	rnemdblock : RNEMD^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
90	;
91
92	minimizerblock : MINIMIZER^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
93	;
94
95	moleculeblock : MOLECULE^ LCURLY! (moleculestatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
96	;
97
98	moleculestatement : assignment
99	\| atomblock
100	\| bondblock
101	\| bendblock
102	\| torsionblock
103	\| inversionblock
104	\| rigidbodyblock
105	\| cutoffgroupblock
106	\| fragmentblock
107	\| constraintblock
108	;
109
110	atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
111	;
112
113	atomstatement : assignment
114	\| POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
115	\| ORIENTATION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
116	;
117
118
119	bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
120	;
121
122	bondstatement : assignment
123	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
124	;
125
126	bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
127	;
128
129	bendstatement : assignment
130	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
131	;
132
133	torsionblock : TORSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
134	;
135
136	torsionstatement : assignment
137	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
138	;
139
140	inversionblock : INVERSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
141	;
142
143	inversionstatement : assignment
144	\| CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
145	\| SATELLITES^ LPAREN! inttuple RPAREN! SEMICOLON!
146	;
147
148	rigidbodyblock : RIGIDBODY^ LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
149	;
150
151	rigidbodystatement : assignment
152	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
153	;
154
155	cutoffgroupblock : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
156	;
157
158	cutoffgroupstatement : assignment
159	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
160	;
161
162	fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
163	;
164
165	fragmentstatement : assignment
166	;
167
168	constraintblock : CONSTRAINT^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(constraintstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
169	;
170
171	constraintstatement : assignment
172	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
173	;
174
175
176	doubleNumberTuple : doubleNumber (COMMA! doubleNumber)*
177	;
178
179	inttuple : intConst (COMMA! intConst)*
180	;
181
182	protected
183	intConst
184	: NUM_INT \| NUM_LONG
185	;
186
187	protected
188	doubleNumber :
189	(intConst \| floatConst)
190	;
191
192	protected
193	floatConst
194	:
195	NUM_FLOAT \| NUM_DOUBLE
196	;
197
198
199	/*
200	protected
201	vectorConst
202	:
203	LPAREN^ doubleNumber COMMA doubleNumber COMMA doubleNumber RPAREN
204	;
205	*/
206
207	protected
208	vectorConst
209	:
210	LPAREN^ doubleNumberTuple RPAREN
211	;
212
213
214	class MDLexer extends Lexer;
215
216	options
217	{
218	k = 3;
219	exportVocab = MD;
220	testLiterals = false;
221	}
222
223	tokens {
224	DOT;
225	}
226
227	{
228
229
230	int deferredLineCount;
231	FilenameObserver* observer;
232
233	public:
234	void setObserver(FilenameObserver* osv) {observer = osv;}
235	void initDeferredLineCount() { deferredLineCount = 0;}
236	void deferredNewline() {
237	deferredLineCount++;
238	}
239
240
241	virtual void newline() {
242	for (;deferredLineCount>0;deferredLineCount--) {
243	CharScanner::newline();
244	}
245	CharScanner::newline();
246	}
247
248	}
249
250
251	// Operators:
252
253	ASSIGNEQUAL : '=' ;
254	COLON : ':' ;
255	COMMA : ',' ;
256	QUESTIONMARK : '?' ;
257	SEMICOLON : ';' ;
258
259	LPAREN : '(' ;
260	RPAREN : ')' ;
261	LBRACKET : '[' ;
262	RBRACKET : ']' ;
263	LCURLY : '{' ;
264	RCURLY : '}' ;
265
266	Whitespace
267	:
268	( // whitespace ignored
269	(' ' \|'\t' \| '\f')
270	\| // handle newlines
271	( '\r' '\n' // MS
272	\| '\r' // Mac
273	\| '\n' // Unix
274	) { newline(); }
275	\| // handle continuation lines
276	( '\\' '\r' '\n' // MS
277	\| '\\' '\r' // Mac
278	\| '\\' '\n' // Unix
279	) {printf("CPP_parser.g continuation line detected\n");
280	deferredNewline();}
281	)
282	{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
283	;
284
285	Comment
286	:
287	"/*"
288	( {LA(2) != '/'}? '*'
289	\| EndOfLine {deferredNewline();}
290	\| ~('*'\| '\r' \| '\n')
291	)*
292	"*/" {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
293	;
294
295	CPPComment
296	:
297	"//" (~('\n' \| '\r'))* EndOfLine
298	{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
299	;
300
301	PREPROC_DIRECTIVE
302	options{paraphrase = "a line directive";}
303	:
304	'#' LineDirective
305	{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
306	;
307
308	protected
309	LineDirective
310	:
311	{
312	deferredLineCount = 0;
313	}
314	("line")? // this would be for if the directive started "#line"
315	(Space)+
316	n:Decimal { setLine(OpenMD::lexi_cast<int>(n->getText()) - 1); }
317	(Space)+
318	(sl:StringLiteral) {std::string filename = sl->getText().substr(1,sl->getText().length()-2); observer->notify(filename);}
319	((Space)+ Decimal)* // To support cpp flags (GNU)
320	EndOfLine
321	;
322
323	protected
324	Space
325	:
326	(' '\|'\t'\|'\f')
327	;
328
329
330	// Literals:
331
332	/*
333	* Note that we do NOT handle tri-graphs nor multi-byte sequences.
334	*/
335
336	/*
337	* Note that we can't have empty character constants (even though we
338	* can have empty strings :-).
339	*/
340	CharLiteral
341	:
342	'\'' (Escape \| ~('\'')) '\''
343	;
344
345	/*
346	* Can't have raw imbedded newlines in string constants. Strict reading of
347	* the standard gives odd dichotomy between newlines & carriage returns.
348	* Go figure.
349	*/
350	StringLiteral
351	:
352	'"'
353	( Escape
354	\|
355	( "\\\r\n" // MS
356	\| "\\\r" // MAC
357	\| "\\\n" // Unix
358	) {deferredNewline();}
359	\|
360	~('"'\|'\r'\|'\n'\|'\\')
361	)*
362	'"'
363	;
364
365	protected
366	EndOfLine
367	:
368	( options{generateAmbigWarnings = false;}:
369	"\r\n" // MS
370	\| '\r' // Mac
371	\| '\n' // Unix
372	)
373	;
374
375	/*
376	* Handle the various escape sequences.
377	*
378	* Note carefully that these numeric escape sequences are not of the
379	* same form as the C language numeric constants.
380	*
381	* There is no such thing as a binary numeric escape sequence.
382	*
383	* Octal escape sequences are either 1, 2, or 3 octal digits exactly.
384	*
385	* There is no such thing as a decimal escape sequence.
386	*
387	* Hexadecimal escape sequences are begun with a leading \x and continue
388	* until a non-hexadecimal character is found.
389	*
390	* No real handling of tri-graph sequences, yet.
391	*/
392
393	protected
394	Escape
395	:
396	'\\'
397	( options{warnWhenFollowAmbig=false;}:
398	'a'
399	\| 'b'
400	\| 'f'
401	\| 'n'
402	\| 'r'
403	\| 't'
404	\| 'v'
405	\| '"'
406	\| '\''
407	\| '\\'
408	\| '?'
409	\| ('0'..'3') (options{warnWhenFollowAmbig=false;}: Digit (options{warnWhenFollowAmbig=false;}: Digit)? )?
410	\| ('4'..'7') (options{warnWhenFollowAmbig=false;}: Digit)?
411	\| 'x' (options{warnWhenFollowAmbig=false;}: Digit \| 'a'..'f' \| 'A'..'F')+
412	)
413	;
414
415
416	protected
417	Vocabulary
418	:
419	'\3'..'\377'
420	;
421
422
423	ID
424	options {testLiterals = true;}
425	:
426	('a'..'z'\|'A'..'Z'\|'_')
427	('a'..'z'\|'A'..'Z'\|'_'\|'0'..'9')*
428	;
429
430
431	protected
432	Digit
433	:
434	'0'..'9'
435	;
436
437	protected
438	Decimal
439	:
440	('0'..'9')+
441	;
442
443	// hexadecimal digit (again, note it's protected!)
444	protected
445	HEX_DIGIT
446	: ('0'..'9'\|'A'..'F'\|'a'..'f')
447	;
448
449
450	// a numeric literal
451	NUM_INT
452	{
453	bool isDecimal = false;
454	ANTLR_USE_NAMESPACE(antlr)RefToken t = ANTLR_USE_NAMESPACE(antlr)nullToken;
455	}
456	: ('+' \| '-')?
457	(
458	'.' {_ttype = DOT;}
459	( ('0'..'9')+ (EXPONENT)? (f1:FLOAT_SUFFIX {t=f1;})?
460	{
461	if ( t &&
462	(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
463	t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
464	_ttype = NUM_FLOAT;
465	}
466	else {
467	_ttype = NUM_DOUBLE; // assume double
468	}
469	}
470	)?
471
472	\| ( '0' {isDecimal = true;} // special case for just '0'
473	( ('x'\|'X')
474	( // hex
475	// the 'e'\|'E' and float suffix stuff look
476	// like hex digits, hence the (...)+ doesn't
477	// know when to stop: ambig. ANTLR resolves
478	// it correctly by matching immediately. It
479	// is therefor ok to hush warning.
480	options {
481	warnWhenFollowAmbig=false;
482	}
483	: HEX_DIGIT
484	)+
485	\| //float or double with leading zero
486	(('0'..'9')+ ('.'\|EXPONENT\|FLOAT_SUFFIX)) => ('0'..'9')+
487	\| ('0'..'7')+ // octal
488	)?
489	\| ('1'..'9') ('0'..'9')* {isDecimal=true;} // non-zero decimal
490	)
491	( ('l'\|'L') { _ttype = NUM_LONG; }
492
493	// only check to see if it's a float if looks like decimal so far
494	\| {isDecimal}?
495	( '.' ('0'..'9')* (EXPONENT)? (f2:FLOAT_SUFFIX {t=f2;})?
496	\| EXPONENT (f3:FLOAT_SUFFIX {t=f3;})?
497	\| f4:FLOAT_SUFFIX {t=f4;}
498	)
499	{
500	if ( t &&
501	(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
502	t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
503	_ttype = NUM_FLOAT;
504	}
505	else {
506	_ttype = NUM_DOUBLE; // assume double
507	}
508	}
509	)?
510	)
511	;
512
513	// a couple protected methods to assist in matching floating point numbers
514	protected
515	EXPONENT
516	: ('e'\|'E'\|'d'\|'D') ('+'\|'-')? ('0'..'9')+
517	;
518
519	protected
520	FLOAT_SUFFIX
521	: 'f'\|'F'\|'d'\|'D'
522	;