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";
  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
                  ;

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
              ;


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
    ;

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:	1902
Committed:	Tue Jul 16 17:16:45 2013 UTC (11 years, 10 months ago) by gezelter
File size:	11336 byte(s)
Log Message:	Added a satellites keyword to the inversion block.
#	User	Rev	Content
1	tim	770	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	cli2	1360	RESTRAINT = "restraint";
32	tim	770	ATOM = "atom";
33			BOND = "bond";
34			BEND = "bend";
35			TORSION = "torsion";
36	cli2	1275	INVERSION = "inversion";
37	tim	770	RIGIDBODY = "rigidBody";
38			CUTOFFGROUP = "cutoffGroup";
39			FRAGMENT = "fragment";
40			MEMBERS = "members";
41	cli2	1275	CENTER = "center";
42	gezelter	1902	SATELLITES = "satellites";
43	tim	770	POSITION = "position";
44			ORIENTATION = "orientation";
45	gezelter	1782	FLUCQ = "flucQ";
46			RNEMD = "RNEMD";
47			MINIMIZER = "minimizer";
48	tim	770	ENDBLOCK;
49			}
50
51
52			mdfile : (statement)*
53			;
54
55			statement : assignment
56	gezelter	1782	\| componentblock
57			\| moleculeblock
58			\| zconstraintblock
59			\| restraintblock
60			\| flucqblock
61			\| rnemdblock
62			\| minimizerblock
63			;
64
65	tim	770	assignment : ID ASSIGNEQUAL^ constant SEMICOLON!
66			;
67
68	tim	814	constant : intConst
69	gezelter	1782	\| floatConst
70			\| vectorConst
71	tim	770	\| ID
72			\| StringLiteral
73			;
74
75			componentblock : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
76			;
77
78			zconstraintblock : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
79			;
80	cli2	1360
81			restraintblock : RESTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
82			;
83	gezelter	1782
84			flucqblock : FLUCQ^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
85			;
86
87			rnemdblock : RNEMD^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
88			;
89
90			minimizerblock : MINIMIZER^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
91			;
92	tim	770
93			moleculeblock : MOLECULE^ LCURLY! (moleculestatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
94			;
95
96			moleculestatement : assignment
97			\| atomblock
98			\| bondblock
99			\| bendblock
100			\| torsionblock
101	gezelter	1277	\| inversionblock
102	tim	770	\| rigidbodyblock
103			\| cutoffgroupblock
104			\| fragmentblock
105			;
106
107			atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
108			;
109
110			atomstatement : assignment
111	tim	814	\| POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
112			\| ORIENTATION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
113	tim	770	;
114
115
116			bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
117			;
118
119			bondstatement : assignment
120			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
121			;
122
123			bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
124			;
125
126			bendstatement : assignment
127			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
128			;
129
130			torsionblock : TORSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
131			;
132
133			torsionstatement : assignment
134			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
135			;
136
137	cli2	1275	inversionblock : INVERSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
138			;
139
140			inversionstatement : assignment
141			\| CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
142	gezelter	1902	\| SATELLITES^ LPAREN! inttuple RPAREN! SEMICOLON!
143	cli2	1275	;
144
145	tim	770	rigidbodyblock : RIGIDBODY^ LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
146			;
147
148			rigidbodystatement : assignment
149			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
150			;
151
152			cutoffgroupblock : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
153			;
154
155			cutoffgroupstatement : assignment
156			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
157			;
158
159			fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
160			;
161
162			fragmentstatement : assignment
163			;
164
165
166
167	tim	814	doubleNumberTuple : doubleNumber (COMMA! doubleNumber)*
168	tim	770	;
169
170			inttuple : intConst (COMMA! intConst)*
171			;
172
173			protected
174			intConst
175	tim	814	: NUM_INT \| NUM_LONG
176	tim	770	;
177
178			protected
179	tim	814	doubleNumber :
180	tim	770	(intConst \| floatConst)
181			;
182
183			protected
184			floatConst
185			:
186	tim	814	NUM_FLOAT \| NUM_DOUBLE
187	tim	770	;
188
189	gezelter	1782	protected
190			vectorConst
191			:
192			LPAREN^ doubleNumber COMMA doubleNumber COMMA doubleNumber RPAREN
193			;
194	tim	770
195			class MDLexer extends Lexer;
196
197			options
198			{
199			k = 3;
200			exportVocab = MD;
201			testLiterals = false;
202			}
203
204			tokens {
205			DOT;
206			}
207
208			{
209
210
211			int deferredLineCount;
212			FilenameObserver* observer;
213
214			public:
215			void setObserver(FilenameObserver* osv) {observer = osv;}
216			void initDeferredLineCount() { deferredLineCount = 0;}
217			void deferredNewline() {
218			deferredLineCount++;
219			}
220
221
222			virtual void newline() {
223			for (;deferredLineCount>0;deferredLineCount--) {
224			CharScanner::newline();
225			}
226			CharScanner::newline();
227			}
228
229			}
230
231
232			// Operators:
233
234			ASSIGNEQUAL : '=' ;
235			COLON : ':' ;
236			COMMA : ',' ;
237			QUESTIONMARK : '?' ;
238			SEMICOLON : ';' ;
239
240			LPAREN : '(' ;
241			RPAREN : ')' ;
242			LBRACKET : '[' ;
243			RBRACKET : ']' ;
244			LCURLY : '{' ;
245			RCURLY : '}' ;
246
247			Whitespace
248			:
249			( // whitespace ignored
250			(' ' \|'\t' \| '\f')
251			\| // handle newlines
252			( '\r' '\n' // MS
253			\| '\r' // Mac
254			\| '\n' // Unix
255			) { newline(); }
256			\| // handle continuation lines
257			( '\\' '\r' '\n' // MS
258			\| '\\' '\r' // Mac
259			\| '\\' '\n' // Unix
260			) {printf("CPP_parser.g continuation line detected\n");
261			deferredNewline();}
262			)
263			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
264			;
265
266			Comment
267			:
268			"/*"
269			( {LA(2) != '/'}? '*'
270			\| EndOfLine {deferredNewline();}
271			\| ~('*'\| '\r' \| '\n')
272			)*
273			"*/" {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
274			;
275
276			CPPComment
277			:
278			"//" (~('\n' \| '\r'))* EndOfLine
279			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
280			;
281
282			PREPROC_DIRECTIVE
283			options{paraphrase = "a line directive";}
284			:
285			'#' LineDirective
286			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
287			;
288
289			protected
290			LineDirective
291			:
292			{
293			deferredLineCount = 0;
294			}
295			("line")? // this would be for if the directive started "#line"
296			(Space)+
297	gezelter	1390	n:Decimal { setLine(OpenMD::lexi_cast<int>(n->getText()) - 1); }
298	tim	770	(Space)+
299			(sl:StringLiteral) {std::string filename = sl->getText().substr(1,sl->getText().length()-2); observer->notify(filename);}
300			((Space)+ Decimal)* // To support cpp flags (GNU)
301			EndOfLine
302			;
303
304			protected
305			Space
306			:
307			(' '\|'\t'\|'\f')
308			;
309
310
311			// Literals:
312
313			/*
314			* Note that we do NOT handle tri-graphs nor multi-byte sequences.
315			*/
316
317			/*
318			* Note that we can't have empty character constants (even though we
319			* can have empty strings :-).
320			*/
321			CharLiteral
322			:
323			'\'' (Escape \| ~('\'')) '\''
324			;
325
326			/*
327			* Can't have raw imbedded newlines in string constants. Strict reading of
328			* the standard gives odd dichotomy between newlines & carriage returns.
329			* Go figure.
330			*/
331			StringLiteral
332			:
333			'"'
334			( Escape
335			\|
336			( "\\\r\n" // MS
337			\| "\\\r" // MAC
338			\| "\\\n" // Unix
339			) {deferredNewline();}
340			\|
341			~('"'\|'\r'\|'\n'\|'\\')
342			)*
343			'"'
344			;
345
346			protected
347			EndOfLine
348			:
349			( options{generateAmbigWarnings = false;}:
350			"\r\n" // MS
351			\| '\r' // Mac
352			\| '\n' // Unix
353			)
354			;
355
356			/*
357			* Handle the various escape sequences.
358			*
359			* Note carefully that these numeric escape sequences are not of the
360			* same form as the C language numeric constants.
361			*
362			* There is no such thing as a binary numeric escape sequence.
363			*
364			* Octal escape sequences are either 1, 2, or 3 octal digits exactly.
365			*
366			* There is no such thing as a decimal escape sequence.
367			*
368			* Hexadecimal escape sequences are begun with a leading \x and continue
369			* until a non-hexadecimal character is found.
370			*
371			* No real handling of tri-graph sequences, yet.
372			*/
373
374			protected
375			Escape
376			:
377			'\\'
378			( options{warnWhenFollowAmbig=false;}:
379			'a'
380			\| 'b'
381			\| 'f'
382			\| 'n'
383			\| 'r'
384			\| 't'
385			\| 'v'
386			\| '"'
387			\| '\''
388			\| '\\'
389			\| '?'
390			\| ('0'..'3') (options{warnWhenFollowAmbig=false;}: Digit (options{warnWhenFollowAmbig=false;}: Digit)? )?
391			\| ('4'..'7') (options{warnWhenFollowAmbig=false;}: Digit)?
392			\| 'x' (options{warnWhenFollowAmbig=false;}: Digit \| 'a'..'f' \| 'A'..'F')+
393			)
394			;
395
396
397			protected
398	tim	814	Vocabulary
399			:
400			'\3'..'\377'
401			;
402
403
404			ID
405			options {testLiterals = true;}
406			:
407			('a'..'z'\|'A'..'Z'\|'_')
408			('a'..'z'\|'A'..'Z'\|'_'\|'0'..'9')*
409			;
410
411
412			protected
413	tim	770	Digit
414			:
415			'0'..'9'
416			;
417
418			protected
419			Decimal
420			:
421			('0'..'9')+
422			;
423
424	tim	814	// hexadecimal digit (again, note it's protected!)
425	tim	770	protected
426	tim	814	HEX_DIGIT
427			: ('0'..'9'\|'A'..'F'\|'a'..'f')
428			;
429	tim	770
430
431	tim	814	// a numeric literal
432			NUM_INT
433			{
434			bool isDecimal = false;
435			ANTLR_USE_NAMESPACE(antlr)RefToken t = ANTLR_USE_NAMESPACE(antlr)nullToken;
436			}
437			: ('+' \| '-')?
438			(
439			'.' {_ttype = DOT;}
440			( ('0'..'9')+ (EXPONENT)? (f1:FLOAT_SUFFIX {t=f1;})?
441			{
442			if ( t &&
443			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
444			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
445			_ttype = NUM_FLOAT;
446			}
447			else {
448			_ttype = NUM_DOUBLE; // assume double
449			}
450			}
451			)?
452	tim	770
453	tim	814	\| ( '0' {isDecimal = true;} // special case for just '0'
454			( ('x'\|'X')
455			( // hex
456			// the 'e'\|'E' and float suffix stuff look
457			// like hex digits, hence the (...)+ doesn't
458			// know when to stop: ambig. ANTLR resolves
459			// it correctly by matching immediately. It
460			// is therefor ok to hush warning.
461			options {
462			warnWhenFollowAmbig=false;
463			}
464			: HEX_DIGIT
465			)+
466			\| //float or double with leading zero
467			(('0'..'9')+ ('.'\|EXPONENT\|FLOAT_SUFFIX)) => ('0'..'9')+
468			\| ('0'..'7')+ // octal
469			)?
470			\| ('1'..'9') ('0'..'9')* {isDecimal=true;} // non-zero decimal
471			)
472			( ('l'\|'L') { _ttype = NUM_LONG; }
473
474			// only check to see if it's a float if looks like decimal so far
475			\| {isDecimal}?
476			( '.' ('0'..'9')* (EXPONENT)? (f2:FLOAT_SUFFIX {t=f2;})?
477			\| EXPONENT (f3:FLOAT_SUFFIX {t=f3;})?
478			\| f4:FLOAT_SUFFIX {t=f4;}
479			)
480			{
481			if ( t &&
482			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
483			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
484			_ttype = NUM_FLOAT;
485			}
486			else {
487			_ttype = NUM_DOUBLE; // assume double
488			}
489			}
490			)?
491			)
492			;
493
494			// a couple protected methods to assist in matching floating point numbers
495	tim	770	protected
496	tim	814	EXPONENT
497	gezelter	848	: ('e'\|'E'\|'d'\|'D') ('+'\|'-')? ('0'..'9')+
498	tim	814	;
499	tim	770
500			protected
501	tim	814	FLOAT_SUFFIX
502			: 'f'\|'F'\|'d'\|'D'
503			;