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";
  POSITION    = "position";
  ORIENTATION = "orientation";
  ENDBLOCK;
}


mdfile  : (statement)*
        ;

statement : assignment
          | componentblock
          | moleculeblock
          | zconstraintblock
          | restraintblock
          ;
            
assignment  : ID ASSIGNEQUAL^ constant SEMICOLON!
            ;
            
constant    : intConst
                                                | floatConst
            | 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);}
                  ;
  
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!
              ;

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
        ;


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:	1465
Committed:	Fri Jul 9 23:08:25 2010 UTC (14 years, 9 months ago) by chuckv
File size:	10750 byte(s)
Log Message:	Creating busticated version of OpenMD
#	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	tim	770	POSITION = "position";
43			ORIENTATION = "orientation";
44			ENDBLOCK;
45			}
46
47
48			mdfile : (statement)*
49			;
50
51			statement : assignment
52			\| componentblock
53			\| moleculeblock
54			\| zconstraintblock
55	cli2	1360	\| restraintblock
56	tim	770	;
57
58			assignment : ID ASSIGNEQUAL^ constant SEMICOLON!
59			;
60
61	tim	814	constant : intConst
62			\| floatConst
63	tim	770	\| ID
64			\| StringLiteral
65			;
66
67			componentblock : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
68			;
69
70			zconstraintblock : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
71			;
72	cli2	1360
73			restraintblock : RESTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
74			;
75	tim	770
76			moleculeblock : MOLECULE^ LCURLY! (moleculestatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
77			;
78
79			moleculestatement : assignment
80			\| atomblock
81			\| bondblock
82			\| bendblock
83			\| torsionblock
84	gezelter	1277	\| inversionblock
85	tim	770	\| rigidbodyblock
86			\| cutoffgroupblock
87			\| fragmentblock
88			;
89
90			atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
91			;
92
93			atomstatement : assignment
94	tim	814	\| POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
95			\| ORIENTATION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
96	tim	770	;
97
98
99			bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
100			;
101
102			bondstatement : assignment
103			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
104			;
105
106			bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
107			;
108
109			bendstatement : assignment
110			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
111			;
112
113			torsionblock : TORSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
114			;
115
116			torsionstatement : assignment
117			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
118			;
119
120	cli2	1275	inversionblock : INVERSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
121			;
122
123			inversionstatement : assignment
124			\| CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
125			;
126
127	tim	770	rigidbodyblock : RIGIDBODY^ LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
128			;
129
130			rigidbodystatement : assignment
131			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
132			;
133
134			cutoffgroupblock : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
135			;
136
137			cutoffgroupstatement : assignment
138			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
139			;
140
141			fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
142			;
143
144			fragmentstatement : assignment
145			;
146
147
148
149	tim	814	doubleNumberTuple : doubleNumber (COMMA! doubleNumber)*
150	tim	770	;
151
152			inttuple : intConst (COMMA! intConst)*
153			;
154
155			protected
156			intConst
157	tim	814	: NUM_INT \| NUM_LONG
158	tim	770	;
159
160			protected
161	tim	814	doubleNumber :
162	tim	770	(intConst \| floatConst)
163			;
164
165			protected
166			floatConst
167			:
168	tim	814	NUM_FLOAT \| NUM_DOUBLE
169	tim	770	;
170
171
172
173			class MDLexer extends Lexer;
174
175			options
176			{
177			k = 3;
178			exportVocab = MD;
179			testLiterals = false;
180			}
181
182			tokens {
183			DOT;
184			}
185
186			{
187
188
189			int deferredLineCount;
190			FilenameObserver* observer;
191
192			public:
193			void setObserver(FilenameObserver* osv) {observer = osv;}
194			void initDeferredLineCount() { deferredLineCount = 0;}
195			void deferredNewline() {
196			deferredLineCount++;
197			}
198
199
200			virtual void newline() {
201			for (;deferredLineCount>0;deferredLineCount--) {
202			CharScanner::newline();
203			}
204			CharScanner::newline();
205			}
206
207			}
208
209
210			// Operators:
211
212			ASSIGNEQUAL : '=' ;
213			COLON : ':' ;
214			COMMA : ',' ;
215			QUESTIONMARK : '?' ;
216			SEMICOLON : ';' ;
217
218			LPAREN : '(' ;
219			RPAREN : ')' ;
220			LBRACKET : '[' ;
221			RBRACKET : ']' ;
222			LCURLY : '{' ;
223			RCURLY : '}' ;
224
225			Whitespace
226			:
227			( // whitespace ignored
228			(' ' \|'\t' \| '\f')
229			\| // handle newlines
230			( '\r' '\n' // MS
231			\| '\r' // Mac
232			\| '\n' // Unix
233			) { newline(); }
234			\| // handle continuation lines
235			( '\\' '\r' '\n' // MS
236			\| '\\' '\r' // Mac
237			\| '\\' '\n' // Unix
238			) {printf("CPP_parser.g continuation line detected\n");
239			deferredNewline();}
240			)
241			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
242			;
243
244			Comment
245			:
246			"/*"
247			( {LA(2) != '/'}? '*'
248			\| EndOfLine {deferredNewline();}
249			\| ~('*'\| '\r' \| '\n')
250			)*
251			"*/" {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
252			;
253
254			CPPComment
255			:
256			"//" (~('\n' \| '\r'))* EndOfLine
257			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
258			;
259
260			PREPROC_DIRECTIVE
261			options{paraphrase = "a line directive";}
262			:
263			'#' LineDirective
264			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
265			;
266
267			protected
268			LineDirective
269			:
270			{
271			deferredLineCount = 0;
272			}
273			("line")? // this would be for if the directive started "#line"
274			(Space)+
275	gezelter	1390	n:Decimal { setLine(OpenMD::lexi_cast<int>(n->getText()) - 1); }
276	tim	770	(Space)+
277			(sl:StringLiteral) {std::string filename = sl->getText().substr(1,sl->getText().length()-2); observer->notify(filename);}
278			((Space)+ Decimal)* // To support cpp flags (GNU)
279			EndOfLine
280			;
281
282			protected
283			Space
284			:
285			(' '\|'\t'\|'\f')
286			;
287
288
289			// Literals:
290
291			/*
292			* Note that we do NOT handle tri-graphs nor multi-byte sequences.
293			*/
294
295			/*
296			* Note that we can't have empty character constants (even though we
297			* can have empty strings :-).
298			*/
299			CharLiteral
300			:
301			'\'' (Escape \| ~('\'')) '\''
302			;
303
304			/*
305			* Can't have raw imbedded newlines in string constants. Strict reading of
306			* the standard gives odd dichotomy between newlines & carriage returns.
307			* Go figure.
308			*/
309			StringLiteral
310			:
311			'"'
312			( Escape
313			\|
314			( "\\\r\n" // MS
315			\| "\\\r" // MAC
316			\| "\\\n" // Unix
317			) {deferredNewline();}
318			\|
319			~('"'\|'\r'\|'\n'\|'\\')
320			)*
321			'"'
322			;
323
324			protected
325			EndOfLine
326			:
327			( options{generateAmbigWarnings = false;}:
328			"\r\n" // MS
329			\| '\r' // Mac
330			\| '\n' // Unix
331			)
332			;
333
334			/*
335			* Handle the various escape sequences.
336			*
337			* Note carefully that these numeric escape sequences are not of the
338			* same form as the C language numeric constants.
339			*
340			* There is no such thing as a binary numeric escape sequence.
341			*
342			* Octal escape sequences are either 1, 2, or 3 octal digits exactly.
343			*
344			* There is no such thing as a decimal escape sequence.
345			*
346			* Hexadecimal escape sequences are begun with a leading \x and continue
347			* until a non-hexadecimal character is found.
348			*
349			* No real handling of tri-graph sequences, yet.
350			*/
351
352			protected
353			Escape
354			:
355			'\\'
356			( options{warnWhenFollowAmbig=false;}:
357			'a'
358			\| 'b'
359			\| 'f'
360			\| 'n'
361			\| 'r'
362			\| 't'
363			\| 'v'
364			\| '"'
365			\| '\''
366			\| '\\'
367			\| '?'
368			\| ('0'..'3') (options{warnWhenFollowAmbig=false;}: Digit (options{warnWhenFollowAmbig=false;}: Digit)? )?
369			\| ('4'..'7') (options{warnWhenFollowAmbig=false;}: Digit)?
370			\| 'x' (options{warnWhenFollowAmbig=false;}: Digit \| 'a'..'f' \| 'A'..'F')+
371			)
372			;
373
374
375			protected
376	tim	814	Vocabulary
377			:
378			'\3'..'\377'
379			;
380
381
382			ID
383			options {testLiterals = true;}
384			:
385			('a'..'z'\|'A'..'Z'\|'_')
386			('a'..'z'\|'A'..'Z'\|'_'\|'0'..'9')*
387			;
388
389
390			protected
391	tim	770	Digit
392			:
393			'0'..'9'
394			;
395
396			protected
397			Decimal
398			:
399			('0'..'9')+
400			;
401
402	tim	814	// hexadecimal digit (again, note it's protected!)
403	tim	770	protected
404	tim	814	HEX_DIGIT
405			: ('0'..'9'\|'A'..'F'\|'a'..'f')
406			;
407	tim	770
408
409	tim	814	// a numeric literal
410			NUM_INT
411			{
412			bool isDecimal = false;
413			ANTLR_USE_NAMESPACE(antlr)RefToken t = ANTLR_USE_NAMESPACE(antlr)nullToken;
414			}
415			: ('+' \| '-')?
416			(
417			'.' {_ttype = DOT;}
418			( ('0'..'9')+ (EXPONENT)? (f1:FLOAT_SUFFIX {t=f1;})?
419			{
420			if ( t &&
421			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
422			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
423			_ttype = NUM_FLOAT;
424			}
425			else {
426			_ttype = NUM_DOUBLE; // assume double
427			}
428			}
429			)?
430	tim	770
431	tim	814	\| ( '0' {isDecimal = true;} // special case for just '0'
432			( ('x'\|'X')
433			( // hex
434			// the 'e'\|'E' and float suffix stuff look
435			// like hex digits, hence the (...)+ doesn't
436			// know when to stop: ambig. ANTLR resolves
437			// it correctly by matching immediately. It
438			// is therefor ok to hush warning.
439			options {
440			warnWhenFollowAmbig=false;
441			}
442			: HEX_DIGIT
443			)+
444			\| //float or double with leading zero
445			(('0'..'9')+ ('.'\|EXPONENT\|FLOAT_SUFFIX)) => ('0'..'9')+
446			\| ('0'..'7')+ // octal
447			)?
448			\| ('1'..'9') ('0'..'9')* {isDecimal=true;} // non-zero decimal
449			)
450			( ('l'\|'L') { _ttype = NUM_LONG; }
451
452			// only check to see if it's a float if looks like decimal so far
453			\| {isDecimal}?
454			( '.' ('0'..'9')* (EXPONENT)? (f2:FLOAT_SUFFIX {t=f2;})?
455			\| EXPONENT (f3:FLOAT_SUFFIX {t=f3;})?
456			\| f4:FLOAT_SUFFIX {t=f4;}
457			)
458			{
459			if ( t &&
460			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
461			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
462			_ttype = NUM_FLOAT;
463			}
464			else {
465			_ttype = NUM_DOUBLE; // assume double
466			}
467			}
468			)?
469			)
470			;
471
472			// a couple protected methods to assist in matching floating point numbers
473	tim	770	protected
474	tim	814	EXPONENT
475	gezelter	848	: ('e'\|'E'\|'d'\|'D') ('+'\|'-')? ('0'..'9')+
476	tim	814	;
477	tim	770
478			protected
479	tim	814	FLOAT_SUFFIX
480			: 'f'\|'F'\|'d'\|'D'
481			;