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

moleculestatement : assignment
                  | atomblock
                  | bondblock
                  | bendblock
                  | torsionblock
                  | 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(oopse::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:	1275
Committed:	Fri Jul 4 20:54:29 2008 UTC (16 years, 10 months ago) by cli2
Original Path:	*trunk/src/mdParser/MDParser.g*
File size:	10549 byte(s)
Log Message:	Changes required for Inversions and Base Atom types. This will break OOPSE badly for a few days or so...
#	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			ATOM = "atom";
32			BOND = "bond";
33			BEND = "bend";
34			TORSION = "torsion";
35	cli2	1275	INVERSION = "inversion";
36	tim	770	RIGIDBODY = "rigidBody";
37			CUTOFFGROUP = "cutoffGroup";
38			FRAGMENT = "fragment";
39			MEMBERS = "members";
40	cli2	1275	CENTER = "center";
41	tim	770	POSITION = "position";
42			ORIENTATION = "orientation";
43			ENDBLOCK;
44			}
45
46
47			mdfile : (statement)*
48			;
49
50			statement : assignment
51			\| componentblock
52			\| moleculeblock
53			\| zconstraintblock
54			;
55
56			assignment : ID ASSIGNEQUAL^ constant SEMICOLON!
57			;
58
59	tim	814	constant : intConst
60			\| floatConst
61	tim	770	\| ID
62			\| StringLiteral
63			;
64
65			componentblock : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
66			;
67
68			zconstraintblock : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
69			;
70
71			moleculeblock : MOLECULE^ LCURLY! (moleculestatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
72			;
73
74			moleculestatement : assignment
75			\| atomblock
76			\| bondblock
77			\| bendblock
78			\| torsionblock
79			\| rigidbodyblock
80			\| cutoffgroupblock
81			\| fragmentblock
82			;
83
84			atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
85			;
86
87			atomstatement : assignment
88	tim	814	\| POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
89			\| ORIENTATION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
90	tim	770	;
91
92
93			bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
94			;
95
96			bondstatement : assignment
97			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
98			;
99
100			bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
101			;
102
103			bendstatement : assignment
104			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
105			;
106
107			torsionblock : TORSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
108			;
109
110			torsionstatement : assignment
111			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
112			;
113
114	cli2	1275	inversionblock : INVERSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
115			;
116
117			inversionstatement : assignment
118			\| CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
119			;
120
121	tim	770	rigidbodyblock : RIGIDBODY^ LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
122			;
123
124			rigidbodystatement : assignment
125			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
126			;
127
128			cutoffgroupblock : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
129			;
130
131			cutoffgroupstatement : assignment
132			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
133			;
134
135			fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
136			;
137
138			fragmentstatement : assignment
139			;
140
141
142
143	tim	814	doubleNumberTuple : doubleNumber (COMMA! doubleNumber)*
144	tim	770	;
145
146			inttuple : intConst (COMMA! intConst)*
147			;
148
149			protected
150			intConst
151	tim	814	: NUM_INT \| NUM_LONG
152	tim	770	;
153
154			protected
155	tim	814	doubleNumber :
156	tim	770	(intConst \| floatConst)
157			;
158
159			protected
160			floatConst
161			:
162	tim	814	NUM_FLOAT \| NUM_DOUBLE
163	tim	770	;
164
165
166
167			class MDLexer extends Lexer;
168
169			options
170			{
171			k = 3;
172			exportVocab = MD;
173			testLiterals = false;
174			}
175
176			tokens {
177			DOT;
178			}
179
180			{
181
182
183			int deferredLineCount;
184			FilenameObserver* observer;
185
186			public:
187			void setObserver(FilenameObserver* osv) {observer = osv;}
188			void initDeferredLineCount() { deferredLineCount = 0;}
189			void deferredNewline() {
190			deferredLineCount++;
191			}
192
193
194			virtual void newline() {
195			for (;deferredLineCount>0;deferredLineCount--) {
196			CharScanner::newline();
197			}
198			CharScanner::newline();
199			}
200
201			}
202
203
204			// Operators:
205
206			ASSIGNEQUAL : '=' ;
207			COLON : ':' ;
208			COMMA : ',' ;
209			QUESTIONMARK : '?' ;
210			SEMICOLON : ';' ;
211
212			LPAREN : '(' ;
213			RPAREN : ')' ;
214			LBRACKET : '[' ;
215			RBRACKET : ']' ;
216			LCURLY : '{' ;
217			RCURLY : '}' ;
218
219			Whitespace
220			:
221			( // whitespace ignored
222			(' ' \|'\t' \| '\f')
223			\| // handle newlines
224			( '\r' '\n' // MS
225			\| '\r' // Mac
226			\| '\n' // Unix
227			) { newline(); }
228			\| // handle continuation lines
229			( '\\' '\r' '\n' // MS
230			\| '\\' '\r' // Mac
231			\| '\\' '\n' // Unix
232			) {printf("CPP_parser.g continuation line detected\n");
233			deferredNewline();}
234			)
235			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
236			;
237
238			Comment
239			:
240			"/*"
241			( {LA(2) != '/'}? '*'
242			\| EndOfLine {deferredNewline();}
243			\| ~('*'\| '\r' \| '\n')
244			)*
245			"*/" {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
246			;
247
248			CPPComment
249			:
250			"//" (~('\n' \| '\r'))* EndOfLine
251			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
252			;
253
254			PREPROC_DIRECTIVE
255			options{paraphrase = "a line directive";}
256			:
257			'#' LineDirective
258			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
259			;
260
261			protected
262			LineDirective
263			:
264			{
265			deferredLineCount = 0;
266			}
267			("line")? // this would be for if the directive started "#line"
268			(Space)+
269			n:Decimal { setLine(oopse::lexi_cast<int>(n->getText()) - 1); }
270			(Space)+
271			(sl:StringLiteral) {std::string filename = sl->getText().substr(1,sl->getText().length()-2); observer->notify(filename);}
272			((Space)+ Decimal)* // To support cpp flags (GNU)
273			EndOfLine
274			;
275
276			protected
277			Space
278			:
279			(' '\|'\t'\|'\f')
280			;
281
282
283			// Literals:
284
285			/*
286			* Note that we do NOT handle tri-graphs nor multi-byte sequences.
287			*/
288
289			/*
290			* Note that we can't have empty character constants (even though we
291			* can have empty strings :-).
292			*/
293			CharLiteral
294			:
295			'\'' (Escape \| ~('\'')) '\''
296			;
297
298			/*
299			* Can't have raw imbedded newlines in string constants. Strict reading of
300			* the standard gives odd dichotomy between newlines & carriage returns.
301			* Go figure.
302			*/
303			StringLiteral
304			:
305			'"'
306			( Escape
307			\|
308			( "\\\r\n" // MS
309			\| "\\\r" // MAC
310			\| "\\\n" // Unix
311			) {deferredNewline();}
312			\|
313			~('"'\|'\r'\|'\n'\|'\\')
314			)*
315			'"'
316			;
317
318			protected
319			EndOfLine
320			:
321			( options{generateAmbigWarnings = false;}:
322			"\r\n" // MS
323			\| '\r' // Mac
324			\| '\n' // Unix
325			)
326			;
327
328			/*
329			* Handle the various escape sequences.
330			*
331			* Note carefully that these numeric escape sequences are not of the
332			* same form as the C language numeric constants.
333			*
334			* There is no such thing as a binary numeric escape sequence.
335			*
336			* Octal escape sequences are either 1, 2, or 3 octal digits exactly.
337			*
338			* There is no such thing as a decimal escape sequence.
339			*
340			* Hexadecimal escape sequences are begun with a leading \x and continue
341			* until a non-hexadecimal character is found.
342			*
343			* No real handling of tri-graph sequences, yet.
344			*/
345
346			protected
347			Escape
348			:
349			'\\'
350			( options{warnWhenFollowAmbig=false;}:
351			'a'
352			\| 'b'
353			\| 'f'
354			\| 'n'
355			\| 'r'
356			\| 't'
357			\| 'v'
358			\| '"'
359			\| '\''
360			\| '\\'
361			\| '?'
362			\| ('0'..'3') (options{warnWhenFollowAmbig=false;}: Digit (options{warnWhenFollowAmbig=false;}: Digit)? )?
363			\| ('4'..'7') (options{warnWhenFollowAmbig=false;}: Digit)?
364			\| 'x' (options{warnWhenFollowAmbig=false;}: Digit \| 'a'..'f' \| 'A'..'F')+
365			)
366			;
367
368
369			protected
370	tim	814	Vocabulary
371			:
372			'\3'..'\377'
373			;
374
375
376			ID
377			options {testLiterals = true;}
378			:
379			('a'..'z'\|'A'..'Z'\|'_')
380			('a'..'z'\|'A'..'Z'\|'_'\|'0'..'9')*
381			;
382
383
384			protected
385	tim	770	Digit
386			:
387			'0'..'9'
388			;
389
390			protected
391			Decimal
392			:
393			('0'..'9')+
394			;
395
396	tim	814	// hexadecimal digit (again, note it's protected!)
397	tim	770	protected
398	tim	814	HEX_DIGIT
399			: ('0'..'9'\|'A'..'F'\|'a'..'f')
400			;
401	tim	770
402
403	tim	814	// a numeric literal
404			NUM_INT
405			{
406			bool isDecimal = false;
407			ANTLR_USE_NAMESPACE(antlr)RefToken t = ANTLR_USE_NAMESPACE(antlr)nullToken;
408			}
409			: ('+' \| '-')?
410			(
411			'.' {_ttype = DOT;}
412			( ('0'..'9')+ (EXPONENT)? (f1:FLOAT_SUFFIX {t=f1;})?
413			{
414			if ( t &&
415			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
416			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
417			_ttype = NUM_FLOAT;
418			}
419			else {
420			_ttype = NUM_DOUBLE; // assume double
421			}
422			}
423			)?
424	tim	770
425	tim	814	\| ( '0' {isDecimal = true;} // special case for just '0'
426			( ('x'\|'X')
427			( // hex
428			// the 'e'\|'E' and float suffix stuff look
429			// like hex digits, hence the (...)+ doesn't
430			// know when to stop: ambig. ANTLR resolves
431			// it correctly by matching immediately. It
432			// is therefor ok to hush warning.
433			options {
434			warnWhenFollowAmbig=false;
435			}
436			: HEX_DIGIT
437			)+
438			\| //float or double with leading zero
439			(('0'..'9')+ ('.'\|EXPONENT\|FLOAT_SUFFIX)) => ('0'..'9')+
440			\| ('0'..'7')+ // octal
441			)?
442			\| ('1'..'9') ('0'..'9')* {isDecimal=true;} // non-zero decimal
443			)
444			( ('l'\|'L') { _ttype = NUM_LONG; }
445
446			// only check to see if it's a float if looks like decimal so far
447			\| {isDecimal}?
448			( '.' ('0'..'9')* (EXPONENT)? (f2:FLOAT_SUFFIX {t=f2;})?
449			\| EXPONENT (f3:FLOAT_SUFFIX {t=f3;})?
450			\| f4:FLOAT_SUFFIX {t=f4;}
451			)
452			{
453			if ( t &&
454			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
455			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
456			_ttype = NUM_FLOAT;
457			}
458			else {
459			_ttype = NUM_DOUBLE; // assume double
460			}
461			}
462			)?
463			)
464			;
465
466			// a couple protected methods to assist in matching floating point numbers
467	tim	770	protected
468	tim	814	EXPONENT
469	gezelter	848	: ('e'\|'E'\|'d'\|'D') ('+'\|'-')? ('0'..'9')+
470	tim	814	;
471	tim	770
472			protected
473	tim	814	FLOAT_SUFFIX
474			: 'f'\|'F'\|'d'\|'D'
475			;