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";
  FLUCQ       = "flucQ";
  RNEMD       = "RNEMD";
  ENDBLOCK;
}


mdfile  : (statement)*
        ;

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

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);}
                  ;

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

rnemdblock  : RNEMD^ 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:	1731
Committed:	Thu May 31 12:25:30 2012 UTC (12 years, 11 months ago) by gezelter
File size:	10966 byte(s)
Log Message:	Reorganized source directories, added RNEMD and flucQ blocks for options parsing.
#	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	gezelter	1731	FLUCQ = "flucQ";
45			RNEMD = "RNEMD";
46	tim	770	ENDBLOCK;
47			}
48
49
50			mdfile : (statement)*
51			;
52
53			statement : assignment
54	gezelter	1731	\| componentblock
55			\| moleculeblock
56			\| zconstraintblock
57			\| restraintblock
58			\| flucqblock
59			\| rnemdblock
60			;
61
62	tim	770	assignment : ID ASSIGNEQUAL^ constant SEMICOLON!
63			;
64
65	tim	814	constant : intConst
66			\| floatConst
67	tim	770	\| ID
68			\| StringLiteral
69			;
70
71			componentblock : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
72			;
73
74			zconstraintblock : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
75			;
76	cli2	1360
77			restraintblock : RESTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
78			;
79	gezelter	1731
80			flucqblock : FLUCQ^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
81			;
82
83			rnemdblock : RNEMD^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
84			;
85	tim	770
86			moleculeblock : MOLECULE^ LCURLY! (moleculestatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
87			;
88
89			moleculestatement : assignment
90			\| atomblock
91			\| bondblock
92			\| bendblock
93			\| torsionblock
94	gezelter	1277	\| inversionblock
95	tim	770	\| rigidbodyblock
96			\| cutoffgroupblock
97			\| fragmentblock
98			;
99
100			atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
101			;
102
103			atomstatement : assignment
104	tim	814	\| POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
105			\| ORIENTATION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
106	tim	770	;
107
108
109			bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
110			;
111
112			bondstatement : assignment
113			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
114			;
115
116			bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
117			;
118
119			bendstatement : assignment
120			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
121			;
122
123			torsionblock : TORSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
124			;
125
126			torsionstatement : assignment
127			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
128			;
129
130	cli2	1275	inversionblock : INVERSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
131			;
132
133			inversionstatement : assignment
134			\| CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
135			;
136
137	tim	770	rigidbodyblock : RIGIDBODY^ LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
138			;
139
140			rigidbodystatement : assignment
141			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
142			;
143
144			cutoffgroupblock : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
145			;
146
147			cutoffgroupstatement : assignment
148			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
149			;
150
151			fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
152			;
153
154			fragmentstatement : assignment
155			;
156
157
158
159	tim	814	doubleNumberTuple : doubleNumber (COMMA! doubleNumber)*
160	tim	770	;
161
162			inttuple : intConst (COMMA! intConst)*
163			;
164
165			protected
166			intConst
167	tim	814	: NUM_INT \| NUM_LONG
168	tim	770	;
169
170			protected
171	tim	814	doubleNumber :
172	tim	770	(intConst \| floatConst)
173			;
174
175			protected
176			floatConst
177			:
178	tim	814	NUM_FLOAT \| NUM_DOUBLE
179	tim	770	;
180
181
182
183			class MDLexer extends Lexer;
184
185			options
186			{
187			k = 3;
188			exportVocab = MD;
189			testLiterals = false;
190			}
191
192			tokens {
193			DOT;
194			}
195
196			{
197
198
199			int deferredLineCount;
200			FilenameObserver* observer;
201
202			public:
203			void setObserver(FilenameObserver* osv) {observer = osv;}
204			void initDeferredLineCount() { deferredLineCount = 0;}
205			void deferredNewline() {
206			deferredLineCount++;
207			}
208
209
210			virtual void newline() {
211			for (;deferredLineCount>0;deferredLineCount--) {
212			CharScanner::newline();
213			}
214			CharScanner::newline();
215			}
216
217			}
218
219
220			// Operators:
221
222			ASSIGNEQUAL : '=' ;
223			COLON : ':' ;
224			COMMA : ',' ;
225			QUESTIONMARK : '?' ;
226			SEMICOLON : ';' ;
227
228			LPAREN : '(' ;
229			RPAREN : ')' ;
230			LBRACKET : '[' ;
231			RBRACKET : ']' ;
232			LCURLY : '{' ;
233			RCURLY : '}' ;
234
235			Whitespace
236			:
237			( // whitespace ignored
238			(' ' \|'\t' \| '\f')
239			\| // handle newlines
240			( '\r' '\n' // MS
241			\| '\r' // Mac
242			\| '\n' // Unix
243			) { newline(); }
244			\| // handle continuation lines
245			( '\\' '\r' '\n' // MS
246			\| '\\' '\r' // Mac
247			\| '\\' '\n' // Unix
248			) {printf("CPP_parser.g continuation line detected\n");
249			deferredNewline();}
250			)
251			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
252			;
253
254			Comment
255			:
256			"/*"
257			( {LA(2) != '/'}? '*'
258			\| EndOfLine {deferredNewline();}
259			\| ~('*'\| '\r' \| '\n')
260			)*
261			"*/" {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
262			;
263
264			CPPComment
265			:
266			"//" (~('\n' \| '\r'))* EndOfLine
267			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
268			;
269
270			PREPROC_DIRECTIVE
271			options{paraphrase = "a line directive";}
272			:
273			'#' LineDirective
274			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
275			;
276
277			protected
278			LineDirective
279			:
280			{
281			deferredLineCount = 0;
282			}
283			("line")? // this would be for if the directive started "#line"
284			(Space)+
285	gezelter	1390	n:Decimal { setLine(OpenMD::lexi_cast<int>(n->getText()) - 1); }
286	tim	770	(Space)+
287			(sl:StringLiteral) {std::string filename = sl->getText().substr(1,sl->getText().length()-2); observer->notify(filename);}
288			((Space)+ Decimal)* // To support cpp flags (GNU)
289			EndOfLine
290			;
291
292			protected
293			Space
294			:
295			(' '\|'\t'\|'\f')
296			;
297
298
299			// Literals:
300
301			/*
302			* Note that we do NOT handle tri-graphs nor multi-byte sequences.
303			*/
304
305			/*
306			* Note that we can't have empty character constants (even though we
307			* can have empty strings :-).
308			*/
309			CharLiteral
310			:
311			'\'' (Escape \| ~('\'')) '\''
312			;
313
314			/*
315			* Can't have raw imbedded newlines in string constants. Strict reading of
316			* the standard gives odd dichotomy between newlines & carriage returns.
317			* Go figure.
318			*/
319			StringLiteral
320			:
321			'"'
322			( Escape
323			\|
324			( "\\\r\n" // MS
325			\| "\\\r" // MAC
326			\| "\\\n" // Unix
327			) {deferredNewline();}
328			\|
329			~('"'\|'\r'\|'\n'\|'\\')
330			)*
331			'"'
332			;
333
334			protected
335			EndOfLine
336			:
337			( options{generateAmbigWarnings = false;}:
338			"\r\n" // MS
339			\| '\r' // Mac
340			\| '\n' // Unix
341			)
342			;
343
344			/*
345			* Handle the various escape sequences.
346			*
347			* Note carefully that these numeric escape sequences are not of the
348			* same form as the C language numeric constants.
349			*
350			* There is no such thing as a binary numeric escape sequence.
351			*
352			* Octal escape sequences are either 1, 2, or 3 octal digits exactly.
353			*
354			* There is no such thing as a decimal escape sequence.
355			*
356			* Hexadecimal escape sequences are begun with a leading \x and continue
357			* until a non-hexadecimal character is found.
358			*
359			* No real handling of tri-graph sequences, yet.
360			*/
361
362			protected
363			Escape
364			:
365			'\\'
366			( options{warnWhenFollowAmbig=false;}:
367			'a'
368			\| 'b'
369			\| 'f'
370			\| 'n'
371			\| 'r'
372			\| 't'
373			\| 'v'
374			\| '"'
375			\| '\''
376			\| '\\'
377			\| '?'
378			\| ('0'..'3') (options{warnWhenFollowAmbig=false;}: Digit (options{warnWhenFollowAmbig=false;}: Digit)? )?
379			\| ('4'..'7') (options{warnWhenFollowAmbig=false;}: Digit)?
380			\| 'x' (options{warnWhenFollowAmbig=false;}: Digit \| 'a'..'f' \| 'A'..'F')+
381			)
382			;
383
384
385			protected
386	tim	814	Vocabulary
387			:
388			'\3'..'\377'
389			;
390
391
392			ID
393			options {testLiterals = true;}
394			:
395			('a'..'z'\|'A'..'Z'\|'_')
396			('a'..'z'\|'A'..'Z'\|'_'\|'0'..'9')*
397			;
398
399
400			protected
401	tim	770	Digit
402			:
403			'0'..'9'
404			;
405
406			protected
407			Decimal
408			:
409			('0'..'9')+
410			;
411
412	tim	814	// hexadecimal digit (again, note it's protected!)
413	tim	770	protected
414	tim	814	HEX_DIGIT
415			: ('0'..'9'\|'A'..'F'\|'a'..'f')
416			;
417	tim	770
418
419	tim	814	// a numeric literal
420			NUM_INT
421			{
422			bool isDecimal = false;
423			ANTLR_USE_NAMESPACE(antlr)RefToken t = ANTLR_USE_NAMESPACE(antlr)nullToken;
424			}
425			: ('+' \| '-')?
426			(
427			'.' {_ttype = DOT;}
428			( ('0'..'9')+ (EXPONENT)? (f1:FLOAT_SUFFIX {t=f1;})?
429			{
430			if ( t &&
431			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
432			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
433			_ttype = NUM_FLOAT;
434			}
435			else {
436			_ttype = NUM_DOUBLE; // assume double
437			}
438			}
439			)?
440	tim	770
441	tim	814	\| ( '0' {isDecimal = true;} // special case for just '0'
442			( ('x'\|'X')
443			( // hex
444			// the 'e'\|'E' and float suffix stuff look
445			// like hex digits, hence the (...)+ doesn't
446			// know when to stop: ambig. ANTLR resolves
447			// it correctly by matching immediately. It
448			// is therefor ok to hush warning.
449			options {
450			warnWhenFollowAmbig=false;
451			}
452			: HEX_DIGIT
453			)+
454			\| //float or double with leading zero
455			(('0'..'9')+ ('.'\|EXPONENT\|FLOAT_SUFFIX)) => ('0'..'9')+
456			\| ('0'..'7')+ // octal
457			)?
458			\| ('1'..'9') ('0'..'9')* {isDecimal=true;} // non-zero decimal
459			)
460			( ('l'\|'L') { _ttype = NUM_LONG; }
461
462			// only check to see if it's a float if looks like decimal so far
463			\| {isDecimal}?
464			( '.' ('0'..'9')* (EXPONENT)? (f2:FLOAT_SUFFIX {t=f2;})?
465			\| EXPONENT (f3:FLOAT_SUFFIX {t=f3;})?
466			\| f4:FLOAT_SUFFIX {t=f4;}
467			)
468			{
469			if ( t &&
470			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
471			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
472			_ttype = NUM_FLOAT;
473			}
474			else {
475			_ttype = NUM_DOUBLE; // assume double
476			}
477			}
478			)?
479			)
480			;
481
482			// a couple protected methods to assist in matching floating point numbers
483	tim	770	protected
484	tim	814	EXPONENT
485	gezelter	848	: ('e'\|'E'\|'d'\|'D') ('+'\|'-')? ('0'..'9')+
486	tim	814	;
487	tim	770
488			protected
489	tim	814	FLOAT_SUFFIX
490			: 'f'\|'F'\|'d'\|'D'
491			;