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";
  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!
              ;

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:	1776
Committed:	Thu Aug 9 15:52:59 2012 UTC (12 years, 8 months ago) by gezelter
File size:	11242 byte(s)
Log Message:	Fixes to mdParser to handle vector assignments, fixes for VelocityVerletIntegrator deleting rnemd_ when it doesn't exist yet.
#	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	gezelter	1746	MINIMIZER = "minimizer";
47	tim	770	ENDBLOCK;
48			}
49
50
51			mdfile : (statement)*
52			;
53
54			statement : assignment
55	gezelter	1731	\| componentblock
56			\| moleculeblock
57			\| zconstraintblock
58			\| restraintblock
59			\| flucqblock
60			\| rnemdblock
61	gezelter	1746	\| minimizerblock
62	gezelter	1731	;
63
64	tim	770	assignment : ID ASSIGNEQUAL^ constant SEMICOLON!
65			;
66
67	tim	814	constant : intConst
68	gezelter	1776	\| floatConst
69			\| vectorConst
70	tim	770	\| ID
71			\| StringLiteral
72			;
73
74			componentblock : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
75			;
76
77			zconstraintblock : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
78			;
79	cli2	1360
80			restraintblock : RESTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
81			;
82	gezelter	1731
83			flucqblock : FLUCQ^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
84			;
85
86			rnemdblock : RNEMD^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
87			;
88	gezelter	1746
89			minimizerblock : MINIMIZER^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
90			;
91	tim	770
92			moleculeblock : MOLECULE^ LCURLY! (moleculestatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
93			;
94
95			moleculestatement : assignment
96			\| atomblock
97			\| bondblock
98			\| bendblock
99			\| torsionblock
100	gezelter	1277	\| inversionblock
101	tim	770	\| rigidbodyblock
102			\| cutoffgroupblock
103			\| fragmentblock
104			;
105
106			atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
107			;
108
109			atomstatement : assignment
110	tim	814	\| POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
111			\| ORIENTATION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
112	tim	770	;
113
114
115			bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
116			;
117
118			bondstatement : assignment
119			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
120			;
121
122			bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
123			;
124
125			bendstatement : assignment
126			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
127			;
128
129			torsionblock : TORSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
130			;
131
132			torsionstatement : assignment
133			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
134			;
135
136	cli2	1275	inversionblock : INVERSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
137			;
138
139			inversionstatement : assignment
140			\| CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
141			;
142
143	tim	770	rigidbodyblock : RIGIDBODY^ LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
144			;
145
146			rigidbodystatement : assignment
147			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
148			;
149
150			cutoffgroupblock : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
151			;
152
153			cutoffgroupstatement : assignment
154			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
155			;
156
157			fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
158			;
159
160			fragmentstatement : assignment
161			;
162
163
164
165	tim	814	doubleNumberTuple : doubleNumber (COMMA! doubleNumber)*
166	tim	770	;
167
168			inttuple : intConst (COMMA! intConst)*
169			;
170
171			protected
172			intConst
173	tim	814	: NUM_INT \| NUM_LONG
174	tim	770	;
175
176			protected
177	tim	814	doubleNumber :
178	tim	770	(intConst \| floatConst)
179			;
180
181			protected
182			floatConst
183			:
184	tim	814	NUM_FLOAT \| NUM_DOUBLE
185	tim	770	;
186
187	gezelter	1776	protected
188			vectorConst
189			:
190			LPAREN^ doubleNumber COMMA doubleNumber COMMA doubleNumber RPAREN
191			;
192	tim	770
193			class MDLexer extends Lexer;
194
195			options
196			{
197			k = 3;
198			exportVocab = MD;
199			testLiterals = false;
200			}
201
202			tokens {
203			DOT;
204			}
205
206			{
207
208
209			int deferredLineCount;
210			FilenameObserver* observer;
211
212			public:
213			void setObserver(FilenameObserver* osv) {observer = osv;}
214			void initDeferredLineCount() { deferredLineCount = 0;}
215			void deferredNewline() {
216			deferredLineCount++;
217			}
218
219
220			virtual void newline() {
221			for (;deferredLineCount>0;deferredLineCount--) {
222			CharScanner::newline();
223			}
224			CharScanner::newline();
225			}
226
227			}
228
229
230			// Operators:
231
232			ASSIGNEQUAL : '=' ;
233			COLON : ':' ;
234			COMMA : ',' ;
235			QUESTIONMARK : '?' ;
236			SEMICOLON : ';' ;
237
238			LPAREN : '(' ;
239			RPAREN : ')' ;
240			LBRACKET : '[' ;
241			RBRACKET : ']' ;
242			LCURLY : '{' ;
243			RCURLY : '}' ;
244
245			Whitespace
246			:
247			( // whitespace ignored
248			(' ' \|'\t' \| '\f')
249			\| // handle newlines
250			( '\r' '\n' // MS
251			\| '\r' // Mac
252			\| '\n' // Unix
253			) { newline(); }
254			\| // handle continuation lines
255			( '\\' '\r' '\n' // MS
256			\| '\\' '\r' // Mac
257			\| '\\' '\n' // Unix
258			) {printf("CPP_parser.g continuation line detected\n");
259			deferredNewline();}
260			)
261			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
262			;
263
264			Comment
265			:
266			"/*"
267			( {LA(2) != '/'}? '*'
268			\| EndOfLine {deferredNewline();}
269			\| ~('*'\| '\r' \| '\n')
270			)*
271			"*/" {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
272			;
273
274			CPPComment
275			:
276			"//" (~('\n' \| '\r'))* EndOfLine
277			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
278			;
279
280			PREPROC_DIRECTIVE
281			options{paraphrase = "a line directive";}
282			:
283			'#' LineDirective
284			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
285			;
286
287			protected
288			LineDirective
289			:
290			{
291			deferredLineCount = 0;
292			}
293			("line")? // this would be for if the directive started "#line"
294			(Space)+
295	gezelter	1390	n:Decimal { setLine(OpenMD::lexi_cast<int>(n->getText()) - 1); }
296	tim	770	(Space)+
297			(sl:StringLiteral) {std::string filename = sl->getText().substr(1,sl->getText().length()-2); observer->notify(filename);}
298			((Space)+ Decimal)* // To support cpp flags (GNU)
299			EndOfLine
300			;
301
302			protected
303			Space
304			:
305			(' '\|'\t'\|'\f')
306			;
307
308
309			// Literals:
310
311			/*
312			* Note that we do NOT handle tri-graphs nor multi-byte sequences.
313			*/
314
315			/*
316			* Note that we can't have empty character constants (even though we
317			* can have empty strings :-).
318			*/
319			CharLiteral
320			:
321			'\'' (Escape \| ~('\'')) '\''
322			;
323
324			/*
325			* Can't have raw imbedded newlines in string constants. Strict reading of
326			* the standard gives odd dichotomy between newlines & carriage returns.
327			* Go figure.
328			*/
329			StringLiteral
330			:
331			'"'
332			( Escape
333			\|
334			( "\\\r\n" // MS
335			\| "\\\r" // MAC
336			\| "\\\n" // Unix
337			) {deferredNewline();}
338			\|
339			~('"'\|'\r'\|'\n'\|'\\')
340			)*
341			'"'
342			;
343
344			protected
345			EndOfLine
346			:
347			( options{generateAmbigWarnings = false;}:
348			"\r\n" // MS
349			\| '\r' // Mac
350			\| '\n' // Unix
351			)
352			;
353
354			/*
355			* Handle the various escape sequences.
356			*
357			* Note carefully that these numeric escape sequences are not of the
358			* same form as the C language numeric constants.
359			*
360			* There is no such thing as a binary numeric escape sequence.
361			*
362			* Octal escape sequences are either 1, 2, or 3 octal digits exactly.
363			*
364			* There is no such thing as a decimal escape sequence.
365			*
366			* Hexadecimal escape sequences are begun with a leading \x and continue
367			* until a non-hexadecimal character is found.
368			*
369			* No real handling of tri-graph sequences, yet.
370			*/
371
372			protected
373			Escape
374			:
375			'\\'
376			( options{warnWhenFollowAmbig=false;}:
377			'a'
378			\| 'b'
379			\| 'f'
380			\| 'n'
381			\| 'r'
382			\| 't'
383			\| 'v'
384			\| '"'
385			\| '\''
386			\| '\\'
387			\| '?'
388			\| ('0'..'3') (options{warnWhenFollowAmbig=false;}: Digit (options{warnWhenFollowAmbig=false;}: Digit)? )?
389			\| ('4'..'7') (options{warnWhenFollowAmbig=false;}: Digit)?
390			\| 'x' (options{warnWhenFollowAmbig=false;}: Digit \| 'a'..'f' \| 'A'..'F')+
391			)
392			;
393
394
395			protected
396	tim	814	Vocabulary
397			:
398			'\3'..'\377'
399			;
400
401
402			ID
403			options {testLiterals = true;}
404			:
405			('a'..'z'\|'A'..'Z'\|'_')
406			('a'..'z'\|'A'..'Z'\|'_'\|'0'..'9')*
407			;
408
409
410			protected
411	tim	770	Digit
412			:
413			'0'..'9'
414			;
415
416			protected
417			Decimal
418			:
419			('0'..'9')+
420			;
421
422	tim	814	// hexadecimal digit (again, note it's protected!)
423	tim	770	protected
424	tim	814	HEX_DIGIT
425			: ('0'..'9'\|'A'..'F'\|'a'..'f')
426			;
427	tim	770
428
429	tim	814	// a numeric literal
430			NUM_INT
431			{
432			bool isDecimal = false;
433			ANTLR_USE_NAMESPACE(antlr)RefToken t = ANTLR_USE_NAMESPACE(antlr)nullToken;
434			}
435			: ('+' \| '-')?
436			(
437			'.' {_ttype = DOT;}
438			( ('0'..'9')+ (EXPONENT)? (f1:FLOAT_SUFFIX {t=f1;})?
439			{
440			if ( t &&
441			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
442			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
443			_ttype = NUM_FLOAT;
444			}
445			else {
446			_ttype = NUM_DOUBLE; // assume double
447			}
448			}
449			)?
450	tim	770
451	tim	814	\| ( '0' {isDecimal = true;} // special case for just '0'
452			( ('x'\|'X')
453			( // hex
454			// the 'e'\|'E' and float suffix stuff look
455			// like hex digits, hence the (...)+ doesn't
456			// know when to stop: ambig. ANTLR resolves
457			// it correctly by matching immediately. It
458			// is therefor ok to hush warning.
459			options {
460			warnWhenFollowAmbig=false;
461			}
462			: HEX_DIGIT
463			)+
464			\| //float or double with leading zero
465			(('0'..'9')+ ('.'\|EXPONENT\|FLOAT_SUFFIX)) => ('0'..'9')+
466			\| ('0'..'7')+ // octal
467			)?
468			\| ('1'..'9') ('0'..'9')* {isDecimal=true;} // non-zero decimal
469			)
470			( ('l'\|'L') { _ttype = NUM_LONG; }
471
472			// only check to see if it's a float if looks like decimal so far
473			\| {isDecimal}?
474			( '.' ('0'..'9')* (EXPONENT)? (f2:FLOAT_SUFFIX {t=f2;})?
475			\| EXPONENT (f3:FLOAT_SUFFIX {t=f3;})?
476			\| f4:FLOAT_SUFFIX {t=f4;}
477			)
478			{
479			if ( t &&
480			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
481			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
482			_ttype = NUM_FLOAT;
483			}
484			else {
485			_ttype = NUM_DOUBLE; // assume double
486			}
487			}
488			)?
489			)
490			;
491
492			// a couple protected methods to assist in matching floating point numbers
493	tim	770	protected
494	tim	814	EXPONENT
495	gezelter	848	: ('e'\|'E'\|'d'\|'D') ('+'\|'-')? ('0'..'9')+
496	tim	814	;
497	tim	770
498			protected
499	tim	814	FLOAT_SUFFIX
500			: 'f'\|'F'\|'d'\|'D'
501			;