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";
  CONSTRAINT  = "constraint";
  DISTANCE    = "distance";
  FRAGMENT    = "fragment";
  MEMBERS     = "members";
  CENTER      = "center";
  SATELLITES  = "satellites";
  POSITION    = "position";
  ORIENTATION = "orientation";
  FLUCQ       = "flucQ";
  RNEMD       = "RNEMD";
  MINIMIZER   = "minimizer";
  ENDBLOCK;
}


mdfile  : (statement)*
        ;

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

assignment  : ID ASSIGNEQUAL^ constant SEMICOLON!
            ;
            
constant    : intConst
                        | floatConst
            | vectorConst
            | ID
            | StringLiteral
            ;

componentblock  : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
                ;
    
zconstraintblock  : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
                  ;

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

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

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

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

moleculestatement : assignment
                  | atomblock
                  | bondblock
                  | bendblock
                  | torsionblock
                  | inversionblock
                  | rigidbodyblock
                  | cutoffgroupblock
                  | fragmentblock
                  | constraintblock
                  ;

atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

atomstatement : assignment
              | POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
              | ORIENTATION^  LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
              ;

                      
bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)?  LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

bondstatement : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!  (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

bendstatement : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

torsionblock  : TORSION^ (LBRACKET! intConst! RBRACKET!)?  LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

torsionstatement  : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

inversionblock  : INVERSION^ (LBRACKET! intConst! RBRACKET!)?  LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

inversionstatement  : assignment
              | CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
              | SATELLITES^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

rigidbodyblock  : RIGIDBODY^  LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
                ;

rigidbodystatement  : assignment
              | MEMBERS^ LPAREN!  inttuple  RPAREN! SEMICOLON!
              ;

cutoffgroupblock  : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
                  ;

cutoffgroupstatement  : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
              ;

fragmentstatement : assignment
              ;

constraintblock : CONSTRAINT^ (LBRACKET! intConst! RBRACKET!)?  LCURLY!(constraintstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
          ;

constraintstatement : assignment
              | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
              ;

              
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:	1979
Committed:	Sat Apr 5 20:56:01 2014 UTC (11 years, 1 month ago) by gezelter
File size:	11684 byte(s)
Log Message:	* Changed the stdDev printouts in RNEMD stats to the 95% confidence intervals * Added the ability to specify non-bonded constraints in a molecule * Added the ability to do selection offsets in the pAngle staticProps module
#	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	gezelter	1979	CONSTRAINT = "constraint";
40			DISTANCE = "distance";
41	tim	770	FRAGMENT = "fragment";
42			MEMBERS = "members";
43	cli2	1275	CENTER = "center";
44	gezelter	1902	SATELLITES = "satellites";
45	tim	770	POSITION = "position";
46			ORIENTATION = "orientation";
47	gezelter	1782	FLUCQ = "flucQ";
48			RNEMD = "RNEMD";
49			MINIMIZER = "minimizer";
50	tim	770	ENDBLOCK;
51			}
52
53
54			mdfile : (statement)*
55			;
56
57			statement : assignment
58	gezelter	1782	\| componentblock
59			\| moleculeblock
60			\| zconstraintblock
61			\| restraintblock
62			\| flucqblock
63			\| rnemdblock
64			\| minimizerblock
65			;
66
67	tim	770	assignment : ID ASSIGNEQUAL^ constant SEMICOLON!
68			;
69
70	tim	814	constant : intConst
71	gezelter	1782	\| floatConst
72			\| vectorConst
73	tim	770	\| ID
74			\| StringLiteral
75			;
76
77			componentblock : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
78			;
79
80			zconstraintblock : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
81			;
82	cli2	1360
83			restraintblock : RESTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
84			;
85	gezelter	1782
86			flucqblock : FLUCQ^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
87			;
88
89			rnemdblock : RNEMD^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
90			;
91
92			minimizerblock : MINIMIZER^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
93			;
94	tim	770
95			moleculeblock : MOLECULE^ LCURLY! (moleculestatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
96			;
97
98			moleculestatement : assignment
99			\| atomblock
100			\| bondblock
101			\| bendblock
102			\| torsionblock
103	gezelter	1277	\| inversionblock
104	tim	770	\| rigidbodyblock
105			\| cutoffgroupblock
106			\| fragmentblock
107	gezelter	1979	\| constraintblock
108	tim	770	;
109
110			atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
111			;
112
113			atomstatement : assignment
114	tim	814	\| POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
115			\| ORIENTATION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
116	tim	770	;
117
118
119			bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
120			;
121
122			bondstatement : assignment
123			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
124			;
125
126			bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
127			;
128
129			bendstatement : assignment
130			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
131			;
132
133			torsionblock : TORSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
134			;
135
136			torsionstatement : assignment
137			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
138			;
139
140	cli2	1275	inversionblock : INVERSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
141			;
142
143			inversionstatement : assignment
144			\| CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
145	gezelter	1902	\| SATELLITES^ LPAREN! inttuple RPAREN! SEMICOLON!
146	cli2	1275	;
147
148	tim	770	rigidbodyblock : RIGIDBODY^ LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
149			;
150
151			rigidbodystatement : assignment
152			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
153			;
154
155			cutoffgroupblock : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
156			;
157
158			cutoffgroupstatement : assignment
159			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
160			;
161
162			fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
163			;
164
165			fragmentstatement : assignment
166			;
167
168	gezelter	1979	constraintblock : CONSTRAINT^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(constraintstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
169			;
170	tim	770
171	gezelter	1979	constraintstatement : assignment
172			\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
173			;
174
175	tim	770
176	tim	814	doubleNumberTuple : doubleNumber (COMMA! doubleNumber)*
177	tim	770	;
178
179			inttuple : intConst (COMMA! intConst)*
180			;
181
182			protected
183			intConst
184	tim	814	: NUM_INT \| NUM_LONG
185	tim	770	;
186
187			protected
188	tim	814	doubleNumber :
189	tim	770	(intConst \| floatConst)
190			;
191
192			protected
193			floatConst
194			:
195	tim	814	NUM_FLOAT \| NUM_DOUBLE
196	tim	770	;
197
198	gezelter	1782	protected
199			vectorConst
200			:
201			LPAREN^ doubleNumber COMMA doubleNumber COMMA doubleNumber RPAREN
202			;
203	tim	770
204			class MDLexer extends Lexer;
205
206			options
207			{
208			k = 3;
209			exportVocab = MD;
210			testLiterals = false;
211			}
212
213			tokens {
214			DOT;
215			}
216
217			{
218
219
220			int deferredLineCount;
221			FilenameObserver* observer;
222
223			public:
224			void setObserver(FilenameObserver* osv) {observer = osv;}
225			void initDeferredLineCount() { deferredLineCount = 0;}
226			void deferredNewline() {
227			deferredLineCount++;
228			}
229
230
231			virtual void newline() {
232			for (;deferredLineCount>0;deferredLineCount--) {
233			CharScanner::newline();
234			}
235			CharScanner::newline();
236			}
237
238			}
239
240
241			// Operators:
242
243			ASSIGNEQUAL : '=' ;
244			COLON : ':' ;
245			COMMA : ',' ;
246			QUESTIONMARK : '?' ;
247			SEMICOLON : ';' ;
248
249			LPAREN : '(' ;
250			RPAREN : ')' ;
251			LBRACKET : '[' ;
252			RBRACKET : ']' ;
253			LCURLY : '{' ;
254			RCURLY : '}' ;
255
256			Whitespace
257			:
258			( // whitespace ignored
259			(' ' \|'\t' \| '\f')
260			\| // handle newlines
261			( '\r' '\n' // MS
262			\| '\r' // Mac
263			\| '\n' // Unix
264			) { newline(); }
265			\| // handle continuation lines
266			( '\\' '\r' '\n' // MS
267			\| '\\' '\r' // Mac
268			\| '\\' '\n' // Unix
269			) {printf("CPP_parser.g continuation line detected\n");
270			deferredNewline();}
271			)
272			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
273			;
274
275			Comment
276			:
277			"/*"
278			( {LA(2) != '/'}? '*'
279			\| EndOfLine {deferredNewline();}
280			\| ~('*'\| '\r' \| '\n')
281			)*
282			"*/" {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
283			;
284
285			CPPComment
286			:
287			"//" (~('\n' \| '\r'))* EndOfLine
288			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
289			;
290
291			PREPROC_DIRECTIVE
292			options{paraphrase = "a line directive";}
293			:
294			'#' LineDirective
295			{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
296			;
297
298			protected
299			LineDirective
300			:
301			{
302			deferredLineCount = 0;
303			}
304			("line")? // this would be for if the directive started "#line"
305			(Space)+
306	gezelter	1390	n:Decimal { setLine(OpenMD::lexi_cast<int>(n->getText()) - 1); }
307	tim	770	(Space)+
308			(sl:StringLiteral) {std::string filename = sl->getText().substr(1,sl->getText().length()-2); observer->notify(filename);}
309			((Space)+ Decimal)* // To support cpp flags (GNU)
310			EndOfLine
311			;
312
313			protected
314			Space
315			:
316			(' '\|'\t'\|'\f')
317			;
318
319
320			// Literals:
321
322			/*
323			* Note that we do NOT handle tri-graphs nor multi-byte sequences.
324			*/
325
326			/*
327			* Note that we can't have empty character constants (even though we
328			* can have empty strings :-).
329			*/
330			CharLiteral
331			:
332			'\'' (Escape \| ~('\'')) '\''
333			;
334
335			/*
336			* Can't have raw imbedded newlines in string constants. Strict reading of
337			* the standard gives odd dichotomy between newlines & carriage returns.
338			* Go figure.
339			*/
340			StringLiteral
341			:
342			'"'
343			( Escape
344			\|
345			( "\\\r\n" // MS
346			\| "\\\r" // MAC
347			\| "\\\n" // Unix
348			) {deferredNewline();}
349			\|
350			~('"'\|'\r'\|'\n'\|'\\')
351			)*
352			'"'
353			;
354
355			protected
356			EndOfLine
357			:
358			( options{generateAmbigWarnings = false;}:
359			"\r\n" // MS
360			\| '\r' // Mac
361			\| '\n' // Unix
362			)
363			;
364
365			/*
366			* Handle the various escape sequences.
367			*
368			* Note carefully that these numeric escape sequences are not of the
369			* same form as the C language numeric constants.
370			*
371			* There is no such thing as a binary numeric escape sequence.
372			*
373			* Octal escape sequences are either 1, 2, or 3 octal digits exactly.
374			*
375			* There is no such thing as a decimal escape sequence.
376			*
377			* Hexadecimal escape sequences are begun with a leading \x and continue
378			* until a non-hexadecimal character is found.
379			*
380			* No real handling of tri-graph sequences, yet.
381			*/
382
383			protected
384			Escape
385			:
386			'\\'
387			( options{warnWhenFollowAmbig=false;}:
388			'a'
389			\| 'b'
390			\| 'f'
391			\| 'n'
392			\| 'r'
393			\| 't'
394			\| 'v'
395			\| '"'
396			\| '\''
397			\| '\\'
398			\| '?'
399			\| ('0'..'3') (options{warnWhenFollowAmbig=false;}: Digit (options{warnWhenFollowAmbig=false;}: Digit)? )?
400			\| ('4'..'7') (options{warnWhenFollowAmbig=false;}: Digit)?
401			\| 'x' (options{warnWhenFollowAmbig=false;}: Digit \| 'a'..'f' \| 'A'..'F')+
402			)
403			;
404
405
406			protected
407	tim	814	Vocabulary
408			:
409			'\3'..'\377'
410			;
411
412
413			ID
414			options {testLiterals = true;}
415			:
416			('a'..'z'\|'A'..'Z'\|'_')
417			('a'..'z'\|'A'..'Z'\|'_'\|'0'..'9')*
418			;
419
420
421			protected
422	tim	770	Digit
423			:
424			'0'..'9'
425			;
426
427			protected
428			Decimal
429			:
430			('0'..'9')+
431			;
432
433	tim	814	// hexadecimal digit (again, note it's protected!)
434	tim	770	protected
435	tim	814	HEX_DIGIT
436			: ('0'..'9'\|'A'..'F'\|'a'..'f')
437			;
438	tim	770
439
440	tim	814	// a numeric literal
441			NUM_INT
442			{
443			bool isDecimal = false;
444			ANTLR_USE_NAMESPACE(antlr)RefToken t = ANTLR_USE_NAMESPACE(antlr)nullToken;
445			}
446			: ('+' \| '-')?
447			(
448			'.' {_ttype = DOT;}
449			( ('0'..'9')+ (EXPONENT)? (f1:FLOAT_SUFFIX {t=f1;})?
450			{
451			if ( t &&
452			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
453			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
454			_ttype = NUM_FLOAT;
455			}
456			else {
457			_ttype = NUM_DOUBLE; // assume double
458			}
459			}
460			)?
461	tim	770
462	tim	814	\| ( '0' {isDecimal = true;} // special case for just '0'
463			( ('x'\|'X')
464			( // hex
465			// the 'e'\|'E' and float suffix stuff look
466			// like hex digits, hence the (...)+ doesn't
467			// know when to stop: ambig. ANTLR resolves
468			// it correctly by matching immediately. It
469			// is therefor ok to hush warning.
470			options {
471			warnWhenFollowAmbig=false;
472			}
473			: HEX_DIGIT
474			)+
475			\| //float or double with leading zero
476			(('0'..'9')+ ('.'\|EXPONENT\|FLOAT_SUFFIX)) => ('0'..'9')+
477			\| ('0'..'7')+ // octal
478			)?
479			\| ('1'..'9') ('0'..'9')* {isDecimal=true;} // non-zero decimal
480			)
481			( ('l'\|'L') { _ttype = NUM_LONG; }
482
483			// only check to see if it's a float if looks like decimal so far
484			\| {isDecimal}?
485			( '.' ('0'..'9')* (EXPONENT)? (f2:FLOAT_SUFFIX {t=f2;})?
486			\| EXPONENT (f3:FLOAT_SUFFIX {t=f3;})?
487			\| f4:FLOAT_SUFFIX {t=f4;}
488			)
489			{
490			if ( t &&
491			(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
492			t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
493			_ttype = NUM_FLOAT;
494			}
495			else {
496			_ttype = NUM_DOUBLE; // assume double
497			}
498			}
499			)?
500			)
501			;
502
503			// a couple protected methods to assist in matching floating point numbers
504	tim	770	protected
505	tim	814	EXPONENT
506	gezelter	848	: ('e'\|'E'\|'d'\|'D') ('+'\|'-')? ('0'..'9')+
507	tim	814	;
508	tim	770
509			protected
510	tim	814	FLOAT_SUFFIX
511			: 'f'\|'F'\|'d'\|'D'
512			;