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";
  FIXED             = "Fixed";
  HARMONIC          = "Harmonic";
  CUBIC             = "Cubic";
  QUARTIC           = "Quartic";
  POLYNOMIAL        = "Polynomial";
  MORSE             = "Morse";
  GHOSTBEND         = "GhostBend";
  UREYBRADLEY       = "UreyBradley";
  COSINE            = "Cosine";
  GHOSTTORSION      = "GhostTorsion";
  CHARMM            = "Charmm";
  OPLS              = "Opls";
  TRAPPE            = "Trappe";
  AMBERIMPROPER     = "AmberImproper";
  IMPROPERCOSINE    = "ImproperCosine";
  CENTRALATOMHEIGHT = "CentralAtomHeight";
  DREIDING          = "Dreiding";
  CHARGE            = "charge";
  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!
    | CHARGE^ LPAREN! floatConst RPAREN! SEMICOLON!
    ;

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

bondstatement : assignment
    | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
    | FIXED^ LPAREN! floatConst RPAREN! SEMICOLON!
    | HARMONIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | CUBIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | QUARTIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | POLYNOMIAL^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | MORSE^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!        
    ;

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

bendstatement : assignment
    | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
    | HARMONIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | GHOSTBEND^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | UREYBRADLEY^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | CUBIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | QUARTIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | POLYNOMIAL^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | COSINE^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!            
    ;

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

torsionstatement  : assignment
    | MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
    | GHOSTTORSION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | CUBIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | QUARTIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | POLYNOMIAL^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | CHARMM^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | OPLS^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | TRAPPE^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | HARMONIC^ LPAREN! doubleNumberTuple 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!
    | AMBERIMPROPER^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | IMPROPERCOSINE^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | HARMONIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | CENTRALATOMHEIGHT^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
    | DREIDING^ LPAREN! doubleNumberTuple 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
    ;
*/

protected
vectorConst
    :
        LPAREN^ doubleNumberTuple 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:	2046
Committed:	Tue Dec 2 22:11:04 2014 UTC (10 years, 7 months ago) by gezelter
File size:	13843 byte(s)
Log Message:	Fixed some broken comments for use with Doxygen. Made changes to allow topology-based force-field overrides in include files. Fixed a calculation of box quadrupole moments for molecules with point dipoles.
#	Content
1	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	RESTRAINT = "restraint";
32	ATOM = "atom";
33	BOND = "bond";
34	BEND = "bend";
35	TORSION = "torsion";
36	INVERSION = "inversion";
37	RIGIDBODY = "rigidBody";
38	CUTOFFGROUP = "cutoffGroup";
39	CONSTRAINT = "constraint";
40	DISTANCE = "distance";
41	FRAGMENT = "fragment";
42	MEMBERS = "members";
43	CENTER = "center";
44	SATELLITES = "satellites";
45	POSITION = "position";
46	ORIENTATION = "orientation";
47	FLUCQ = "flucQ";
48	RNEMD = "RNEMD";
49	MINIMIZER = "minimizer";
50	FIXED = "Fixed";
51	HARMONIC = "Harmonic";
52	CUBIC = "Cubic";
53	QUARTIC = "Quartic";
54	POLYNOMIAL = "Polynomial";
55	MORSE = "Morse";
56	GHOSTBEND = "GhostBend";
57	UREYBRADLEY = "UreyBradley";
58	COSINE = "Cosine";
59	GHOSTTORSION = "GhostTorsion";
60	CHARMM = "Charmm";
61	OPLS = "Opls";
62	TRAPPE = "Trappe";
63	AMBERIMPROPER = "AmberImproper";
64	IMPROPERCOSINE = "ImproperCosine";
65	CENTRALATOMHEIGHT = "CentralAtomHeight";
66	DREIDING = "Dreiding";
67	CHARGE = "charge";
68	ENDBLOCK;
69	}
70
71
72	mdfile : (statement)*
73	;
74
75	statement : assignment
76	\| componentblock
77	\| moleculeblock
78	\| zconstraintblock
79	\| restraintblock
80	\| flucqblock
81	\| rnemdblock
82	\| minimizerblock
83	;
84
85	assignment : ID ASSIGNEQUAL^ constant SEMICOLON!
86	;
87
88	constant : intConst
89	\| floatConst
90	\| vectorConst
91	\| ID
92	\| StringLiteral
93	;
94
95	componentblock : COMPONENT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
96	;
97
98	zconstraintblock : ZCONSTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
99	;
100
101	restraintblock : RESTRAINT^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
102	;
103
104	flucqblock : FLUCQ^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
105	;
106
107	rnemdblock : RNEMD^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
108	;
109
110	minimizerblock : MINIMIZER^ LCURLY! (assignment)* RCURLY {#RCURLY->setType(ENDBLOCK);}
111	;
112
113	moleculeblock : MOLECULE^ LCURLY! (moleculestatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
114	;
115
116	moleculestatement : assignment
117	\| atomblock
118	\| bondblock
119	\| bendblock
120	\| torsionblock
121	\| inversionblock
122	\| rigidbodyblock
123	\| cutoffgroupblock
124	\| fragmentblock
125	\| constraintblock
126	;
127
128	atomblock : ATOM^ LBRACKET! intConst RBRACKET! LCURLY! (atomstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
129	;
130
131	atomstatement : assignment
132	\| POSITION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
133	\| ORIENTATION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
134	\| CHARGE^ LPAREN! floatConst RPAREN! SEMICOLON!
135	;
136
137
138	bondblock : BOND^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(bondstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
139	;
140
141	bondstatement : assignment
142	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
143	\| FIXED^ LPAREN! floatConst RPAREN! SEMICOLON!
144	\| HARMONIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
145	\| CUBIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
146	\| QUARTIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
147	\| POLYNOMIAL^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
148	\| MORSE^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
149	;
150
151	bendblock : BEND^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (bendstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
152	;
153
154	bendstatement : assignment
155	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
156	\| HARMONIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
157	\| GHOSTBEND^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
158	\| UREYBRADLEY^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
159	\| CUBIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
160	\| QUARTIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
161	\| POLYNOMIAL^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
162	\| COSINE^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
163	;
164
165	torsionblock : TORSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(torsionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
166	;
167
168	torsionstatement : assignment
169	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
170	\| GHOSTTORSION^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
171	\| CUBIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
172	\| QUARTIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
173	\| POLYNOMIAL^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
174	\| CHARMM^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
175	\| OPLS^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
176	\| TRAPPE^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
177	\| HARMONIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
178	;
179
180	inversionblock : INVERSION^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(inversionstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
181	;
182
183	inversionstatement : assignment
184	\| CENTER^ LPAREN! intConst RPAREN! SEMICOLON!
185	\| SATELLITES^ LPAREN! inttuple RPAREN! SEMICOLON!
186	\| AMBERIMPROPER^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
187	\| IMPROPERCOSINE^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
188	\| HARMONIC^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
189	\| CENTRALATOMHEIGHT^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
190	\| DREIDING^ LPAREN! doubleNumberTuple RPAREN! SEMICOLON!
191	;
192
193	rigidbodyblock : RIGIDBODY^ LBRACKET! intConst RBRACKET! LCURLY!(rigidbodystatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
194	;
195
196	rigidbodystatement : assignment
197	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
198	;
199
200	cutoffgroupblock : CUTOFFGROUP^ (LBRACKET! intConst! RBRACKET!)? LCURLY! (cutoffgroupstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
201	;
202
203	cutoffgroupstatement : assignment
204	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
205	;
206
207	fragmentblock : FRAGMENT^ LBRACKET! intConst RBRACKET! LCURLY! (fragmentstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
208	;
209
210	fragmentstatement : assignment
211	;
212
213	constraintblock : CONSTRAINT^ (LBRACKET! intConst! RBRACKET!)? LCURLY!(constraintstatement)* RCURLY {#RCURLY->setType(ENDBLOCK);}
214	;
215
216	constraintstatement : assignment
217	\| MEMBERS^ LPAREN! inttuple RPAREN! SEMICOLON!
218	;
219
220
221	doubleNumberTuple : doubleNumber (COMMA! doubleNumber)*
222	;
223
224	inttuple : intConst (COMMA! intConst)*
225	;
226
227	protected
228	intConst
229	:
230	NUM_INT \| NUM_LONG
231	;
232
233	protected
234	doubleNumber
235	:
236	(intConst \| floatConst)
237	;
238
239	protected
240	floatConst
241	:
242	NUM_FLOAT \| NUM_DOUBLE
243	;
244
245
246	/*
247	protected
248	vectorConst
249	:
250	LPAREN^ doubleNumber COMMA doubleNumber COMMA doubleNumber RPAREN
251	;
252	*/
253
254	protected
255	vectorConst
256	:
257	LPAREN^ doubleNumberTuple RPAREN
258	;
259
260
261	class MDLexer extends Lexer;
262
263	options
264	{
265	k = 3;
266	exportVocab = MD;
267	testLiterals = false;
268	}
269
270	tokens {
271	DOT;
272	}
273
274	{
275
276
277	int deferredLineCount;
278	FilenameObserver* observer;
279
280	public:
281	void setObserver(FilenameObserver* osv) {observer = osv;}
282	void initDeferredLineCount() { deferredLineCount = 0;}
283	void deferredNewline() {
284	deferredLineCount++;
285	}
286
287
288	virtual void newline() {
289	for (;deferredLineCount>0;deferredLineCount--) {
290	CharScanner::newline();
291	}
292	CharScanner::newline();
293	}
294
295	}
296
297
298	// Operators:
299
300	ASSIGNEQUAL : '=' ;
301	COLON : ':' ;
302	COMMA : ',' ;
303	QUESTIONMARK : '?' ;
304	SEMICOLON : ';' ;
305
306	LPAREN : '(' ;
307	RPAREN : ')' ;
308	LBRACKET : '[' ;
309	RBRACKET : ']' ;
310	LCURLY : '{' ;
311	RCURLY : '}' ;
312
313	Whitespace
314	:
315	( // whitespace ignored
316	(' ' \|'\t' \| '\f')
317	\| // handle newlines
318	( '\r' '\n' // MS
319	\| '\r' // Mac
320	\| '\n' // Unix
321	) { newline(); }
322	\| // handle continuation lines
323	( '\\' '\r' '\n' // MS
324	\| '\\' '\r' // Mac
325	\| '\\' '\n' // Unix
326	) {printf("CPP_parser.g continuation line detected\n");
327	deferredNewline();}
328	)
329	{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
330	;
331
332	Comment
333	:
334	"/*"
335	( {LA(2) != '/'}? '*'
336	\| EndOfLine {deferredNewline();}
337	\| ~('*'\| '\r' \| '\n')
338	)*
339	"*/" {_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP;}
340	;
341
342	CPPComment
343	:
344	"//" (~('\n' \| '\r'))* EndOfLine
345	{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
346	;
347
348	PREPROC_DIRECTIVE
349	options{paraphrase = "a line directive";}
350	:
351	'#' LineDirective
352	{_ttype = ANTLR_USE_NAMESPACE(antlr)Token::SKIP; newline();}
353	;
354
355	protected
356	LineDirective
357	:
358	{
359	deferredLineCount = 0;
360	}
361	("line")? // this would be for if the directive started "#line"
362	(Space)+
363	n:Decimal { setLine(OpenMD::lexi_cast<int>(n->getText()) - 1); }
364	(Space)+
365	(sl:StringLiteral) {std::string filename = sl->getText().substr(1,sl->getText().length()-2); observer->notify(filename);}
366	((Space)+ Decimal)* // To support cpp flags (GNU)
367	EndOfLine
368	;
369
370	protected
371	Space
372	:
373	(' '\|'\t'\|'\f')
374	;
375
376
377	// Literals:
378
379	/*
380	* Note that we do NOT handle tri-graphs nor multi-byte sequences.
381	*/
382
383	/*
384	* Note that we can't have empty character constants (even though we
385	* can have empty strings :-).
386	*/
387	CharLiteral
388	:
389	'\'' (Escape \| ~('\'')) '\''
390	;
391
392	/*
393	* Can't have raw imbedded newlines in string constants. Strict reading of
394	* the standard gives odd dichotomy between newlines & carriage returns.
395	* Go figure.
396	*/
397	StringLiteral
398	:
399	'"'
400	( Escape
401	\|
402	( "\\\r\n" // MS
403	\| "\\\r" // MAC
404	\| "\\\n" // Unix
405	) {deferredNewline();}
406	\|
407	~('"'\|'\r'\|'\n'\|'\\')
408	)*
409	'"'
410	;
411
412	protected
413	EndOfLine
414	:
415	( options{generateAmbigWarnings = false;}:
416	"\r\n" // MS
417	\| '\r' // Mac
418	\| '\n' // Unix
419	)
420	;
421
422	/*
423	* Handle the various escape sequences.
424	*
425	* Note carefully that these numeric escape sequences are not of the
426	* same form as the C language numeric constants.
427	*
428	* There is no such thing as a binary numeric escape sequence.
429	*
430	* Octal escape sequences are either 1, 2, or 3 octal digits exactly.
431	*
432	* There is no such thing as a decimal escape sequence.
433	*
434	* Hexadecimal escape sequences are begun with a leading \x and continue
435	* until a non-hexadecimal character is found.
436	*
437	* No real handling of tri-graph sequences, yet.
438	*/
439
440	protected
441	Escape
442	:
443	'\\'
444	( options{warnWhenFollowAmbig=false;}:
445	'a'
446	\| 'b'
447	\| 'f'
448	\| 'n'
449	\| 'r'
450	\| 't'
451	\| 'v'
452	\| '"'
453	\| '\''
454	\| '\\'
455	\| '?'
456	\| ('0'..'3') (options{warnWhenFollowAmbig=false;}: Digit (options{warnWhenFollowAmbig=false;}: Digit)? )?
457	\| ('4'..'7') (options{warnWhenFollowAmbig=false;}: Digit)?
458	\| 'x' (options{warnWhenFollowAmbig=false;}: Digit \| 'a'..'f' \| 'A'..'F')+
459	)
460	;
461
462
463	protected
464	Vocabulary
465	:
466	'\3'..'\377'
467	;
468
469
470	ID
471	options {testLiterals = true;}
472	:
473	('a'..'z'\|'A'..'Z'\|'_')
474	('a'..'z'\|'A'..'Z'\|'_'\|'0'..'9')*
475	;
476
477
478	protected
479	Digit
480	:
481	'0'..'9'
482	;
483
484	protected
485	Decimal
486	:
487	('0'..'9')+
488	;
489
490	// hexadecimal digit (again, note it's protected!)
491	protected
492	HEX_DIGIT
493	: ('0'..'9'\|'A'..'F'\|'a'..'f')
494	;
495
496
497	// a numeric literal
498	NUM_INT
499	{
500	bool isDecimal = false;
501	ANTLR_USE_NAMESPACE(antlr)RefToken t = ANTLR_USE_NAMESPACE(antlr)nullToken;
502	}
503	: ('+' \| '-')?
504	(
505	'.' {_ttype = DOT;}
506	( ('0'..'9')+ (EXPONENT)? (f1:FLOAT_SUFFIX {t=f1;})?
507	{
508	if ( t &&
509	(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
510	t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
511	_ttype = NUM_FLOAT;
512	}
513	else {
514	_ttype = NUM_DOUBLE; // assume double
515	}
516	}
517	)?
518
519	\| ( '0' {isDecimal = true;} // special case for just '0'
520	( ('x'\|'X')
521	( // hex
522	// the 'e'\|'E' and float suffix stuff look
523	// like hex digits, hence the (...)+ doesn't
524	// know when to stop: ambig. ANTLR resolves
525	// it correctly by matching immediately. It
526	// is therefor ok to hush warning.
527	options {
528	warnWhenFollowAmbig=false;
529	}
530	: HEX_DIGIT
531	)+
532	\| //float or double with leading zero
533	(('0'..'9')+ ('.'\|EXPONENT\|FLOAT_SUFFIX)) => ('0'..'9')+
534	\| ('0'..'7')+ // octal
535	)?
536	\| ('1'..'9') ('0'..'9')* {isDecimal=true;} // non-zero decimal
537	)
538	( ('l'\|'L') { _ttype = NUM_LONG; }
539
540	// only check to see if it's a float if looks like decimal so far
541	\| {isDecimal}?
542	( '.' ('0'..'9')* (EXPONENT)? (f2:FLOAT_SUFFIX {t=f2;})?
543	\| EXPONENT (f3:FLOAT_SUFFIX {t=f3;})?
544	\| f4:FLOAT_SUFFIX {t=f4;}
545	)
546	{
547	if ( t &&
548	(t->getText().find('f') != ANTLR_USE_NAMESPACE(std)string::npos \|\|
549	t->getText().find('F') != ANTLR_USE_NAMESPACE(std)string::npos ) ) {
550	_ttype = NUM_FLOAT;
551	}
552	else {
553	_ttype = NUM_DOUBLE; // assume double
554	}
555	}
556	)?
557	)
558	;
559
560	// a couple protected methods to assist in matching floating point numbers
561	protected
562	EXPONENT
563	: ('e'\|'E'\|'d'\|'D') ('+'\|'-')? ('0'..'9')+
564	;
565
566	protected
567	FLOAT_SUFFIX
568	: 'f'\|'F'\|'d'\|'D'
569	;