cpp.g

C Preprocessor,antlr的grammar语言描述,用于学习词法分析,语法分析

            |   ('\\\n') => ('\\\n')
            |   ('\\\r\n') => ('\\\n')
            |   ~'\n'   
            ) +
        ;

SIZEOF	:	'sizeof'
		;

DEFINED	:	'defined'
			{	
				if(!inDirective)
				_type = CKEYWORD;
			}
		;
	
IDENTIFIER		
		:	LETTER (LETTER|'0'..'9')* 
			{
				if(inDirective == true)
				{
					if(cpp == 'D') 
					{
						if(input.LA(1) != '(' )
						{
				 			inDefineMode= true; 
				 			inDirective = false; 
				 			cpp = 'X';
						}
						else 
						{  
							cpp = '('; 
						}
					}
				}
			}
		;

fragment
LETTER	:   '$'
		|   'A'..'Z'
		|   'a'..'z'
		|   '_'
		;

/* Operators: */
					
STRINGIFICATION	:	'#_#_'
				;

SHARPSHARP		:	'##';

ASSIGNEQUAL		:	'='	;

COLON			:	':' ;

COMMA			:	',' ;

QUESTIONMARK	:	'?' ;

SEMICOLON		:	';' ;

POINTERTO		:	'->' ;

LPAREN			:	'(' ;

RPAREN	:	')' 
			{
				if(cpp=='(')
				{
					if(input.LA(1) != '\n')
					{
						inDefineMode = true; 
						inDirective = false; 
					}
					cpp ='X';
				}
			}
		;

LSQUARE			:	'[' ;

RSQUARE			:	']' ;

LCURLY			:	'{' ;

RCURLY			:	'}' ;

EQUAL			:	'==';

NOTEQUAL		:	'!=';

LESSTHANOREQUALTO
				:	'<=';

LESSTHAN 		:	'<' ;

GREATERTHANOREQUALTO
				:	'>=';

GREATERTHAN		:	'>' ;

DIVIDE			:	'/' ;

DIVIDEEQUAL		:	'/=';

PLUS			:	'+' ;

PLUSEQUAL		:	'+=';

PLUSPLUS		:	'++';

MINUS			:	'-' ;

MINUSEQUAL		:	'-=';

MINUSMINUS		:	'--';

STAR			:	'*' ;

TIMESEQUAL		:	'*=';

MOD				:	'%' ;

MODEQUAL		:	'%=';

SHIFTRIGHT		:	'>>';

SHIFTRIGHTEQUAL	:	'>>=';

SHIFTLEFT		:	'<<';

SHIFTLEFTEQUAL	:	'<<=';

AND				:	'&&';

NOT				:	'!' ;

OR				:	'||';

AMPERSAND		:	'&' ;

BITWISEANDEQUAL	:	'&=';

TILDE			:	'~' ;

BITWISEOR		:	'|' ;

BITWISEOREQUAL	:	'|=';

BITWISEXOR		:	'^' ;

BITWISEXOREQUAL	:	'^=';

DOT				:	'.' ;	

POINTERTOMBR	:	'->*' ;

DOTMBR			:	'.*'  ;

SCOPE			:	'::'  ;

ELLIPSIS		:	'...' ;

fragment
INCLUDE_FILE 	
		: '<' ( ' ' | '!' | '#' ..';' | '=' | '?' .. '[' | ']' .. '\u00FF')+ '>'
     	|   '"' ( ' ' | '!' | '#' ..';' | '=' | '?' .. '[' | ']' .. '\u00FF')+ '"'
     	;

CHARACTER_LITERAL
		:	('L')? '\'' ( EscapeSequence | ~('\''|'\\') ) '\''
		;

//STRING_LITERAL	:	'"' (options {greedy=false;} : EscapeSequence | ~('"'))* '"'
STRING_LITERAL	
		:	'"' (EscapeSequence | ~('"'))* '"'
		;

HEX_LITERAL		
		:	'0' ('x'|'X') HexDigit+ IntegerTypeSuffix? 
		;

DECIMAL_LITERAL	
		:	('0' | '1'..'9' '0'..'9'*) IntegerTypeSuffix?
		;

OCTAL_LITERAL	
		:	'0' ('0'..'7')+ IntegerTypeSuffix?
		;

fragment HexDigit		
		:	('0'..'9'|'a'..'f'|'A'..'F') 
		;

fragment IntegerTypeSuffix
		:	'u' | 'ul' | 'U' | 'UL' | 'l' | 'L'	
		;

FLOATING_POINT_LITERAL
		:	('0'..'9')+ '.' ('0'..'9')* Exponent? FloatTypeSuffix?
		|	'.' ('0'..'9')+ Exponent? FloatTypeSuffix?
		|	('0'..'9')+ Exponent FloatTypeSuffix?
		|	('0'..'9')+ FloatTypeSuffix
		;

fragment Exponent		
		:	('e'|'E') ('+'|'-')? ('0'..'9')+ 
		;

fragment FloatTypeSuffix	
		:	('f'|'F'|'d'|'D') 
		;

fragment EscapeSequence	
		: 	'\\' ('b'|'t'|'n'|'f'|'r'|'v'|'\"'|'\''|'\\')
		|	'\\' 'x' HexDigit+
		|   OctalEscape
		;

fragment OctalEscape		
		:	'\\' ('0'..'3') ('0'..'7') ('0'..'7')
		|	'\\' ('0'..'7') ('0'..'7')
		|	'\\' ('0'..'7')
		;

fragment UnicodeEscape	
		:	'\\' 'u' HexDigit HexDigit HexDigit HexDigit
		;

COMMENT			
		:	'/*'  ( options {greedy=false;}:.)* '*/' 			  {_channel=99;}
		|	'/' '\\' '\n' '*'  ( options {greedy=false;}:.)* '*/' {_channel=99;}
		;

LINE_COMMENT	
		:	'//' ~('\n'|'\r')* '\r'? '\n'
		{
			if(!inDirective)
			{
				try{					
						if(input.LT(1) != '#' && input.LT(1) != -1 )
						{
							_type = TEXT_END;
						}
						else
						{
							_type = End;
						}
				}
				catch(Exception e)
				{
					_type = End;
				}
			}
			else
			{
				 _type=End;
			}

			inDirective=false;
			inDefineMode=false;
			discardSpace = true;
		}
		;

WS		:	
		{ 
			if( this.getLine() != 1)
			{
				if(input.LA(-1) == '\n')
					ltoken = End;
				else
					ltoken = input.LA(-1);
			}
			else
			{
				try
				{
					if(input.LA(-1) == '\n')
						ltoken = End;
					else
						ltoken = input.LA(-1);	
				}
				catch (Exception e)
				{
					ltoken = End;
				}
			}

		}
		(' '|'\r'|'\t'|'\f')+
		{			 
			if(inDirective ==  true)
			{
				if(discardSpace == true)
				{
					_channel = 99;
				}
				else
				{
					_type = WS;
				}
			}
			else
			{
				if(!inDefineMode)
				{
					try
					{
						if(input.LA(1) != EOF  && input.LA(1) == '#' &&   ltoken == End )
						{
							_type = End;
						}
					}
					catch (Exception e)
					{
						_channel = 99;
					}
				}
			}
		}
		;

macro_text		
		:	source_text+ -> ^(MACRO_DEFINE source_text+)
		;

End		:	WS?	 '\n'
		{
			if(!inDirective)
			{
				if(inDefineMode) 
				{
					_type=End;
				}
				else
				{
						try{					
								if(input.LT(1) != '#' && input.LT(1) != -1 )
								{
									_type = TEXT_END;
								}
						}
						catch(Exception e)
						{
							_type = End;
						}
				}
			} else {_type=End;}
			ltoken =End;	
			inDirective=false;
			inDefineMode=false;
			discardSpace=true;
		}
		;

ESCAPED_NEWLINE	
		:	('\\\n' | '\\\r\n')
		{
			_channel=99;
		}
		;

COPERATOR       
		:   {!inDirective}? 
        (   COLON               |   QUESTIONMARK		
        |   POINTERTO			|	LCURLY				|	RCURLY			
		|	LSQUARE				|	RSQUARE           	|	STAR
        |   EQUAL               |   NOTEQUAL            |   LESSTHANOREQUALTO   |   LESSTHAN        
        |   GREATERTHANOREQUALTO|   GREATERTHAN         |   DIVIDE              |   PLUSPLUS
        |   MINUSMINUS          |   MOD                 |   SHIFTRIGHT          |   SHIFTLEFT 
        |   AND                 |   OR                  |   BITWISEOR           |   BITWISEXOR      
        |   DOT                 |   POINTERTOMBR        |   DOTMBR              |   SCOPE        
        |   AMPERSAND           |   PLUS                |   MINUS        
        |   TILDE               |   ASSIGNEQUAL         |   TIMESEQUAL          |   DIVIDEEQUAL
        |   MODEQUAL            |   PLUSEQUAL           |   MINUSEQUAL          |   SHIFTLEFTEQUAL 
        |   SHIFTRIGHTEQUAL     |   BITWISEANDEQUAL     |   BITWISEXOREQUAL     |   BITWISEOREQUAL
		|	NOT					|	ELLIPSIS			
        )
        ;

CKEYWORD	:	{!inDirective}?
					'typedef'	| '__va_list__'	| 'extern'	| 'static'	| 'auto'		| 'register' 
				|	'void' 		| 'char'		| 'short'	| 'int'		| 'long'		| 'float' 
				|	'double'	| 'signed'		| 'unsigned'| '__fpreg' | '__float80' 	| 'struct' 
				|	'union'		| 'enum'		| 'const'	| 'volatile'|  'case' 		| 'default'	
				|	'switch'	| 'while'		| 'do' 		| 'for' 	|	'goto'		| 'continue'	
				|	'break'		| 'return'  	| 'if' 		| 'else'	
				;