【发布时间】：2016-03-11 00:59:29
【问题描述】：
全部，
我一直在尝试根据标准文档 N4567 创建 C++ 语法，这是我能找到的最新版本。我相信语法是完整的，但我需要测试它。我试图解决的一个问题是让词法分析器从标准中识别原始字符串。我已经使用 Actions & Semantic Predicates 实现了一个可能的解决方案。我需要帮助确定它是否真的有效。我已经阅读了关于动作和谓词之间交互的 ANTLR4 参考，但可以确定我的解决方案是否有效。下面包括一个精简语法。任何想法将不胜感激。我试图在示例中包含我的想法。
grammar SampleRaw;
@lexer::members {
string d_char_seq = "";
}
string_literal
: ENCODING_PREFIX? '\"' S_CHAR* '\"'
| ENCODING_PREFIX? 'R' Raw_String
;
ENCODING_PREFIX // one of
: 'u8'
| [uUL]
;
S_CHAR /* any member of the source character set except the
double_quote ", backslash \, or NEW_LINE character
*/
: ~[\"\\\n\r]
| ESCAPE_SEQUENCE
| UNIV_CHAR_NAME
;
fragment ESCAPE_SEQUENCE
: SIMPLE_ESCAPE_SEQ
| OCT_ESCAPE_SEQ
| HEX_ESCAPE_SEQ
;
fragment SIMPLE_ESCAPE_SEQ // one of
: '\\' '\''
| '\\' '\"'
| '\\' '?'
| '\\' '\\'
| '\\' 'a'
| '\\' 'b'
| '\\' 'f'
| '\\' 'n'
| '\\' 'r'
| '\\' 't'
| '\\' 'v'
;
fragment OCT_ESCAPE_SEQ
: [0-3] ( OCT_DIGIT OCT_DIGIT? )?
| [4-7] ( OCT_DIGIT )?
;
fragment HEX_ESCAPE_SEQ
: '\\' 'x' HEX_DIGIT+
;
fragment UNIV_CHAR_NAME
: '\\' 'u' HEX_QUAD
| '\\' 'U' HEX_QUAD HEX_QUAD
;
fragment HEX_QUAD
: HEX_DIGIT HEX_DIGIT HEX_DIGIT HEX_DIGIT
;
fragment HEX_DIGIT
: [a-zA-Z0-9]
;
fragment OCT_DIGIT
: [0-7]
;
/*
Raw_String
: '\"' D_CHAR* '(' R_CHAR* ')' D_CHAR* '\"'
;
*/
Raw_String
: ( /* CASE when D_CHAR is empty
ACTION in D_CHAR_SEQ attempts to reset variable d_char_seq
if it is empty, so handle it staticly
*/
'\"'
'('
( ~[)] // Anything but )
| [)] ~[\"] // ) Actually OK, can't be followed by "
// - )" - these are the terminating chars
)*
')'
'\"'
| '\"'
D_CHAR_SEQ /* Will the ACTION in D_CHAR_SEQ be an issue for
the Semantic Predicates Below????
*/
'('
( ~[)] // Anything but )
| [)] D_CHAR_SEQ { ( getText() != d_char_seq ) }?
/* ) Actually OK, can't be followed D_CHAR_SEQ match
IF D_CHAR_SEQs match, turn OFF the Alternative
*/
| [)] D_CHAR_SEQ { ( getText() == d_char_seq ) }? ~[\"]
/* ) Actually OK, must be followed D_CHAR_SEQ match
IF D_CHAR_SEQs match, turn ON the Alternative
Cant't match the final " , but
WE HAVE MATCHED OUR TERMINATING CHARS
*/
)*
')'
D_CHAR_SEQ /* No need to check here,
Matching Terminating CHARS is only way to get out
of loop above
*/
'\"'
)
{ d_char_seq = ""; } // Reset Variable
;
/*
fragment R_CHAR
// any member of the source character set, except a right
// parenthesis ) followed by the initial D_CHAR*
// (which may be empty) followed by a double quote ".
//
: ~[)]
;
*/
fragment D_CHAR
/* any member of the basic source character set except
space, the left parenthesis (, the right parenthesis ),
the backslash \, and the control characters representing
horizontal tab, vertical tab, form feed, and newline.
*/
: ~[ )(\\\t\v\f\n\r]
;
fragment D_CHAR_SEQ
: D_CHAR+ { d_char_seq = ( d_char_seq == "" ) ? getText() : d_char_seq ; }
;
【问题讨论】：
-
希望你只是在修补。一个完整的 C++ 解析器实际上是很多 棘手的 工作：请参阅stackoverflow.com/questions/243383/…。你知道 ANTLR3 已经有部分 C++ 语法可用了吗？
-
是的，你是对的。一个完整的 C++ 解析器需要做很多工作，而我可能永远无法完成这项任务。然而，像我这样的齿轮头/极客需要在我的停机时间做点什么。
标签： c++ c++11 antlr antlr4 lexer