lex.hpp File Reference

Detailed Description

Tools for parsing C-like input.

Functions in this file use objects of opaque type lexer_t. To create a lexer_t instance, use create_lexer().

Classes
struct	token_t
	Parser token. More...
struct	cast_t
	Preprocessor cast. More...
struct	lex_value_t

Functions
idaman lexer_t *ida_export	create_lexer (const char *const *keys, size_t size, void *ud=NULL)
	Create new lexical analyzer and set its keyword table. More...
idaman void ida_export	destroy_lexer (lexer_t *lx)
	Destroy a lexical analyzer.
idaman error_t ida_export	lex_define_macro (lexer_t *lx, const char *macro, const char *body, int nargs=0, bool isfunc=false)
	Define a macro.
idaman void ida_export	lex_undefine_macro (lexer_t *lx, const char *macro)
	Undefine a macro.
idaman int ida_export	lex_set_options (lexer_t *lx, int options)
	Set lexer options. More...
idaman error_t ida_export	lex_get_token (lexer_t *lx, token_t *t)
	Get next token.
idaman int ida_export	lex_enum_macros (const lexer_t *lx, int idaapi cb(const char *name, const char *body, int nargs, bool isfunc, void *ud), void *ud=NULL)
	Enumerate all macros. More...
idaman const char *ida_export	lex_print_token (qstring *buf, const token_t *t)
	Debug: get text representation of token.
String oriented functions
idaman error_t ida_export	lex_init_string (lexer_t *lx, const char *line, macro_t **macros=NULL)
	Set the input line and the macro table. More...
File oriented functions
idaman error_t ida_export	lex_init_file (lexer_t *lx, const char *file)
	Initialization: file may be NULL. More...
idaman const char *ida_export	lex_get_file_line (lexer_t *lx, int32 *linenum, const char **lineptr, int level=0)
	Error handling. More...
idaman void ida_export	lex_term_file (lexer_t *lx, bool del_macros)
	Termination: also see lex_init_file()

Token stack
typedef qstack< token_t >	tokenstack_t
	see get_token(), unget_token()
bool	get_token (token_t *t, lexer_t *lx, tokenstack_t &buf)
	Retrieve token from a stack or lexer. More...
void	unget_token (const token_t &t, tokenstack_t &buf)
	Push a token back onto the token stack.

Macros
#define	LXOPT_PARSE_FLOATS   0x0001
	enable floating point constants
#define	LXOPT_REQ_SEPARATOR   0x0002
	require a separator between a number and an ident or a character/string constant or dot
#define	LXOPT_NOCASE_FILES   0x0004
	case-insensitive file search
#define	LXOPT_C99_CONSTANTS   0x0008
	the size and sign of constants depend on the value itself and the 'U', 'L', and 'LL' More...

Typedefs
typedef ushort	lxtype
	see Parser token types
typedef error_t	lx_resolver_t(lexer_t *lx, void *ud, token_t *curtok, sval_t *res)
	Preprocessor callback for unknown tokens. More...
typedef error_t	lx_parse_cast_t(lexer_t *lx, cast_t *cast, token_t *ct)
	Preprocessor callbacks for casts. More...
typedef int idaapi	lx_preprocessor_cb(void *ud, const char *fname, int nl, const char *line)
	Preprocessor callback. More...
typedef int idaapi	lx_pragma_cb(void *ud, const char *line)
	Callback for #pragma directives. More...
typedef int idaapi	lx_warning_cb(void *ud, const char *line)
	Callback for #warning directives. More...
typedef int idaapi	lx_macro_cb(void *ud, const char *name, const char *body, int nargs, bool isfunc, bool is_new_macro)
	Callback for #define directives. More...
typedef int idaapi	lx_undef_cb(void *ud, const char *name)
	Callback for #undef directives. More...

Variables
const lxtype	lx_end = 1
	no more tokens
const lxtype	lx_ident = 2
	ident
const lxtype	lx_number = 3
	long constant
const lxtype	lx_string = 4
	string constant (token_t.chr != 0 => unicode string)
const lxtype	lx_char = 5
	char constant
const lxtype	lx_typename = 6
	user-defined type
const lxtype	lx_float = 7
	IEEE floating point constant.
const lxtype	lx_int64 = 8
	int64 constant
const lxtype	lx_key = 128
	keywords start. More...

Typedef Documentation

typedef error_t lx_resolver_t(lexer_t *lx, void *ud, token_t *curtok, sval_t *res)

Preprocessor callback for unknown tokens.

Will be called when preprocessor calculates the value of #if expression.

typedef error_t lx_parse_cast_t(lexer_t *lx, cast_t *cast, token_t *ct)

Preprocessor callbacks for casts.

Will be called when preprocessor calculates the value of #if expression.

typedef int idaapi lx_preprocessor_cb(void *ud, const char *fname, int nl, const char *line)

Preprocessor callback.

It will be called for each input line.

Returns

an error code (0-ok)

typedef int idaapi lx_pragma_cb(void *ud, const char *line)

Callback for #pragma directives.

Returns

an error code (0-ok)

typedef int idaapi lx_warning_cb(void *ud, const char *line)

Callback for #warning directives.

Returns

an error code (0-ok)

typedef int idaapi lx_macro_cb(void *ud, const char *name, const char *body, int nargs, bool isfunc, bool is_new_macro)

Callback for #define directives.

Returns

an error code (0-ok)

typedef int idaapi lx_undef_cb(void *ud, const char *name)

Callback for #undef directives.

Returns

an error code (0-ok)

Function Documentation

idaman lexer_t* ida_export create_lexer	(	const char *const *	keys,
		size_t	size,
		void *	ud = NULL
	)

Create new lexical analyzer and set its keyword table.

If keys==NULL, then set the default C keyword table

idaman int ida_export lex_set_options	(	lexer_t *	lx,
		int	options
	)

Set lexer options.

Parameters

options

Lexer options

Returns

the old options

idaman int ida_export lex_enum_macros	(	const lexer_t *	lx,
		int idaapi	cbconst char *name, const char *body, int nargs, bool isfunc, void *ud,
		void *	ud = NULL
	)

Enumerate all macros.

Do so until 'cb' returns non-zero.

idaman error_t ida_export lex_init_string	(	lexer_t *	lx,
		const char *	line,
		macro_t **	macros = NULL
	)

Set the input line and the macro table.

if macros==NULL, the macro table will not be changed.

idaman error_t ida_export lex_init_file	(	lexer_t *	lx,
		const char *	file
	)

Initialization: file may be NULL.

Also see lex_term_file().

idaman const char* ida_export lex_get_file_line	(	lexer_t *	lx,
		int32 *	linenum,
		const char **	lineptr,
		int	level = 0
	)

Error handling.

if level > 0, then return information about the enclosing file which included the current one.

bool get_token ( token_t *  t, lexer_t *  lx, tokenstack_t &  buf  )

inline

Retrieve token from a stack or lexer.

If buf is not empty then get the token on top of the stack. If buf is empty then gen the next token from the lexer.

Returns

success