avoid recursion in the scanner
This commit is contained in:
Niels
parent
b64367e2f7
commit
2478be807c
2 changed files with 705 additions and 678 deletions
1256
src/json.hpp
1256
src/json.hpp
File diff suppressed because it is too large
Load diff
|
|
@ -7393,76 +7393,81 @@ class basic_json
|
|||
*/
|
||||
token_type scan() noexcept
|
||||
{
|
||||
// pointer for backtracking information
|
||||
m_marker = nullptr;
|
||||
while (true)
|
||||
{
|
||||
// pointer for backtracking information
|
||||
m_marker = nullptr;
|
||||
|
||||
// remember the begin of the token
|
||||
m_start = m_cursor;
|
||||
assert(m_start != nullptr);
|
||||
// remember the begin of the token
|
||||
m_start = m_cursor;
|
||||
assert(m_start != nullptr);
|
||||
|
||||
/*!re2c
|
||||
re2c:define:YYCTYPE = lexer_char_t;
|
||||
re2c:define:YYCURSOR = m_cursor;
|
||||
re2c:define:YYLIMIT = m_limit;
|
||||
re2c:define:YYMARKER = m_marker;
|
||||
re2c:define:YYFILL = "yyfill(); // LCOV_EXCL_LINE";
|
||||
re2c:yyfill:parameter = 0;
|
||||
re2c:indent:string = " ";
|
||||
re2c:indent:top = 1;
|
||||
re2c:labelprefix = "basic_json_parser_";
|
||||
/*!re2c
|
||||
re2c:define:YYCTYPE = lexer_char_t;
|
||||
re2c:define:YYCURSOR = m_cursor;
|
||||
re2c:define:YYLIMIT = m_limit;
|
||||
re2c:define:YYMARKER = m_marker;
|
||||
re2c:define:YYFILL = "yyfill(); // LCOV_EXCL_LINE";
|
||||
re2c:yyfill:parameter = 0;
|
||||
re2c:indent:string = " ";
|
||||
re2c:indent:top = 1;
|
||||
re2c:labelprefix = "basic_json_parser_";
|
||||
|
||||
// ignore whitespace
|
||||
ws = [ \t\n\r]+;
|
||||
ws { return scan(); }
|
||||
// ignore whitespace
|
||||
ws = [ \t\n\r]+;
|
||||
ws { continue; }
|
||||
|
||||
// ignore byte-order-mark
|
||||
bom = "\xEF\xBB\xBF";
|
||||
bom { return scan(); }
|
||||
// ignore byte-order-mark
|
||||
bom = "\xEF\xBB\xBF";
|
||||
bom { continue; }
|
||||
|
||||
// structural characters
|
||||
"[" { return token_type::begin_array; }
|
||||
"]" { return token_type::end_array; }
|
||||
"{" { return token_type::begin_object; }
|
||||
"}" { return token_type::end_object; }
|
||||
"," { return token_type::value_separator; }
|
||||
":" { return token_type::name_separator; }
|
||||
// structural characters
|
||||
"[" { last_token_type = token_type::begin_array; break; }
|
||||
"]" { last_token_type = token_type::end_array; break; }
|
||||
"{" { last_token_type = token_type::begin_object; break; }
|
||||
"}" { last_token_type = token_type::end_object; break; }
|
||||
"," { last_token_type = token_type::value_separator; break; }
|
||||
":" { last_token_type = token_type::name_separator; break; }
|
||||
|
||||
// literal names
|
||||
"null" { return token_type::literal_null; }
|
||||
"true" { return token_type::literal_true; }
|
||||
"false" { return token_type::literal_false; }
|
||||
// literal names
|
||||
"null" { last_token_type = token_type::literal_null; break; }
|
||||
"true" { last_token_type = token_type::literal_true; break; }
|
||||
"false" { last_token_type = token_type::literal_false; break; }
|
||||
|
||||
// number
|
||||
decimal_point = [.];
|
||||
digit = [0-9];
|
||||
digit_1_9 = [1-9];
|
||||
e = [eE];
|
||||
minus = [-];
|
||||
plus = [+];
|
||||
zero = [0];
|
||||
exp = e (minus|plus)? digit+;
|
||||
frac = decimal_point digit+;
|
||||
int = (zero|digit_1_9 digit*);
|
||||
number = minus? int frac? exp?;
|
||||
number { return token_type::value_number; }
|
||||
// number
|
||||
decimal_point = [.];
|
||||
digit = [0-9];
|
||||
digit_1_9 = [1-9];
|
||||
e = [eE];
|
||||
minus = [-];
|
||||
plus = [+];
|
||||
zero = [0];
|
||||
exp = e (minus|plus)? digit+;
|
||||
frac = decimal_point digit+;
|
||||
int = (zero|digit_1_9 digit*);
|
||||
number = minus? int frac? exp?;
|
||||
number { last_token_type = token_type::value_number; break; }
|
||||
|
||||
// string
|
||||
quotation_mark = ["];
|
||||
escape = [\\];
|
||||
unescaped = [^"\\\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F];
|
||||
single_escaped = ["\\/bfnrt];
|
||||
unicode_escaped = [u][0-9a-fA-F]{4};
|
||||
escaped = escape (single_escaped | unicode_escaped);
|
||||
char = unescaped | escaped;
|
||||
string = quotation_mark char* quotation_mark;
|
||||
string { return token_type::value_string; }
|
||||
// string
|
||||
quotation_mark = ["];
|
||||
escape = [\\];
|
||||
unescaped = [^"\\\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F];
|
||||
single_escaped = ["\\/bfnrt];
|
||||
unicode_escaped = [u][0-9a-fA-F]{4};
|
||||
escaped = escape (single_escaped | unicode_escaped);
|
||||
char = unescaped | escaped;
|
||||
string = quotation_mark char* quotation_mark;
|
||||
string { last_token_type = token_type::value_string; break; }
|
||||
|
||||
// end of file
|
||||
'\000' { return token_type::end_of_input; }
|
||||
// end of file
|
||||
'\000' { last_token_type = token_type::end_of_input; break; }
|
||||
|
||||
// anything else is an error
|
||||
. { return token_type::parse_error; }
|
||||
*/
|
||||
// anything else is an error
|
||||
. { last_token_type = token_type::parse_error; break; }
|
||||
*/
|
||||
}
|
||||
|
||||
return last_token_type;
|
||||
}
|
||||
|
||||
/// append data from the stream to the internal buffer
|
||||
|
|
@ -7810,6 +7815,8 @@ class basic_json
|
|||
const lexer_char_t* m_cursor = nullptr;
|
||||
/// pointer to the end of the buffer
|
||||
const lexer_char_t* m_limit = nullptr;
|
||||
/// the last token type
|
||||
token_type last_token_type = token_type::end_of_input;
|
||||
};
|
||||
|
||||
/*!
|
||||
|
|
|
|||
Loading…
Reference in a new issue