clean up
This commit is contained in:
Niels
parent
b21bf95620
commit
2fc82358ce
3 changed files with 759 additions and 779 deletions
1355
src/json.hpp
1355
src/json.hpp
File diff suppressed because it is too large
Load diff
|
|
@ -2456,7 +2456,10 @@ class basic_json
|
|||
end_of_input
|
||||
};
|
||||
|
||||
inline lexer(const char* s) : m_content(s)
|
||||
/// the char type to use in the lexer
|
||||
using lexer_char_t = typename string_t::value_type;
|
||||
|
||||
inline lexer(const typename string_t::value_type* s) : m_content(s)
|
||||
{
|
||||
m_start = m_cursor = m_content;
|
||||
m_limit = m_content + strlen(m_content);
|
||||
|
|
@ -2464,46 +2467,39 @@ class basic_json
|
|||
|
||||
inline lexer() = default;
|
||||
|
||||
template<typename CharT>
|
||||
inline static std::basic_string<CharT> to_unicode(const long codepoint)
|
||||
inline static string_t to_unicode(const long codepoint)
|
||||
{
|
||||
std::string result;
|
||||
string_t result;
|
||||
|
||||
if (codepoint <= 0x7f)
|
||||
{
|
||||
// 1-byte (ASCII) characters: 0xxxxxxx
|
||||
result.append(1, static_cast<char>(codepoint));
|
||||
// 1-byte characters: 0xxxxxxx (ASCI)
|
||||
result.append(1, static_cast<typename string_t::value_type>(codepoint));
|
||||
}
|
||||
else if (codepoint <= 0x7ff)
|
||||
{
|
||||
// 2-byte characters: 110xxxxx 10xxxxxx
|
||||
// the 0xC0 enables the two most significant bits to make this
|
||||
// a 2-byte UTF-8 character
|
||||
result.append(1, static_cast<CharT>(0xC0 | ((codepoint >> 6) & 0x1F)));
|
||||
result.append(1, static_cast<CharT>(0x80 | (codepoint & 0x3F)));
|
||||
result.append(1, static_cast<typename string_t::value_type>(0xC0 | ((codepoint >> 6) & 0x1F)));
|
||||
result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F)));
|
||||
}
|
||||
else if (codepoint <= 0xffff)
|
||||
{
|
||||
// 3-byte characters: 1110xxxx 10xxxxxx 10xxxxxx
|
||||
// the 0xE0 enables the three most significant bits to make
|
||||
// this a 3-byte UTF-8 character
|
||||
result.append(1, static_cast<CharT>(0xE0 | ((codepoint >> 12) & 0x0F)));
|
||||
result.append(1, static_cast<CharT>(0x80 | ((codepoint >> 6) & 0x3F)));
|
||||
result.append(1, static_cast<CharT>(0x80 | (codepoint & 0x3F)));
|
||||
result.append(1, static_cast<typename string_t::value_type>(0xE0 | ((codepoint >> 12) & 0x0F)));
|
||||
result.append(1, static_cast<typename string_t::value_type>(0x80 | ((codepoint >> 6) & 0x3F)));
|
||||
result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F)));
|
||||
}
|
||||
else if (codepoint <= 0x10ffff)
|
||||
{
|
||||
// 4-byte characters: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
|
||||
// the 0xF0 enables the four most significant bits to make this
|
||||
// a 4-byte UTF-8 character
|
||||
result.append(1, static_cast<CharT>(0xF0 | ((codepoint >> 18) & 0x07)));
|
||||
result.append(1, static_cast<CharT>(0x80 | ((codepoint >> 12) & 0x3F)));
|
||||
result.append(1, static_cast<CharT>(0x80 | ((codepoint >> 6) & 0x3F)));
|
||||
result.append(1, static_cast<CharT>(0x80 | (codepoint & 0x3F)));
|
||||
result.append(1, static_cast<typename string_t::value_type>(0xF0 | ((codepoint >> 18) & 0x07)));
|
||||
result.append(1, static_cast<typename string_t::value_type>(0x80 | ((codepoint >> 12) & 0x3F)));
|
||||
result.append(1, static_cast<typename string_t::value_type>(0x80 | ((codepoint >> 6) & 0x3F)));
|
||||
result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F)));
|
||||
}
|
||||
else
|
||||
{
|
||||
throw std::out_of_range("code point is invalid");
|
||||
throw std::out_of_range("code points above 0x10FFFF are invalid");
|
||||
}
|
||||
|
||||
return result;
|
||||
|
|
@ -2557,77 +2553,74 @@ class basic_json
|
|||
inline token_type scan()
|
||||
{
|
||||
// pointer for backtracking information
|
||||
const char* m_marker = nullptr;
|
||||
const typename string_t::value_type* m_marker = nullptr;
|
||||
|
||||
while (true)
|
||||
{
|
||||
// remember the begin of the token
|
||||
m_start = m_cursor;
|
||||
// remember the begin of the token
|
||||
m_start = m_cursor;
|
||||
|
||||
/*!re2c
|
||||
re2c:define:YYCTYPE = char;
|
||||
re2c:define:YYCURSOR = m_cursor;
|
||||
re2c:define:YYLIMIT = m_limit;
|
||||
re2c:define:YYMARKER = m_marker;
|
||||
re2c:indent:string = " ";
|
||||
re2c:indent:top = 1;
|
||||
re2c:labelprefix = "basic_json_parser_";
|
||||
re2c:yyfill:enable = 0;
|
||||
/*!re2c
|
||||
re2c:define:YYCTYPE = lexer_char_t;
|
||||
re2c:define:YYCURSOR = m_cursor;
|
||||
re2c:define:YYLIMIT = m_limit;
|
||||
re2c:define:YYMARKER = m_marker;
|
||||
re2c:indent:string = " ";
|
||||
re2c:indent:top = 1;
|
||||
re2c:labelprefix = "basic_json_parser_";
|
||||
re2c:yyfill:enable = 0;
|
||||
|
||||
// whitespace
|
||||
ws = [ \t\n\r]+;
|
||||
ws { continue; }
|
||||
// whitespace
|
||||
ws = [ \t\n\r]+;
|
||||
ws { return scan(); }
|
||||
|
||||
// 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
|
||||
"[" { 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; }
|
||||
|
||||
// literal names
|
||||
"null" { return token_type::literal_null; }
|
||||
"true" { return token_type::literal_true; }
|
||||
"false" { return token_type::literal_false; }
|
||||
// literal names
|
||||
"null" { return token_type::literal_null; }
|
||||
"true" { return token_type::literal_true; }
|
||||
"false" { return token_type::literal_false; }
|
||||
|
||||
// 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 { return token_type::value_number; }
|
||||
|
||||
// string
|
||||
quotation_mark = [\"];
|
||||
escape = [\\];
|
||||
unescaped = [^\"\\\000];
|
||||
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 = [^\"\\\000];
|
||||
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; }
|
||||
|
||||
// end of file
|
||||
'\000' { return token_type::end_of_input; }
|
||||
// end of file
|
||||
'\000' { return token_type::end_of_input; }
|
||||
|
||||
// anything else is an error
|
||||
. { return token_type::parse_error; }
|
||||
*/
|
||||
}
|
||||
// anything else is an error
|
||||
. { return token_type::parse_error; }
|
||||
*/
|
||||
}
|
||||
|
||||
inline std::string get_token() const
|
||||
inline string_t get_token() const
|
||||
{
|
||||
return std::string(m_start, static_cast<size_t>(m_cursor - m_start));
|
||||
return string_t(m_start, static_cast<size_t>(m_cursor - m_start));
|
||||
}
|
||||
|
||||
/*!
|
||||
|
|
@ -2638,16 +2631,14 @@ class basic_json
|
|||
from the pointer difference of the two pointers).
|
||||
|
||||
@return string value of current token without opening and closing quotes
|
||||
|
||||
@todo Take care of Unicode.
|
||||
*/
|
||||
inline std::string get_string() const
|
||||
inline string_t get_string() const
|
||||
{
|
||||
std::string result;
|
||||
string_t result;
|
||||
result.reserve(static_cast<size_t>(m_cursor - m_start - 2));
|
||||
|
||||
// iterate the result between the quotes
|
||||
for (const char* i = m_start + 1; i < m_cursor - 1; ++i)
|
||||
for (const typename string_t::value_type* i = m_start + 1; i < m_cursor - 1; ++i)
|
||||
{
|
||||
// process escaped characters
|
||||
if (*i == '\\')
|
||||
|
|
@ -2707,7 +2698,7 @@ class basic_json
|
|||
// get code xxxx from \uxxxx
|
||||
auto codepoint = strtol(i + 1, nullptr, 16);
|
||||
// add unicode character(s)
|
||||
result += to_unicode<char>(codepoint);
|
||||
result += to_unicode(codepoint);
|
||||
// skip the next four characters (\uxxxx)
|
||||
i += 4;
|
||||
break;
|
||||
|
|
@ -2746,20 +2737,20 @@ class basic_json
|
|||
|
||||
private:
|
||||
/// the buffer
|
||||
const char* m_content = nullptr;
|
||||
const typename string_t::value_type* m_content = nullptr;
|
||||
/// pointer to he beginning of the current symbol
|
||||
const char* m_start = nullptr;
|
||||
const typename string_t::value_type* m_start = nullptr;
|
||||
/// pointer to the current symbol
|
||||
const char* m_cursor = nullptr;
|
||||
const typename string_t::value_type* m_cursor = nullptr;
|
||||
/// pointer to the end of the buffer
|
||||
const char* m_limit = nullptr;
|
||||
const typename string_t::value_type* m_limit = nullptr;
|
||||
};
|
||||
|
||||
class parser
|
||||
{
|
||||
public:
|
||||
/// constructor for strings
|
||||
inline parser(const std::string& s) : m_buffer(s), m_lexer(m_buffer.c_str())
|
||||
inline parser(const string_t& s) : m_buffer(s), m_lexer(m_buffer.c_str())
|
||||
{
|
||||
// read first token
|
||||
get_token();
|
||||
|
|
@ -2770,7 +2761,7 @@ class basic_json
|
|||
{
|
||||
while (_is)
|
||||
{
|
||||
std::string input_line;
|
||||
string_t input_line;
|
||||
std::getline(_is, input_line);
|
||||
m_buffer += input_line;
|
||||
}
|
||||
|
|
@ -2964,7 +2955,7 @@ class basic_json
|
|||
|
||||
private:
|
||||
/// the buffer
|
||||
std::string m_buffer;
|
||||
string_t m_buffer;
|
||||
/// the type of the last read token
|
||||
typename lexer::token_type last_token = lexer::token_type::uninitialized;
|
||||
/// the lexer
|
||||
|
|
|
|||
|
|
@ -5517,8 +5517,8 @@ TEST_CASE("lexer class")
|
|||
|
||||
SECTION("to_unicode")
|
||||
{
|
||||
CHECK(json::lexer::to_unicode<char>(0x1F4A9) == "💩");
|
||||
CHECK_THROWS_AS(json::lexer::to_unicode<char>(0x110000), std::out_of_range);
|
||||
CHECK(json::lexer::to_unicode(0x1F4A9) == "💩");
|
||||
CHECK_THROWS_AS(json::lexer::to_unicode(0x200000), std::out_of_range);
|
||||
}
|
||||
}
|
||||
|
||||
|
|
|
|||
Loading…
Reference in a new issue