j3d1/json
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Merge pull request #534 from TedLyngmo/utf8_encoding_enhancement

Browse source 
UTF8 encoding enhancement
This commit is contained in:
Niels Lohmann 2017-03-24 19:38:29 +01:00 committed by GitHub
commit bba688f53e
2 changed files with 6 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
src/json.hpp
View file

@ -10344,20 +10344,20 @@ class basic_json
else if (codepoint <= 0x7ff) else if (codepoint <= 0x7ff)
{ {
// 2-byte characters: 110xxxxx 10xxxxxx // 2-byte characters: 110xxxxx 10xxxxxx
result.append(1, static_cast<typename string_t::value_type>(0xC0 | ((codepoint >> 6) & 0x1F))); result.append(1, static_cast<typename string_t::value_type>(0xC0 | (codepoint >> 6)));
result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F))); result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F)));
} }
else if (codepoint <= 0xffff) else if (codepoint <= 0xffff)
{ {
// 3-byte characters: 1110xxxx 10xxxxxx 10xxxxxx // 3-byte characters: 1110xxxx 10xxxxxx 10xxxxxx
result.append(1, static_cast<typename string_t::value_type>(0xE0 | ((codepoint >> 12) & 0x0F))); result.append(1, static_cast<typename string_t::value_type>(0xE0 | (codepoint >> 12)));
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 >> 6) & 0x3F)));
result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F))); result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F)));
} }
else if (codepoint <= 0x10ffff) else if (codepoint <= 0x10ffff)
{ {
// 4-byte characters: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx // 4-byte characters: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
result.append(1, static_cast<typename string_t::value_type>(0xF0 | ((codepoint >> 18) & 0x07))); result.append(1, static_cast<typename string_t::value_type>(0xF0 | (codepoint >> 18)));
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 >> 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 >> 6) & 0x3F)));
result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F))); result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F)));

6
src/json.hpp.re2c
View file

@ -10344,20 +10344,20 @@ class basic_json
else if (codepoint <= 0x7ff) else if (codepoint <= 0x7ff)
{ {
// 2-byte characters: 110xxxxx 10xxxxxx // 2-byte characters: 110xxxxx 10xxxxxx
result.append(1, static_cast<typename string_t::value_type>(0xC0 | ((codepoint >> 6) & 0x1F))); result.append(1, static_cast<typename string_t::value_type>(0xC0 | (codepoint >> 6)));
result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F))); result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F)));
} }
else if (codepoint <= 0xffff) else if (codepoint <= 0xffff)
{ {
// 3-byte characters: 1110xxxx 10xxxxxx 10xxxxxx // 3-byte characters: 1110xxxx 10xxxxxx 10xxxxxx
result.append(1, static_cast<typename string_t::value_type>(0xE0 | ((codepoint >> 12) & 0x0F))); result.append(1, static_cast<typename string_t::value_type>(0xE0 | (codepoint >> 12)));
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 >> 6) & 0x3F)));
result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F))); result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F)));
} }
else if (codepoint <= 0x10ffff) else if (codepoint <= 0x10ffff)
{ {
// 4-byte characters: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx // 4-byte characters: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
result.append(1, static_cast<typename string_t::value_type>(0xF0 | ((codepoint >> 18) & 0x07))); result.append(1, static_cast<typename string_t::value_type>(0xF0 | (codepoint >> 18)));
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 >> 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 >> 6) & 0x3F)));
result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F))); result.append(1, static_cast<typename string_t::value_type>(0x80 | (codepoint & 0x3F)));