zhangyiss/toml11
1
0
Fork 0
mirror of https://github.com/ToruNiina/toml11.git synced 2025-09-17 17:58:09 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request #39 from ToruNiina/throw-incorrect-unicode

Browse Source 
Throw syntax_error when parser encounter an incorrect utf-8 codepoint
This commit is contained in:
Toru Niina
2019-03-17 13:12:16 +09:00
committed by GitHub
parent d86870e038 76863cb27f
commit f7380c6e32
1 changed files with 6 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
toml/parser.hpp
View File

@@ -250,11 +250,11 @@ std::string read_utf8_codepoint(const region<Container>& reg,
{ {
if(0xD800 <= codepoint && codepoint <= 0xDFFF) if(0xD800 <= codepoint && codepoint <= 0xDFFF)
{ {
std::cerr << format_underline("[warning] " throw syntax_error(format_underline("[error] "
"toml::read_utf8_codepoint: codepoints in the range " "toml::read_utf8_codepoint: codepoints in the range "
"[0xD800, 0xDFFF] are not valid UTF-8.", {{ "[0xD800, 0xDFFF] are not valid UTF-8.", {{
std::addressof(loc), "not a valid UTF-8 codepoint" std::addressof(loc), "not a valid UTF-8 codepoint"
}}) << std::endl; }}));
} }
assert(codepoint < 0xD800 || 0xDFFF < codepoint); assert(codepoint < 0xD800 || 0xDFFF < codepoint);
// 1110yyyy 10yxxxxx 10xxxxxx // 1110yyyy 10yxxxxx 10xxxxxx
@@ -262,15 +262,8 @@ std::string read_utf8_codepoint(const region<Container>& reg,
character += static_cast<unsigned char>(0x80|(codepoint >> 6 & 0x3F)); character += static_cast<unsigned char>(0x80|(codepoint >> 6 & 0x3F));
character += static_cast<unsigned char>(0x80|(codepoint & 0x3F)); character += static_cast<unsigned char>(0x80|(codepoint & 0x3F));
} }
else if(codepoint < 0x200000) // U+010000 ... U+1FFFFF else if(codepoint < 0x110000) // U+010000 ... U+10FFFF
{ {
if(0x10FFFF < codepoint) // out of Unicode region
{
std::cerr << format_underline("[error] "
"toml::read_utf8_codepoint: input codepoint is too large to "
"decode as a unicode character.", {{std::addressof(loc),
"should be in [0x00..0x10FFFF]"}}) << std::endl;
}
// 11110yyy 10yyxxxx 10xxxxxx 10xxxxxx // 11110yyy 10yyxxxx 10xxxxxx 10xxxxxx
character += static_cast<unsigned char>(0xF0| codepoint >> 18); character += static_cast<unsigned char>(0xF0| codepoint >> 18);
character += static_cast<unsigned char>(0x80|(codepoint >> 12 & 0x3F)); character += static_cast<unsigned char>(0x80|(codepoint >> 12 & 0x3F));
@@ -279,9 +272,9 @@ std::string read_utf8_codepoint(const region<Container>& reg,
} }
else // out of UTF-8 region else // out of UTF-8 region
{ {
throw std::range_error(format_underline(concat_to_string("[error] " throw syntax_error(format_underline("[error] toml::read_utf8_codepoint:"
"input codepoint (", str, ") is too large to encode as utf-8."), " input codepoint is too large.",
{{std::addressof(reg), "should be in [0x00..0x10FFFF]"}})); {{std::addressof(loc), "should be in [0x00..0x10FFFF]"}}));
} }
return character; return character;
} }