Skip to content

J
json
Commit ec7a1d83 by Robert Marki
Options

Fix character skipping after a surrogate pair

In a string the first character following a surrogate pair is skipped by the lexer, but the rest of the string is parsed as usual.
parent 39486303
Showing with 9 additions and 4 deletions
src/json.hpp
...@@ -6856,8 +6856,8 @@ basic_json_parser_59: ...@@ -6856,8 +6856,8 @@ basic_json_parser_59:
auto codepoint2 = std::strtoul(std::string(reinterpret_cast<typename string_t::const_pointer> auto codepoint2 = std::strtoul(std::string(reinterpret_cast<typename string_t::const_pointer>
(i + 7), 4).c_str(), nullptr, 16); (i + 7), 4).c_str(), nullptr, 16);
result += to_unicode(codepoint, codepoint2); result += to_unicode(codepoint, codepoint2);
// skip the next 11 characters (xxxx\uyyyy) // skip the next 10 characters (xxxx\uyyyy)
i += 11; i += 10;
} }
else else
{ {
......
src/json.hpp.re2c
...@@ -6162,8 +6162,8 @@ class basic_json ...@@ -6162,8 +6162,8 @@ class basic_json
auto codepoint2 = std::strtoul(std::string(reinterpret_cast<typename string_t::const_pointer> auto codepoint2 = std::strtoul(std::string(reinterpret_cast<typename string_t::const_pointer>
(i + 7), 4).c_str(), nullptr, 16); (i + 7), 4).c_str(), nullptr, 16);
result += to_unicode(codepoint, codepoint2); result += to_unicode(codepoint, codepoint2);
// skip the next 11 characters (xxxx\uyyyy) // skip the next 10 characters (xxxx\uyyyy)
i += 11; i += 10;
} }
else else
{ {
......
test/unit.cpp
...@@ -10205,4 +10205,9 @@ TEST_CASE("regression tests") ...@@ -10205,4 +10205,9 @@ TEST_CASE("regression tests")
j["string"] = bytes; j["string"] = bytes;
CHECK(j["string"] == "\u0007\u0007"); CHECK(j["string"] == "\u0007\u0007");
} }
SECTION("character following a surrogate pair is skipped")
{
CHECK(json::parse("\"\\ud80c\\udc60abc\"").get<json::string_t>() == u8"\U00013060abc");
}
} }
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment