Compare commits

...
Author SHA1 Message Date
Claude cf14cbdd02 Deduplicate UBJSON/BJData length-type error message
The "expected length type specification" parse error was built with an
identical if/else block in both get_ubjson_string() and
get_ubjson_size_value(), each branching on the input format to pick the
marker list. Extract a single unexpected_length_type_message() helper that
selects the markers via a ternary and appends the caller-supplied infix,
removing four near-duplicate message strings and centralizing the marker
list. Error messages are unchanged.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Tv6SnfYbcx8y3y4PrECcto
2026-07-22 21:03:03 +00:00
Angadi56andGitHub 3565f40229 reject negative UBJSON/BJData string length (#5284) 2026-07-20 20:32:38 +00:00
4 changed files with 136 additions and 52 deletions
+49 -26
View File
@@ -1846,6 +1846,47 @@ class binary_reader
return get_ubjson_value(get_char ? get_ignore_noop() : current);
}
/*!
@brief reject a negative UBJSON/BJData string length
String and key lengths are written with signed integer markers (i, I, l,
L). A negative value is malformed; without this check get_string() would
silently treat it as an empty string and leave the following bytes to be
misread as the next value. This mirrors the non-negative check the
optimized-container count path already performs in get_ubjson_size_value.
@param[in] len the string length read from the input
@return whether the length is valid (non-negative)
*/
template<typename NumberType>
bool check_ubjson_string_length(const NumberType len)
{
if (JSON_HEDLEY_UNLIKELY(len < 0))
{
return sax->parse_error(chars_read, get_token_string(), parse_error::create(113, chars_read,
exception_message(input_format, "string length must not be negative", "string"), nullptr));
}
return true;
}
/*!
@brief create the error message for a missing/invalid length-type marker
Used both when reading a UBJSON/BJData string and when reading an optimized
container size. The accepted markers depend on the input format, and the
size variant appends " after '#'" via @a infix.
@param[in] last_token the offending byte as returned by get_token_string()
@param[in] infix extra context inserted after the marker list
@return the formatted error message
*/
std::string unexpected_length_type_message(const std::string& last_token, const char* infix) const
{
return concat("expected length type specification (",
input_format == input_format_t::bjdata ? "U, i, u, I, m, l, M, L" : "U, i, I, l, L",
")", infix, "; last byte: 0x", last_token);
}
/*!
@brief reads a UBJSON string
@@ -1883,25 +1924,25 @@ class binary_reader
case 'i':
{
std::int8_t len{};
return get_number(input_format, len) && get_string(input_format, len, result);
return get_number(input_format, len) && check_ubjson_string_length(len) && get_string(input_format, len, result);
}
case 'I':
{
std::int16_t len{};
return get_number(input_format, len) && get_string(input_format, len, result);
return get_number(input_format, len) && check_ubjson_string_length(len) && get_string(input_format, len, result);
}
case 'l':
{
std::int32_t len{};
return get_number(input_format, len) && get_string(input_format, len, result);
return get_number(input_format, len) && check_ubjson_string_length(len) && get_string(input_format, len, result);
}
case 'L':
{
std::int64_t len{};
return get_number(input_format, len) && get_string(input_format, len, result);
return get_number(input_format, len) && check_ubjson_string_length(len) && get_string(input_format, len, result);
}
case 'u':
@@ -1938,17 +1979,8 @@ class binary_reader
break;
}
auto last_token = get_token_string();
std::string message;
if (input_format != input_format_t::bjdata)
{
message = "expected length type specification (U, i, I, l, L); last byte: 0x" + last_token;
}
else
{
message = "expected length type specification (U, i, u, I, m, l, M, L); last byte: 0x" + last_token;
}
return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, exception_message(input_format, message, "string"), nullptr));
return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read,
exception_message(input_format, unexpected_length_type_message(last_token, ""), "string"), nullptr));
}
/*!
@@ -2227,17 +2259,8 @@ class binary_reader
break;
}
auto last_token = get_token_string();
std::string message;
if (input_format != input_format_t::bjdata)
{
message = "expected length type specification (U, i, I, l, L) after '#'; last byte: 0x" + last_token;
}
else
{
message = "expected length type specification (U, i, u, I, m, l, M, L) after '#'; last byte: 0x" + last_token;
}
return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, exception_message(input_format, message, "size"), nullptr));
return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read,
exception_message(input_format, unexpected_length_type_message(last_token, " after '#'"), "size"), nullptr));
}
/*!
+49 -26
View File
@@ -12395,6 +12395,47 @@ class binary_reader
return get_ubjson_value(get_char ? get_ignore_noop() : current);
}
/*!
@brief reject a negative UBJSON/BJData string length
String and key lengths are written with signed integer markers (i, I, l,
L). A negative value is malformed; without this check get_string() would
silently treat it as an empty string and leave the following bytes to be
misread as the next value. This mirrors the non-negative check the
optimized-container count path already performs in get_ubjson_size_value.
@param[in] len the string length read from the input
@return whether the length is valid (non-negative)
*/
template<typename NumberType>
bool check_ubjson_string_length(const NumberType len)
{
if (JSON_HEDLEY_UNLIKELY(len < 0))
{
return sax->parse_error(chars_read, get_token_string(), parse_error::create(113, chars_read,
exception_message(input_format, "string length must not be negative", "string"), nullptr));
}
return true;
}
/*!
@brief create the error message for a missing/invalid length-type marker
Used both when reading a UBJSON/BJData string and when reading an optimized
container size. The accepted markers depend on the input format, and the
size variant appends " after '#'" via @a infix.
@param[in] last_token the offending byte as returned by get_token_string()
@param[in] infix extra context inserted after the marker list
@return the formatted error message
*/
std::string unexpected_length_type_message(const std::string& last_token, const char* infix) const
{
return concat("expected length type specification (",
input_format == input_format_t::bjdata ? "U, i, u, I, m, l, M, L" : "U, i, I, l, L",
")", infix, "; last byte: 0x", last_token);
}
/*!
@brief reads a UBJSON string
@@ -12432,25 +12473,25 @@ class binary_reader
case 'i':
{
std::int8_t len{};
return get_number(input_format, len) && get_string(input_format, len, result);
return get_number(input_format, len) && check_ubjson_string_length(len) && get_string(input_format, len, result);
}
case 'I':
{
std::int16_t len{};
return get_number(input_format, len) && get_string(input_format, len, result);
return get_number(input_format, len) && check_ubjson_string_length(len) && get_string(input_format, len, result);
}
case 'l':
{
std::int32_t len{};
return get_number(input_format, len) && get_string(input_format, len, result);
return get_number(input_format, len) && check_ubjson_string_length(len) && get_string(input_format, len, result);
}
case 'L':
{
std::int64_t len{};
return get_number(input_format, len) && get_string(input_format, len, result);
return get_number(input_format, len) && check_ubjson_string_length(len) && get_string(input_format, len, result);
}
case 'u':
@@ -12487,17 +12528,8 @@ class binary_reader
break;
}
auto last_token = get_token_string();
std::string message;
if (input_format != input_format_t::bjdata)
{
message = "expected length type specification (U, i, I, l, L); last byte: 0x" + last_token;
}
else
{
message = "expected length type specification (U, i, u, I, m, l, M, L); last byte: 0x" + last_token;
}
return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, exception_message(input_format, message, "string"), nullptr));
return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read,
exception_message(input_format, unexpected_length_type_message(last_token, ""), "string"), nullptr));
}
/*!
@@ -12776,17 +12808,8 @@ class binary_reader
break;
}
auto last_token = get_token_string();
std::string message;
if (input_format != input_format_t::bjdata)
{
message = "expected length type specification (U, i, I, l, L) after '#'; last byte: 0x" + last_token;
}
else
{
message = "expected length type specification (U, i, u, I, m, l, M, L) after '#'; last byte: 0x" + last_token;
}
return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, exception_message(input_format, message, "size"), nullptr));
return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read,
exception_message(input_format, unexpected_length_type_message(last_token, " after '#'"), "size"), nullptr));
}
/*!
+13
View File
@@ -2721,6 +2721,19 @@ TEST_CASE("BJData")
CHECK_THROWS_WITH_AS(_ = json::from_bjdata(v), "[json.exception.parse_error.113] parse error at byte 2: syntax error while parsing BJData string: expected length type specification (U, i, u, I, m, l, M, L); last byte: 0x31", json::parse_error&);
}
SECTION("negative length")
{
json _;
std::vector<uint8_t> const vi = {'S', 'i', 0xFF};
CHECK_THROWS_WITH_AS(_ = json::from_bjdata(vi), "[json.exception.parse_error.113] parse error at byte 3: syntax error while parsing BJData string: string length must not be negative", json::parse_error&);
CHECK(json::from_bjdata(vi, true, false).is_discarded());
std::vector<uint8_t> const vl = {'S', 'l', 0xFF, 0xFF, 0xFF, 0xFF};
CHECK_THROWS_WITH_AS(_ = json::from_bjdata(vl), "[json.exception.parse_error.113] parse error at byte 6: syntax error while parsing BJData string: string length must not be negative", json::parse_error&);
CHECK(json::from_bjdata(vl, true, false).is_discarded());
}
SECTION("parse bjdata markers in ubjson")
{
// create a single-character string for all number types
+25
View File
@@ -1862,6 +1862,31 @@ TEST_CASE("UBJSON")
json _;
CHECK_THROWS_WITH_AS(_ = json::from_ubjson(v), "[json.exception.parse_error.113] parse error at byte 2: syntax error while parsing UBJSON string: expected length type specification (U, i, I, l, L); last byte: 0x31", json::parse_error&);
}
SECTION("negative length")
{
json _;
std::vector<uint8_t> const vi = {'S', 'i', 0xFF};
CHECK_THROWS_WITH_AS(_ = json::from_ubjson(vi), "[json.exception.parse_error.113] parse error at byte 3: syntax error while parsing UBJSON string: string length must not be negative", json::parse_error&);
CHECK(json::from_ubjson(vi, true, false).is_discarded());
std::vector<uint8_t> const vI = {'S', 'I', 0xFF, 0xFF};
CHECK_THROWS_WITH_AS(_ = json::from_ubjson(vI), "[json.exception.parse_error.113] parse error at byte 4: syntax error while parsing UBJSON string: string length must not be negative", json::parse_error&);
CHECK(json::from_ubjson(vI, true, false).is_discarded());
std::vector<uint8_t> const vl = {'S', 'l', 0xFF, 0xFF, 0xFF, 0xFF};
CHECK_THROWS_WITH_AS(_ = json::from_ubjson(vl), "[json.exception.parse_error.113] parse error at byte 6: syntax error while parsing UBJSON string: string length must not be negative", json::parse_error&);
CHECK(json::from_ubjson(vl, true, false).is_discarded());
std::vector<uint8_t> const vL = {'S', 'L', 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF};
CHECK_THROWS_WITH_AS(_ = json::from_ubjson(vL), "[json.exception.parse_error.113] parse error at byte 10: syntax error while parsing UBJSON string: string length must not be negative", json::parse_error&);
CHECK(json::from_ubjson(vL, true, false).is_discarded());
// a length of zero remains valid and yields an empty string
std::vector<uint8_t> const v0 = {'S', 'i', 0};
CHECK(json::from_ubjson(v0) == json(""));
}
}
SECTION("array")