Compare commits

..
Author SHA1 Message Date
Niels Lohmann cc749ac699 Keep full byte2 x byte3 combinatorics in the wrong-fourth-byte sections
The maintainer wants exhaustive coverage of every byte combination here
rather than the representative-prefix reduction, matching the style of
the sibling "wrong second/third byte" sections in the same files.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>
2026-09-05 23:18:51 +02:00
Niels Lohmann 1c6ca81b8a Fix dead ill-formed-fourth-byte UTF-8 test sections (byte3/byte4 typo)
The "ill-formed: wrong fourth byte" SECTIONs in unit-unicode3.cpp,
unit-unicode4.cpp, and unit-unicode5.cpp guarded their loop with a check
on byte3 instead of byte4. Since the enclosing loop already restricts
byte3 to its valid range, the guard was always true and the section's
"continue" fired unconditionally, so check_utf8string()/check_utf8dump()
were never actually invoked for a malformed fourth byte.

Fixing the guard naively (byte3 -> byte4) would also have swept the full
byte2 x byte3 combinatorics for every byte4 value, adding millions of
redundant iterations: the lexer validates continuation bytes strictly in
sequence with early exit (see next_byte_in_range() in lexer.hpp), so once
byte2/byte3 are within their valid range, the byte4 outcome does not
depend on which valid byte2/byte3 values were chosen. Instead, byte2 and
byte3 are now held to a small hedge of representative valid prefixes
(range corners plus a midpoint) while byte4 is still swept exhaustively
over its full 0x00-0xFF range, since that is the actual property under
test. Also fixed the garbled "skip fourth second byte" comment in
unit-unicode3.cpp.

Verified offline: before the fix, the "wrong fourth byte" subcase
executes 0 assertions in all three files (proving it was dead code);
after the fix, it executes 11520 (unicode3), 34560 (unicode4), and 11520
(unicode5) assertions, and a deliberately reintroduced bug in the
lexer's byte4 range check causes it to fail (proving it is now
meaningful). Total per-file assertion counts grow by the same small
amounts, not by millions, and all other sections in these files still
pass unchanged.

Fixes #5416

Signed-off-by: Niels Lohmann <mail@nlohmann.me>
2026-09-05 23:05:42 +02:00
4 changed files with 7 additions and 118 deletions
+3 -114
View File
@@ -38,26 +38,6 @@ class huge_binary_t : public std::vector<std::uint8_t>
using huge_binary_json = nlohmann::basic_json <
std::map, std::vector, std::string, bool, std::int64_t, std::uint64_t,
double, std::allocator, nlohmann::adl_serializer, huge_binary_t, void >;
// a string type that reports a size beyond INT32_MAX without allocating that
// much memory, so BSON length overflow can be tested for strings and
// (embedded) documents as well, following the same idea as huge_binary_t
class huge_string_t : public std::string
{
public:
using std::string::string;
huge_string_t(const std::string& s) : std::string(s) {} // NOLINT(google-explicit-constructor,hicpp-explicit-conversions)
size_type size() const noexcept // NOLINT(readability-convert-member-functions-to-static)
{
// one byte more than the BSON length field can represent
return static_cast<size_type>((std::numeric_limits<std::int32_t>::max)()) + 1;
}
};
using huge_string_json = nlohmann::basic_json <
std::map, std::vector, huge_string_t, bool, std::int64_t, std::uint64_t,
double, std::allocator, nlohmann::adl_serializer, std::vector<std::uint8_t>, void >;
} // namespace
TEST_CASE("BSON")
@@ -125,36 +105,10 @@ TEST_CASE("BSON")
SECTION("lengths exceeding INT32_MAX cannot be serialized to BSON")
{
// out_of_range.412 is thrown from a single shared helper
// (to_bson_length) that guards the BSON length fields of binary
// values, strings, and (embedded) documents alike
SECTION("binary")
{
huge_binary_json j;
j["b"] = huge_binary_json::binary(huge_binary_t{});
huge_binary_json j;
j["b"] = huge_binary_json::binary(huge_binary_t{});
CHECK_THROWS_WITH_AS(huge_binary_json::to_bson(j), "[json.exception.out_of_range.412] BSON length 2147483661 exceeds maximum of 2147483647", huge_binary_json::out_of_range&);
}
SECTION("string")
{
huge_string_json j;
j["s"] = huge_string_json::string_t("value");
CHECK_THROWS_WITH_AS(huge_string_json::to_bson(j), "[json.exception.out_of_range.412] BSON length 4294967308 exceeds maximum of 2147483647", huge_string_json::out_of_range&);
}
SECTION("document")
{
// an oversized string nested one level deep makes the
// *embedded* document's own length exceed INT32_MAX as well
huge_string_json nested;
nested["s"] = huge_string_json::string_t("value");
huge_string_json j;
j["nested"] = nested;
CHECK_THROWS_WITH_AS(huge_string_json::to_bson(j), "[json.exception.out_of_range.412] BSON length 6442450963 exceeds maximum of 2147483647", huge_string_json::out_of_range&);
}
CHECK_THROWS_WITH_AS(huge_binary_json::to_bson(j), "[json.exception.out_of_range.412] BSON length 2147483661 exceeds maximum of 2147483647", huge_binary_json::out_of_range&);
}
SECTION("string length must be at least 1")
@@ -239,23 +193,6 @@ TEST_CASE("BSON")
CHECK(json::from_bson(result, true, false) == j);
}
SECTION("non-empty object with bool from a non-0/1 byte (lenient parsing)")
{
// documented lenient behavior (see gh-5333): any non-zero byte
// is accepted as `true`, not just 0x01
std::vector<std::uint8_t> const input =
{
0x0D, 0x00, 0x00, 0x00, // size (little endian)
0x08, // entry: boolean
'e', 'n', 't', 'r', 'y', '\x00',
0x02, // value = 0x02 (neither 0x00 nor 0x01)
0x00 // end marker
};
const json expected = { { "entry", true } };
CHECK(json::from_bson(input) == expected);
}
SECTION("non-empty object with double")
{
json const j =
@@ -562,29 +499,6 @@ TEST_CASE("BSON")
CHECK(json::from_bson(result, true, false) == j);
}
SECTION("array elements with non-conforming keys (lenient parsing)")
{
// documented lenient behavior (see gh-5333): BSON array element
// keys are not checked against the required decimal sequence
// "0", "1", "2", ... - elements are taken in encoded order
std::vector<std::uint8_t> const input =
{
0x26, 0x00, 0x00, 0x00, // size (little endian)
0x04, 'e', 'n', 't', 'r', 'y', '\x00', // entry: embedded array
0x1A, 0x00, 0x00, 0x00, // size (little endian)
0x10, '5', 0x00, 0x0A, 0x00, 0x00, 0x00, // key "5" (bogus) -> 10
0x10, 'x', 0x00, 0x14, 0x00, 0x00, 0x00, // key "x" (non-numeric) -> 20
0x10, '1', 0x00, 0x1E, 0x00, 0x00, 0x00, // key "1" (out of order) -> 30
0x00, // end marker (embedded array)
0x00 // end marker
};
const json expected = { { "entry", json::array({10, 20, 30}) } };
CHECK(json::from_bson(input) == expected);
}
SECTION("non-empty object with binary member")
{
const size_t N = 10;
@@ -680,31 +594,6 @@ TEST_CASE("BSON")
CHECK(json::from_bson(result, true, false) == j);
}
SECTION("binary member with subtype 0x02 (old binary) keeps its inner length prefix (lenient parsing)")
{
// documented lenient behavior (see gh-5333): the payload for
// binary subtype 0x02 ("old binary") is returned as-is,
// including its own inner 4-byte length prefix; it is not
// stripped or reinterpreted
std::vector<std::uint8_t> const input =
{
0x17, 0x00, 0x00, 0x00, // size (little endian)
0x05, 'e', 'n', 't', 'r', 'y', '\x00', // entry: binary
0x06, 0x00, 0x00, 0x00, // size of binary (little endian)
0x02, // "old binary" subtype
0x02, 0x00, 0x00, 0x00, // inner length prefix (part of the old-binary payload)
0x68, 0x69, // payload ('h', 'i')
0x00 // end marker
};
// the inner length prefix is part of the (unmodified) payload
const std::vector<std::uint8_t> expected_payload = {0x02, 0x00, 0x00, 0x00, 0x68, 0x69};
const json expected = { { "entry", json::binary(expected_payload, 0x02) } };
CHECK(json::from_bson(input) == expected);
}
SECTION("Some more complex document")
{
json const j =
+2 -2
View File
@@ -304,8 +304,8 @@ TEST_CASE("Unicode (3/5)" * doctest::skip())
{
for (int byte4 = 0x00; byte4 <= 0xFF; ++byte4)
{
// skip fourth second byte
if (0x80 <= byte3 && byte3 <= 0xBF)
// skip correct fourth byte
if (0x80 <= byte4 && byte4 <= 0xBF)
{
continue;
}
+1 -1
View File
@@ -305,7 +305,7 @@ TEST_CASE("Unicode (4/5)" * doctest::skip())
for (int byte4 = 0x00; byte4 <= 0xFF; ++byte4)
{
// skip correct fourth byte
if (0x80 <= byte3 && byte3 <= 0xBF)
if (0x80 <= byte4 && byte4 <= 0xBF)
{
continue;
}
+1 -1
View File
@@ -305,7 +305,7 @@ TEST_CASE("Unicode (5/5)" * doctest::skip())
for (int byte4 = 0x00; byte4 <= 0xFF; ++byte4)
{
// skip correct fourth byte
if (0x80 <= byte3 && byte3 <= 0xBF)
if (0x80 <= byte4 && byte4 <= 0xBF)
{
continue;
}