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
8 changed files with 2006 additions and 301 deletions
-18
View File
@@ -177,24 +177,6 @@ json_test_add_test_for(src/unit-comparison.cpp
MAIN test_main CXX_STANDARDS ${test_cxx_standards} ${test_force}
)
# test the parser again with JSON_DIAGNOSTIC_POSITIONS enabled
json_test_set_test_options(test-class_parser_diagnostic_positions
COMPILE_DEFINITIONS JSON_DIAGNOSTIC_POSITIONS=1
)
json_test_add_test_for(src/unit-class_parser.cpp
NAME test-class_parser_diagnostic_positions
MAIN test_main CXX_STANDARDS ${test_cxx_standards} ${test_force}
)
# test diagnostic positions again without regular diagnostics (JSON pointer paths)
json_test_set_test_options(test-diagnostic-positions_only
COMPILE_DEFINITIONS JSON_DIAGNOSTICS=0
)
json_test_add_test_for(src/unit-diagnostic-positions.cpp
NAME test-diagnostic-positions_only
MAIN test_main CXX_STANDARDS ${test_cxx_standards} ${test_force}
)
# *DO NOT* use json_test_set_test_options() below this line
#############################################################################
-266
View File
@@ -344,50 +344,6 @@ void trailing_comma_helper(const std::string& s)
}
}
#if JSON_DIAGNOSTIC_POSITIONS
/**
* Validates that the generated JSON object is the same as expected
* Validates that the start position and end position match the start and end of the string
*
* This check assumes that there is no whitespace around the json object in the original string.
*/
void validate_generated_json_and_start_end_pos_helper(const std::string& original_string, const json& j, const json& check)
{
CHECK(j == check);
CHECK(j.start_pos() == 0);
CHECK(j.end_pos() == original_string.size());
}
/**
* Parses the root object from the given root string and validates that the start and end positions for the nested object are correct.
*
* This checks that whitespace around the nested object is included in the start and end positions of the root object.
*/
void validate_start_end_pos_for_nested_obj_helper(const std::string& nested_type_json_str, const std::string& root_type_json_str, const json& expected_json, const json::parser_callback_t& cb = nullptr)
{
json j;
// 1. If callback is provided, use callback version of parse()
if (cb)
{
j = json::parse(root_type_json_str, cb);
}
else
{
j = json::parse(root_type_json_str);
}
// 2. Check if the generated JSON is as expected
// Assumptions: The root_type_json_str does not have any whitespace around the json object
validate_generated_json_and_start_end_pos_helper(root_type_json_str, j, expected_json);
// 3. Get the nested object
const auto& nested = j["nested"];
// 4. Check if the start and end positions are generated correctly for nested objects and arrays
CHECK(nested_type_json_str == root_type_json_str.substr(nested.start_pos(), nested.end_pos() - nested.start_pos()));
}
#endif
} // namespace
TEST_CASE("parser class")
@@ -1823,228 +1779,6 @@ TEST_CASE("parser class")
CHECK_THROWS_WITH_AS(_ = json::parse("/a", nullptr, true, true), "[json.exception.parse_error.101] parse error at line 1, column 2: syntax error while parsing value - invalid comment; expecting '/' or '*' after '/'; last read: '/a'", json::parse_error);
CHECK_THROWS_WITH_AS(_ = json::parse("/*", nullptr, true, true), "[json.exception.parse_error.101] parse error at line 1, column 3: syntax error while parsing value - invalid comment; missing closing '*/'; last read: '/*<U+0000>'", json::parse_error);
}
#if JSON_DIAGNOSTIC_POSITIONS
// Macro for all test cases for start_pos and end_pos
#define SETUP_TESTCASES() \
SECTION("with callback") \
{ \
SECTION("filter nothing") \
{ \
json::parser_callback_t const cb = [](int /*unused*/, json::parse_event_t /*unused*/, json& /*unused*/) noexcept \
{ \
return true; \
}; \
validate_start_end_pos_for_nested_obj_helper(nested_type_json_str, root_type_json_str, expected, cb); \
} \
SECTION("filter element") \
{ \
json::parser_callback_t const cb = [](int /*unused*/, json::parse_event_t event, json& j) noexcept \
{ \
return (event != json::parse_event_t::key && event != json::parse_event_t::value) || j != json("a"); \
}; \
validate_start_end_pos_for_nested_obj_helper(nested_type_json_str, root_type_json_str, filteredExpected, cb); \
} \
} \
SECTION("without callback") \
{ \
validate_start_end_pos_for_nested_obj_helper(nested_type_json_str, root_type_json_str, expected); \
}
SECTION("retrieve start position and end position")
{
SECTION("for object")
{
// Create an object with spaces to test the start and end positions. Spaces will not be included in the
// JSON object, however, the start and end positions should include the spaces from the input JSON string.
const std::string nested_type_json_str = R"({ "a": 1,"b" : "test1"})";
const std::string root_type_json_str = R"({ "nested": )" + nested_type_json_str + R"(, "anotherValue": "test2"})";
auto expected = json({{"nested", {{"a", 1}, {"b", "test1"}}}, {"anotherValue", "test2"}});
auto filteredExpected = expected;
filteredExpected["nested"].erase("a");
SETUP_TESTCASES()
}
SECTION("for array")
{
const std::string nested_type_json_str = R"(["a", "test", 45])";
const std::string root_type_json_str = R"({ "nested": )" + nested_type_json_str + R"(, "anotherValue": "test" })";
auto expected = json({{"nested", {"a", "test", 45}}, {"anotherValue", "test"}});
auto filteredExpected = expected;
filteredExpected["nested"] = json({"test", 45});
SETUP_TESTCASES()
}
SECTION("for array with objects")
{
const std::string nested_type_json_str = R"([{"a": 1, "b": "test"}, {"c": 2, "d": "test2"}])";
const std::string root_type_json_str = R"({ "nested": )" + nested_type_json_str + R"(, "anotherValue": "test" })";
auto expected = json({{"nested", {{{"a", 1}, {"b", "test"}}, {{"c", 2}, {"d", "test2"}}}}, {"anotherValue", "test"}});
auto filteredExpected = expected;
filteredExpected["nested"][0].erase("a");
SETUP_TESTCASES()
auto j = json::parse(root_type_json_str);
auto nested_array = j["nested"];
const auto& nested_obj = nested_array[0];
CHECK(nested_type_json_str.substr(1, 21) == root_type_json_str.substr(nested_obj.start_pos(), nested_obj.end_pos() - nested_obj.start_pos()));
CHECK(nested_type_json_str.substr(24, 22) == root_type_json_str.substr(nested_array[1].start_pos(), nested_array[1].end_pos() - nested_array[1].start_pos()));
}
SECTION("for two levels of nesting objects")
{
const std::string nested_type_json_str = R"({"nested2": {"b": "test"}})";
const std::string root_type_json_str = R"({ "a": 2, "nested": )" + nested_type_json_str + R"(, "anotherValue": "test" })";
auto expected = json({{"a", 2}, {"nested", {{"nested2", {{"b", "test"}}}}}, {"anotherValue", "test"}});
auto filteredExpected = expected;
filteredExpected.erase("a");
SETUP_TESTCASES()
auto j = json::parse(root_type_json_str);
auto nested_obj = j["nested"]["nested2"];
CHECK(nested_type_json_str.substr(12, 13) == root_type_json_str.substr(nested_obj.start_pos(), nested_obj.end_pos() - nested_obj.start_pos()));
}
SECTION("for simple types")
{
SECTION("no nested")
{
SECTION("with callback")
{
json::parser_callback_t const cb = [](int /*unused*/, json::parse_event_t /*unused*/, json& /*unused*/) noexcept
{
return true;
};
// 1. string type
std::string json_str = R"("test")";
auto j = json::parse(json_str, cb);
validate_generated_json_and_start_end_pos_helper(json_str, j, "test");
// 2. number type
json_str = R"(1)";
j = json::parse(json_str, cb);
validate_generated_json_and_start_end_pos_helper(json_str, j, 1);
// 3. boolean type
json_str = R"(true)";
j = json::parse(json_str, cb);
validate_generated_json_and_start_end_pos_helper(json_str, j, true);
// 4. null type
json_str = R"(null)";
j = json::parse(json_str, cb);
validate_generated_json_and_start_end_pos_helper(json_str, j, nullptr);
}
SECTION("without callback")
{
// 1. string type
std::string json_str = R"("test")";
auto j = json::parse(json_str);
validate_generated_json_and_start_end_pos_helper(json_str, j, "test");
// 2. number type
json_str = R"(1)";
j = json::parse(json_str);
validate_generated_json_and_start_end_pos_helper(json_str, j, 1);
json_str = R"(1.001239923)";
j = json::parse(json_str);
validate_generated_json_and_start_end_pos_helper(json_str, j, 1.001239923);
json_str = R"(1.123812389000000)";
j = json::parse(json_str);
validate_generated_json_and_start_end_pos_helper(json_str, j, 1.123812389);
// 3. boolean type
json_str = R"(true)";
j = json::parse(json_str);
validate_generated_json_and_start_end_pos_helper(json_str, j, true);
json_str = R"(false)";
j = json::parse(json_str);
validate_generated_json_and_start_end_pos_helper(json_str, j, false);
// 4. null type
json_str = R"(null)";
j = json::parse(json_str);
validate_generated_json_and_start_end_pos_helper(json_str, j, nullptr);
}
}
SECTION("string type")
{
const std::string nested_type_json_str = R"("test")";
const std::string root_type_json_str = R"({ "a": 1, "nested": )" + nested_type_json_str + R"(, "anotherValue": "test" })";
auto expected = json({{"nested", "test"}, {"anotherValue", "test"}, {"a", 1}});
auto filteredExpected = expected;
filteredExpected.erase("a");
SETUP_TESTCASES()
}
SECTION("number type")
{
const std::string nested_type_json_str = R"(2)";
const std::string root_type_json_str = R"({ "a": 1, "nested": )" + nested_type_json_str + R"(, "anotherValue": "test" })";
auto expected = json({{"nested", 2}, {"anotherValue", "test"}, {"a", 1}});
auto filteredExpected = expected;
filteredExpected.erase("a");
SETUP_TESTCASES()
}
SECTION("boolean type")
{
const std::string nested_type_json_str = R"(true)";
const std::string root_type_json_str = R"({ "a": 1, "nested": )" + nested_type_json_str + R"(, "anotherValue": "test" })";
auto expected = json({{"nested", true}, {"anotherValue", "test"}, {"a", 1}});
auto filteredExpected = expected;
filteredExpected.erase("a");
SETUP_TESTCASES()
}
SECTION("null type")
{
const std::string nested_type_json_str = R"(null)";
const std::string root_type_json_str = R"({ "a": 1, "nested": )" + nested_type_json_str + R"(, "anotherValue": "test" })";
auto expected = json({{"nested", nullptr}, {"anotherValue", "test"}, {"a", 1}});
auto filteredExpected = expected;
filteredExpected.erase("a");
SETUP_TESTCASES()
}
}
SECTION("with leading whitespace and newlines around root JSON")
{
const std::string initial_whitespace = R"(
)";
const std::string nested_type_json_str = R"({
"a": 1,
"nested": {
"b": "test"
},
"anotherValue": "test"
})";
const std::string end_whitespace = R"(
)";
const std::string root_type_json_str = initial_whitespace + nested_type_json_str + end_whitespace;
auto expected = json({{"a", 1}, {"nested", {{"b", "test"}}}, {"anotherValue", "test"}});
auto j = json::parse(root_type_json_str);
// 2. Check if the generated JSON is as expected
CHECK(j == expected);
// 3. Check if the start and end positions do not include the surrounding whitespace
CHECK(j.start_pos() == initial_whitespace.size());
CHECK(j.end_pos() == root_type_json_str.size() - end_whitespace.size());
}
}
#undef SETUP_TESTCASES
#endif
}
// this test relies on parse errors being thrown, so it is skipped when
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,44 @@
// __ _____ _____ _____
// __| | __| | | | JSON for Modern C++ (supporting code)
// | | |__ | | | | | | version 3.12.0
// |_____|_____|_____|_|___| https://github.com/nlohmann/json
//
// SPDX-FileCopyrightText: 2013-2026 Niels Lohmann <https://nlohmann.me>
// SPDX-License-Identifier: MIT
#include "doctest_compatibility.h"
#ifdef JSON_DIAGNOSTICS
#undef JSON_DIAGNOSTICS
#endif
#define JSON_DIAGNOSTICS 0
#define JSON_DIAGNOSTIC_POSITIONS 1
#include <nlohmann/json.hpp>
using json = nlohmann::json;
TEST_CASE("Better diagnostics with positions only")
{
SECTION("invalid type")
{
const std::string json_invalid_string = R"(
{
"address": {
"street": "Fake Street",
"housenumber": "1"
}
}
)";
json j = json::parse(json_invalid_string);
CHECK_THROWS_WITH_AS(j.at("address").at("housenumber").get<int>(),
"[json.exception.type_error.302] (bytes 108-111) type must be number, but is string", json::type_error);
}
SECTION("invalid type without positions")
{
const json j = "foo";
CHECK_THROWS_WITH_AS(j.get<int>(),
"[json.exception.type_error.302] type must be number, but is string", json::type_error);
}
}
+1 -13
View File
@@ -8,9 +8,7 @@
#include "doctest_compatibility.h"
#ifndef JSON_DIAGNOSTICS
#define JSON_DIAGNOSTICS 1
#endif
#define JSON_DIAGNOSTICS 1
#define JSON_DIAGNOSTIC_POSITIONS 1
#include <nlohmann/json.hpp>
@@ -29,13 +27,8 @@ TEST_CASE("Better diagnostics with positions")
}
)";
json j = json::parse(json_invalid_string);
#if JSON_DIAGNOSTICS
CHECK_THROWS_WITH_AS(j.at("address").at("housenumber").get<int>(),
"[json.exception.type_error.302] (/address/housenumber) (bytes 108-111) type must be number, but is string", json::type_error);
#else
CHECK_THROWS_WITH_AS(j.at("address").at("housenumber").get<int>(),
"[json.exception.type_error.302] (bytes 108-111) type must be number, but is string", json::type_error);
#endif
}
SECTION("invalid type without positions")
@@ -81,12 +74,7 @@ TEST_CASE("Better diagnostics with positions")
// (/foo/bar); the position of that parent is reported in the message
const json doc = json::parse(R"({"foo":{"bar":"a string"}})");
const json patch = json::parse(R"([{"op":"add","path":"/foo/bar/baz","value":1}])");
#if JSON_DIAGNOSTICS
CHECK_THROWS_WITH_AS(doc.patch(patch),
"[json.exception.out_of_range.411] (/foo/bar) (bytes 14-24) cannot add value: the JSON Patch 'add' target's parent is of type string, but must be an object or array", json::out_of_range);
#else
CHECK_THROWS_WITH_AS(doc.patch(patch),
"[json.exception.out_of_range.411] (bytes 14-24) cannot add value: the JSON Patch 'add' target's parent is of type string, but must be an object or array", json::out_of_range);
#endif
}
}
+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;
}