wt/json
1
0 Fork 0
mirror of https://github.com/nlohmann/json.git synced 2026-10-22 11:36:14 +00:00
Code Issues Packages Projects Releases Wiki Activity

Test parsing containers of the sizes where the DOM parser changes storage

Browse Source 
The DOM parser collects the elements of an array on a stack (up to 16) and
the members of an ordered_json object (up to 32) before it moves them into
the container, and uses a hash table for keys from 64 members on. Parse
arrays and objects of the sizes around these limits, nested into each other,
with repeated keys, and as random trees that are built next to their text,
and compare json and ordered_json against the value built through the API.

The tests pass on the code from before these changes as well. The random
trees are what finds the mistakes that are easy to make here: while the
collecting change was being written, a collected array that outgrew the
stack at the moment its parent had just reached the limit left its elements
in the parent, and only these tests (142 failed assertions) showed it.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>
This commit is contained in:
Niels Lohmann committed 2026-10-22 09:28:03 +02:00
1 parent ffa9babc8d
commit e38bb5d6f8
1 file changed
+271
Download Patch File Download Diff File Expand all files Collapse all files
tests/src/unit-ordered_json.cpp
+271
View File
@@ -237,3 +237,274 @@ TEST_CASE("regression test for issue #5673 - ordered_json::emplace with a non-rv
CHECK(oj.dump() == "{\"a\":1}");
}
}
namespace
{
// the sizes around which the DOM parser changes how it stores the elements of
// an array (16) and the members of an object in a vector (32, and 64 for the
// key index)
const std::size_t container_sizes[] = {0, 1, 2, 3, 15, 16, 17, 18, 31, 32, 33, 34, 63, 64, 65, 66, 200};
template<typename Json>
std::string array_text(const std::size_t n, const std::size_t depth)
{
std::string text = "[";
for (std::size_t i = 0; i < n; ++i)
{
text += (i == 0 ? "" : ",");
// alternate scalars, empty containers and nested containers
switch (i % 4)
{
case 0:
text += std::to_string(i);
break;
case 1:
text += "\"s" + std::to_string(i) + "\"";
break;
case 2:
text += depth == 0 ? "[]" : array_text<Json>(n / 2, depth - 1);
break;
default:
text += depth == 0 ? "{}" : "{\"a\":" + array_text<Json>(n / 3, depth - 1) + ",\"b\":null}";
break;
}
}
return text + "]";
}
template<typename Json>
std::string object_text(const std::size_t n, const std::size_t depth, const bool duplicates)
{
std::string text = "{";
for (std::size_t i = 0; i < n; ++i)
{
text += (i == 0 ? "" : ",");
text += "\"key" + std::to_string(i) + "\":";
switch (i % 4)
{
case 0:
text += std::to_string(i);
break;
case 1:
text += depth == 0 ? "[]" : array_text<Json>(n / 2, depth - 1);
break;
case 2:
text += depth == 0 ? "{}" : object_text<Json>(n / 2, depth - 1, duplicates);
break;
default:
text += "\"value\"";
break;
}
}
if (duplicates)
{
// keys that are there already, the first and the last one, with other values
text += n == 0 ? "" : ",";
text += "\"key0\":[1,2],\"key" + std::to_string(n - 1) + "\":{\"again\":true}";
}
return text + "}";
}
template<typename Json>
void check_same_dump(const std::string& text)
{
// the text is what dump() writes for the parsed value, and the value
// survives a second round trip: arrays keep their elements in order, an
// ordered_json keeps its members in order of appearance
const Json parsed = Json::parse(text);
const Json again = Json::parse(parsed.dump());
CHECK(parsed == again);
CHECK(parsed.dump() == again.dump());
CHECK(Json::accept(text));
}
} // namespace
TEST_CASE("parsing containers of sizes around the limits of the DOM parser")
{
SECTION("arrays")
{
for (const std::size_t n : container_sizes)
{
const std::string text = array_text<json>(n, 2);
INFO("size ", n);
const json parsed = json::parse(text);
CHECK(parsed.is_array());
CHECK(parsed.size() == n);
for (std::size_t i = 0; i < n; i += 4)
{
CHECK(parsed[i] == i);
}
check_same_dump<json>(text);
check_same_dump<ordered_json>(text);
CHECK(ordered_json::parse(text).dump() == parsed.dump());
}
}
SECTION("objects")
{
for (const bool duplicates :
{
false, true
})
{
for (const std::size_t n : container_sizes)
{
const std::string text = object_text<json>(n, 2, duplicates);
INFO("size ", n, " duplicates ", duplicates);
const json parsed = json::parse(text);
const ordered_json ordered = ordered_json::parse(text);
CHECK(parsed.is_object());
CHECK(ordered.is_object());
CHECK(parsed.size() == ordered.size());
// the same members, in the order of their first appearance
std::size_t i = 0;
for (auto it = ordered.begin(); it != ordered.end(); ++it, ++i)
{
if (i < n)
{
CHECK(it.key() == "key" + std::to_string(i));
}
CHECK(parsed.at(it.key()) == json(it.value()));
}
if (duplicates && n > 1)
{
// the later value replaced the earlier one in its place
CHECK(ordered.at("key0") == ordered_json::parse("[1,2]"));
CHECK(ordered.at("key" + std::to_string(n - 1)) == ordered_json::parse("{\"again\":true}"));
CHECK(ordered.begin().key() == "key0");
CHECK(ordered.size() == n);
}
check_same_dump<json>(text);
check_same_dump<ordered_json>(text);
}
}
}
SECTION("objects with many different and repeated keys")
{
// 300 members of which a third repeat an earlier key
std::string text = "{";
for (std::size_t i = 0; i < 300; ++i)
{
text += (i == 0 ? "" : ",");
text += "\"k" + std::to_string(i % 200) + "\":" + std::to_string(i);
}
text += "}";
const ordered_json ordered = ordered_json::parse(text);
const json parsed = json::parse(text);
CHECK(ordered.size() == 200);
CHECK(parsed.size() == 200);
std::size_t i = 0;
for (auto it = ordered.begin(); it != ordered.end(); ++it, ++i)
{
CHECK(it.key() == "k" + std::to_string(i));
// the last value written
CHECK(it.value() == (i < 100 ? i + 200 : i));
CHECK(parsed.at(it.key()) == json(it.value()));
}
}
}
namespace
{
// a deterministic source of structure for parsing tests: builds a JSON text
// and, at the same time, the value it has to parse to
class tree_generator
{
public:
explicit tree_generator(std::uint32_t seed) : state(seed) {}
template<typename Json>
Json make(std::string& text, const std::size_t depth)
{
switch (next(depth == 0 ? 5 : 9))
{
case 0:
text += "null";
return Json(nullptr);
case 1:
text += "true";
return Json(true);
case 2:
{
const int n = static_cast<int>(next(2000)) - 1000;
text += std::to_string(n);
return Json(n);
}
case 3:
text += "\"text\"";
return Json("text");
case 4:
text += "2.5";
return Json(2.5);
case 5:
case 6:
case 7:
{
Json array = Json::array();
text += "[";
const std::size_t n = size();
for (std::size_t i = 0; i < n; ++i)
{
text += (i == 0 ? "" : ",");
array.push_back(make<Json>(text, depth - 1));
}
text += "]";
return array;
}
default:
{
Json object = Json::object();
text += "{";
const std::size_t n = size();
for (std::size_t i = 0; i < n; ++i)
{
const std::string key = "k" + std::to_string(i);
text += (i == 0 ? "\"" : ",\"") + key + "\":";
object[key] = make<Json>(text, depth - 1);
}
text += "}";
return object;
}
}
}
private:
std::uint32_t next(const std::uint32_t bound)
{
state = state * 1664525U + 1013904223U;
return (state >> 8) % bound;
}
std::size_t size()
{
return container_sizes[next(sizeof(container_sizes) / sizeof(*container_sizes))];
}
std::uint32_t state;
};
} // namespace
TEST_CASE("parsing random trees of arrays and objects")
{
for (std::uint32_t seed = 1; seed <= 150; ++seed)
{
INFO("seed ", seed);
std::string text;
std::string ordered_text;
const json expected = tree_generator(seed).make<json>(text, 3);
const ordered_json ordered_expected = tree_generator(seed).make<ordered_json>(ordered_text, 3);
CHECK(text == ordered_text);
CHECK(json::parse(text) == expected);
CHECK(json::accept(text));
const ordered_json ordered = ordered_json::parse(text);
CHECK(ordered == ordered_expected);
CHECK(ordered.dump() == ordered_expected.dump());
CHECK(json::parse(text).dump() == expected.dump());
}
}