mirror of
https://github.com/nlohmann/json.git
synced 2026-10-29 12:05:22 +00:00
json_view: source positions in materialize() with JSON_DIAGNOSTIC_POSITIONS (#5945)
* json_view: source positions in materialize() with JSON_DIAGNOSTIC_POSITIONS With JSON_DIAGNOSTIC_POSITIONS, materialize() now gives every value the start_pos()/end_pos() that basic_json::parse() would give it, so exceptions about materialized values name the same bytes (#5864). - json_sax_dom_parser takes the lexer type as a defaulted template parameter; materialize() passes position_source, which reports the positions it derives from the index. diagnostic_positions treats npos as "no position" and keeps ranges all-or-nothing. - Strings with escapes keep the offset of their opening quote plus one in `next` (only with the macro); materialize() scans from there to the closing quote. The end of an array or object is found by scanning from the end of its last element over whitespace, comments and a trailing comma. - In editable documents, values written by edits, arrays and objects whose elements changed, and containers whose last element has no position get none. Images keep the flag is_new, and load() checks that `next` of a string is within the text. Nothing changes on the parse path without the macro. The view's own exceptions still carry no byte range, so the tests that compare them with those of materialized values skip it under the macro, as they skip the path under JSON_DIAGNOSTICS. Signed-off-by: Niels Lohmann <mail@nlohmann.me> * Indent comment in unit-json_view_from_json.cpp Signed-off-by: Niels Lohmann <mail@nlohmann.me> * Address review: clearer position code and docs - position_source gets clear(); set_scalar() returns early for new values, switches on node_kind, and uses no local copies; the one position_source per materialize() call is documented - no npos + 1 for containers without a position; a comment on why the end of a container is reported as its closing bracket - less duplication under #if in materialize() and the builder - diagnostic_positions sets npos explicitly - docs: positions of basic_json values after changes, positions in images depend on how the document was parsed, when next is set, and the cost of positions while parsing Signed-off-by: Niels Lohmann <mail@nlohmann.me> * Share the comment scanners of the view builder with materialize() skip_line_comment() and find_block_comment_end() in scan.hpp are used by the builder's comment() and by position_source::container_end(), which had its own copy. Signed-off-by: Niels Lohmann <mail@nlohmann.me> * Address review: if/else for container positions, clearer next field - the opening bracket of a container sets positions with an if/else instead of a ternary, and the frame takes its start from positions - the closing bracket of a container is found in nested if/else; the parent's last_end is taken from positions, so it is npos whenever no range is set - architecture.md: the next field reads "or 0 if it was not; all other values: 0" instead of "else 0; otherwise 0" Signed-off-by: Niels Lohmann <mail@nlohmann.me> * Fix GCC unused-result and clang-tidy raw-string findings in unit-json_view_positions.cpp Signed-off-by: Niels Lohmann <mail@nlohmann.me> --------- Signed-off-by: Niels Lohmann <mail@nlohmann.me>
This commit is contained in:
Niels Lohmann
@@ -0,0 +1,346 @@
|
||||
// __ _____ _____ _____
|
||||
// __| | __| | | | 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
|
||||
|
||||
// source positions of materialize() (JSON_DIAGNOSTIC_POSITIONS, #5864)
|
||||
|
||||
#include "doctest_compatibility.h"
|
||||
|
||||
#ifndef JSON_DIAGNOSTICS
|
||||
#define JSON_DIAGNOSTICS 1
|
||||
#endif
|
||||
#define JSON_DIAGNOSTIC_POSITIONS 1
|
||||
#include <nlohmann/json_view.hpp>
|
||||
using nlohmann::json;
|
||||
using nlohmann::ordered_json;
|
||||
using nlohmann::json_document;
|
||||
using nlohmann::json_editable_document;
|
||||
|
||||
#include <cstdint>
|
||||
#include <cstring>
|
||||
#include <fstream>
|
||||
#include <sstream>
|
||||
#include <string>
|
||||
#include <vector>
|
||||
|
||||
#include "json_view_test_helpers.hpp"
|
||||
#include <test_data.hpp>
|
||||
|
||||
namespace
|
||||
{
|
||||
constexpr std::size_t npos = std::string::npos;
|
||||
|
||||
// whether every value of a has the source positions of the value of b at the
|
||||
// same place (a and b are equal; shallow test documents only)
|
||||
template<typename BasicJsonType>
|
||||
bool same_positions(const BasicJsonType& a, const BasicJsonType& b)
|
||||
{
|
||||
if (a.start_pos() != b.start_pos() || a.end_pos() != b.end_pos() || a.type() != b.type() || a.size() != b.size())
|
||||
{
|
||||
return false;
|
||||
}
|
||||
if (a.is_object())
|
||||
{
|
||||
for (auto it = a.begin(); it != a.end(); ++it)
|
||||
{
|
||||
if (!same_positions(it.value(), b.at(it.key())))
|
||||
{
|
||||
return false;
|
||||
}
|
||||
}
|
||||
}
|
||||
else if (a.is_array())
|
||||
{
|
||||
for (std::size_t i = 0; i < a.size(); ++i)
|
||||
{
|
||||
if (!same_positions(a[i], b[i]))
|
||||
{
|
||||
return false;
|
||||
}
|
||||
}
|
||||
}
|
||||
return true;
|
||||
}
|
||||
|
||||
// the materialized root of text has the positions of parse()
|
||||
template<typename BasicJsonType>
|
||||
bool positions_as_parse(const std::string& text, bool comments = false, bool trailing_commas = false)
|
||||
{
|
||||
const auto j = BasicJsonType::parse(text, nullptr, true, comments, trailing_commas);
|
||||
const auto d = nlohmann::basic_json_document<BasicJsonType>::parse(text, true, comments, trailing_commas);
|
||||
const auto e = nlohmann::basic_json_document<BasicJsonType, nlohmann::view_mode::editable>::parse(text, true, comments, trailing_commas);
|
||||
return same_positions(d.root().materialize(), j) && same_positions(e.root().materialize(), j);
|
||||
}
|
||||
|
||||
std::string range(const json& j)
|
||||
{
|
||||
return j.start_pos() == npos && j.end_pos() == npos ? "none" : std::to_string(j.start_pos()) + "-" + std::to_string(j.end_pos());
|
||||
}
|
||||
} // namespace
|
||||
|
||||
TEST_CASE("json_view materialize() source positions")
|
||||
{
|
||||
SECTION("as parse()")
|
||||
{
|
||||
const std::vector<std::string> texts =
|
||||
{
|
||||
"null", "true", "false", "0", "-12", "3.25e-2", "18446744073709551615", "-9223372036854775809", R"("a")",
|
||||
R"("\n")", R"("\u00e9")", R"("\ud83d\ude00")", "\"\xf0\x9f\x98\x80\"", R"("a\"b\\")", R"("x\\")",
|
||||
" [ ] ", "{}", "[[[]]]", R"({"a":{"b":[1,{"c":[]}]}})",
|
||||
R"({"\u0041": "v\n", "k": "\"", "\t":[ "\/" ]})",
|
||||
R"([1, 2.5, true, false, null, "s", "\u0000", -0])",
|
||||
"\n\t[\r\n 1 ,\n {\n \"a\" : [ ]\n }\n]\n",
|
||||
"\xef\xbb\xbf[1, {\"a\": \"b\"}]", "\xef\xbb\xbf\"\\n\"",
|
||||
};
|
||||
for (const auto& text : texts)
|
||||
{
|
||||
CAPTURE(text)
|
||||
CHECK(positions_as_parse<json>(text));
|
||||
CHECK(positions_as_parse<ordered_json>(text));
|
||||
}
|
||||
}
|
||||
|
||||
SECTION("comments and trailing commas")
|
||||
{
|
||||
const std::vector<std::string> texts =
|
||||
{
|
||||
"[1 /* ] */]", "[1 // ]\n]", "{\"a\": 1 /* } */ , }", "[1, 2, ]", "[[1,],[2 , /**/ ] , ]",
|
||||
"{\"a\": [] // }\n, \"b\": {} /* } ] */ }", "[ /* empty */ ]", "{ // }\n}", "/* a */ [1] // b",
|
||||
R"(["\n" /* " */ , ])",
|
||||
};
|
||||
for (const auto& text : texts)
|
||||
{
|
||||
CAPTURE(text)
|
||||
CHECK(positions_as_parse<json>(text, true, true));
|
||||
}
|
||||
}
|
||||
|
||||
SECTION("generated documents")
|
||||
{
|
||||
json_view_test::generator gen;
|
||||
for (int i = 0; i < 300; ++i)
|
||||
{
|
||||
std::string text;
|
||||
gen.value(text, 0);
|
||||
CAPTURE(text)
|
||||
CHECK(positions_as_parse<json>(text));
|
||||
CHECK(positions_as_parse<ordered_json>(text));
|
||||
}
|
||||
}
|
||||
|
||||
SECTION("test data")
|
||||
{
|
||||
for (const char* name :
|
||||
{
|
||||
"/json_testsuite/sample.json", "/nativejson-benchmark/canada.json", "/nativejson-benchmark/citm_catalog.json",
|
||||
"/nativejson-benchmark/twitter.json", "/json_tests/pass1.json"
|
||||
})
|
||||
{
|
||||
CAPTURE(name)
|
||||
const std::ifstream f(std::string(TEST_DATA_DIRECTORY) + name, std::ios::binary);
|
||||
std::stringstream ss;
|
||||
ss << f.rdbuf();
|
||||
CHECK(positions_as_parse<json>(ss.str()));
|
||||
}
|
||||
}
|
||||
|
||||
SECTION("subtrees")
|
||||
{
|
||||
const std::string text = R"({"a": [1, "\n", {"b": null}], "s": "x\ty", "n": -1.5})";
|
||||
const json j = json::parse(text);
|
||||
const json_document d = json_document::parse(text);
|
||||
CHECK(same_positions(d.root()["a"].materialize(), j["a"]));
|
||||
CHECK(same_positions(d.root()["a"][2].materialize(), j["a"][2]));
|
||||
CHECK(same_positions(d.root()["a"][1].materialize(), j["a"][1]));
|
||||
CHECK(same_positions(d.root()["s"].materialize(), j["s"]));
|
||||
CHECK(same_positions(d.root()["n"].materialize(), j["n"]));
|
||||
CHECK(range(d.root()["a"][1].materialize()) == "10-14");
|
||||
CHECK(range(d.root()["s"].materialize()) == "35-41");
|
||||
}
|
||||
|
||||
#if !defined(JSON_NOEXCEPTION)
|
||||
SECTION("exception messages")
|
||||
{
|
||||
const std::string text = "{\n \"address\": {\"street\": \"Fake\\tStreet\", \"housenumber\": \"1\"},\n \"list\": [1, [true]]\n}";
|
||||
const json j = json::parse(text);
|
||||
const json_document d = json_document::parse(text);
|
||||
const json m = d.root().materialize();
|
||||
const std::string expected = json_view_test::exception_of_call([&]
|
||||
{
|
||||
static_cast<void>(j.at("address").at("housenumber").get<int>());
|
||||
});
|
||||
CHECK(expected.find("(bytes ") != std::string::npos);
|
||||
CHECK(json_view_test::exception_of_call([&]
|
||||
{
|
||||
static_cast<void>(m.at("address").at("housenumber").get<int>());
|
||||
}) == expected);
|
||||
CHECK(json_view_test::exception_of_call([&]
|
||||
{
|
||||
static_cast<void>(m.at("address").at("street").get<int>());
|
||||
}) == json_view_test::exception_of_call([&]
|
||||
{
|
||||
static_cast<void>(j.at("address").at("street").get<int>());
|
||||
}));
|
||||
CHECK(json_view_test::exception_of_call([&]
|
||||
{
|
||||
static_cast<void>(m.at("list").at(1).get<int>());
|
||||
}) == json_view_test::exception_of_call([&]
|
||||
{
|
||||
static_cast<void>(j.at("list").at(1).get<int>());
|
||||
}));
|
||||
}
|
||||
#endif
|
||||
|
||||
SECTION("edited documents")
|
||||
{
|
||||
const std::string text = R"({"a": [1, 2], "b": {"c": "\n", "d": true}, "e": [3, [4]], "f": "x"})";
|
||||
const json j = json::parse(text);
|
||||
|
||||
SECTION("new scalars have none")
|
||||
{
|
||||
json_editable_document d = json_editable_document::parse(text);
|
||||
d.set(d.root()["f"], 5);
|
||||
d.set(d.root()["b"]["d"], "s");
|
||||
const json m = d.root().materialize();
|
||||
CHECK(range(m["f"]) == "none");
|
||||
CHECK(range(m["b"]["d"]) == "none");
|
||||
// the object whose last member is new has none; the others keep theirs
|
||||
CHECK(range(m) == "none");
|
||||
CHECK(range(m["b"]) == "none");
|
||||
CHECK(same_positions(m["a"], j["a"]));
|
||||
CHECK(same_positions(m["e"], j["e"]));
|
||||
CHECK(same_positions(m["b"]["c"], j["b"]["c"]));
|
||||
}
|
||||
|
||||
SECTION("containers whose elements change have none")
|
||||
{
|
||||
json_editable_document d = json_editable_document::parse(text);
|
||||
d.push_back(d.root()["a"], 7);
|
||||
d.erase(d.root()["e"], 0);
|
||||
const json m = d.root().materialize();
|
||||
CHECK(range(m["a"]) == "none");
|
||||
CHECK(range(m["a"][2]) == "none");
|
||||
CHECK(range(m["e"]) == "none");
|
||||
// the elements kept keep theirs
|
||||
CHECK(range(m["a"][0]) == range(j["a"][0]));
|
||||
CHECK(range(m["a"][1]) == range(j["a"][1]));
|
||||
CHECK(same_positions(m["e"][0], j["e"][1]));
|
||||
CHECK(same_positions(m["b"], j["b"]));
|
||||
// the root's elements did not change
|
||||
CHECK(range(m) == range(j));
|
||||
}
|
||||
|
||||
SECTION("new containers have none")
|
||||
{
|
||||
json_editable_document d = json_editable_document::parse(text);
|
||||
d.set(d.root()["b"], json::parse(R"({"x": [1]})"));
|
||||
const json m = d.root().materialize();
|
||||
CHECK(range(m["b"]) == "none");
|
||||
CHECK(range(m["b"]["x"]) == "none");
|
||||
CHECK(range(m["b"]["x"][0]) == "none");
|
||||
CHECK(range(m) == range(j));
|
||||
}
|
||||
|
||||
SECTION("copies of scalars of the document keep theirs")
|
||||
{
|
||||
// (a copy of an array or object is a new one; its scalars are the
|
||||
// parsed ones)
|
||||
json_editable_document d = json_editable_document::parse(text);
|
||||
d.set(d.root()["f"], d.root()["e"]);
|
||||
d.set(d.root()["a"], d.root()["b"]["c"]);
|
||||
const json m = d.root().materialize();
|
||||
CHECK(range(m["a"]) == range(j["b"]["c"]));
|
||||
CHECK(range(m["f"]) == "none");
|
||||
CHECK(range(m["f"][1]) == "none");
|
||||
CHECK(range(m["f"][0]) == range(j["e"][0]));
|
||||
CHECK(range(m["f"][1][0]) == range(j["e"][1][0]));
|
||||
}
|
||||
}
|
||||
|
||||
#if !(defined(__BYTE_ORDER__) && defined(__ORDER_BIG_ENDIAN__) && __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__)
|
||||
SECTION("images")
|
||||
{
|
||||
const std::string text = R"({"a\n": ["\u00e9", 1.5, {"b": [true, null]}], "c": "d"})";
|
||||
const json j = json::parse(text);
|
||||
|
||||
SECTION("round trip")
|
||||
{
|
||||
const json_document d = json_document::load(json_document::parse(text).save());
|
||||
CHECK(same_positions(d.root().materialize(), j));
|
||||
}
|
||||
|
||||
SECTION("edited")
|
||||
{
|
||||
json_editable_document e = json_editable_document::parse(text);
|
||||
e.set(e.root()["c"], 1);
|
||||
e.push_back(e.root()[std::string("a") + '\n'], false);
|
||||
e.set(e.root()[std::string("a") + '\n'][2]["b"][0], "\"");
|
||||
const json_document d = json_document::load(e.save());
|
||||
const json m = d.root().materialize();
|
||||
CHECK(range(m["c"]) == "none");
|
||||
CHECK(range(m) == "none");
|
||||
const json& a = m[std::string("a") + '\n'];
|
||||
const json& ja = j[std::string("a") + '\n'];
|
||||
CHECK(range(a) == "none");
|
||||
CHECK(range(a[3]) == "none");
|
||||
CHECK(range(a[0]) == range(ja[0]));
|
||||
CHECK(range(a[1]) == range(ja[1]));
|
||||
CHECK(range(a[2]["b"][0]) == "none");
|
||||
CHECK(range(a[2]["b"][1]) == range(ja[2]["b"][1]));
|
||||
// the last element of b is still the parsed one
|
||||
CHECK(range(a[2]["b"]) == range(ja[2]["b"]));
|
||||
CHECK(range(a[2]) == range(ja[2]));
|
||||
}
|
||||
|
||||
SECTION("without positions")
|
||||
{
|
||||
// an image saved without JSON_DIAGNOSTIC_POSITIONS: next is 0 for
|
||||
// the strings with escapes
|
||||
std::vector<std::uint8_t> image = json_document::parse(text).save();
|
||||
std::uint64_t count = 0;
|
||||
std::memcpy(&count, image.data() + 8, sizeof(count));
|
||||
for (std::size_t i = 0; i < count; ++i)
|
||||
{
|
||||
nlohmann::detail::view::node n{};
|
||||
std::memcpy(&n, image.data() + 64 + (i * sizeof(n)), sizeof(n));
|
||||
if (n.kind == nlohmann::detail::view::node_kind::string)
|
||||
{
|
||||
n.next = 0;
|
||||
}
|
||||
std::memcpy(image.data() + 64 + (i * sizeof(n)), &n, sizeof(n));
|
||||
}
|
||||
const json_document d = json_document::load(image);
|
||||
const json m = d.root().materialize();
|
||||
CHECK(range(m[std::string("a") + '\n'][0]) == "none");
|
||||
CHECK(range(m["c"]) == range(j["c"]));
|
||||
CHECK(same_positions(m[std::string("a") + '\n'][2], j[std::string("a") + '\n'][2]));
|
||||
}
|
||||
|
||||
#if !defined(JSON_NOEXCEPTION)
|
||||
SECTION("a decoded string's position must be within the text")
|
||||
{
|
||||
std::vector<std::uint8_t> image = json_document::parse(R"(["\n"])").save();
|
||||
nlohmann::detail::view::node n{};
|
||||
std::memcpy(&n, image.data() + 64 + sizeof(n), sizeof(n));
|
||||
REQUIRE(n.next == 2);
|
||||
n.next = 7;
|
||||
std::memcpy(image.data() + 64 + sizeof(n), &n, sizeof(n));
|
||||
const auto load_image = [&image]
|
||||
{
|
||||
// GCC ignores a cast to void for [[nodiscard]]-like results, so keep the document
|
||||
const json_document d = json_document::load(image);
|
||||
static_cast<void>(d);
|
||||
};
|
||||
CHECK_THROWS_AS(load_image(), json::parse_error&);
|
||||
// without the check, the position is out of bounds and ignored
|
||||
const json_document d = json_document::load(image, json_document::image_check::none);
|
||||
CHECK(range(d.root()[0].materialize()) == "none");
|
||||
}
|
||||
#endif
|
||||
}
|
||||
#endif
|
||||
}
|
||||
Reference in New Issue
Block a user