mirror of
https://github.com/nlohmann/json.git
synced 2026-10-06 14:40:32 +00:00
Write json_view's dump() without a call or conversion per token
The default dump() (no indentation, no ensure_ascii) gets its own writer that makes the same walk and produces the same output: - the write position stays in a local variable instead of a member, so the compiler keeps it in a register across stores through aliasing char pointers; - strings and number tokens are copied with fixed-size 32-byte moves wherever enough source bytes remain, instead of one memcpy call per token; - the innermost open container lives in local variables; a stack that starts as a local array of 32 entries holds the rest; - unedited documents are walked through the node array in order, and integer tokens are read from the source directly. On top of that, float tokens of at most 15 significant digits are written straight from their digits via zmij::to_shortest() and write_shortest(), without converting to a double and back: such decimals are farther apart than a double's rounding interval, so the token's digits are the double's shortest digits. Tokens of 16+ digits, or edited values, still go through decimal_to_float(). The view's own NEON write_decimal() is removed in favor of the shared writer, and the dump output now grows in 64 KiB steps instead of being resized to its estimate at once. Signed-off-by: Niels Lohmann <mail@nlohmann.me>
This commit is contained in:
1 parent
9cb9f7fbad
commit
64c3dbd152
5 files changed
+1101
-14
No files matched your search
@@ -1147,6 +1147,57 @@ TEST_CASE("json_view dump")
|
||||
CHECK(d.root().dump() == json::parse(text).dump());
|
||||
CHECK(d.root().dump() == "[1.5,100.0,0,-0.0,1.2345678901234568e+29,18446744073709551615,-9223372036854775808,0.1,1e-07,5e-324]");
|
||||
CHECK(d.root().dump(-1, ' ', false, json_view::number_format::source) == "[1.50,1E2,-0,-0.0,123456789012345678901234567890,18446744073709551615,-9223372036854775808,0.1,1e-7,5e-324]");
|
||||
// also indented, and with ensure_ascii
|
||||
CHECK(d.root().dump(0, ' ', false, json_view::number_format::source) == "[\n1.50,\n1E2,\n-0,\n-0.0,\n123456789012345678901234567890,\n18446744073709551615,\n-9223372036854775808,\n0.1,\n1e-7,\n5e-324\n]");
|
||||
CHECK(d.root().dump(-1, ' ', true, json_view::number_format::source) == "[1.50,1E2,-0,-0.0,123456789012345678901234567890,18446744073709551615,-9223372036854775808,0.1,1e-7,5e-324]");
|
||||
|
||||
// float tokens of up to 17 significant digits in every spelling: those
|
||||
// of at most 15 digits are written from their digits, the others
|
||||
// through the conversion; both as dump() writes them
|
||||
{
|
||||
std::mt19937_64 tokens(1170); // NOLINT(cert-msc32-c,cert-msc51-cpp,bugprone-random-generator-seed)
|
||||
// a number below n; the remainder is a std::uint64_t, which is
|
||||
// std::size_t on some platforms and wider on others
|
||||
const auto draw = [&tokens](std::size_t n)
|
||||
{
|
||||
const std::uint64_t r = tokens() % n;
|
||||
return static_cast<std::size_t>(r);
|
||||
};
|
||||
std::string many_tokens = "[";
|
||||
for (int i = 0; i < 20000; ++i)
|
||||
{
|
||||
const std::size_t length = 1 + draw(17);
|
||||
std::string digits(1, static_cast<char>('1' + draw(9)));
|
||||
for (std::size_t k = 1; k < length; ++k)
|
||||
{
|
||||
digits += static_cast<char>('0' + draw(10));
|
||||
}
|
||||
digits += std::string(draw(4), '0'); // trailing zeros
|
||||
std::string token = draw(3) == 0 ? "-" : "";
|
||||
const std::size_t point = draw(digits.size() + 1);
|
||||
if (point == 0)
|
||||
{
|
||||
token += "0." + std::string(draw(5), '0') + digits;
|
||||
}
|
||||
else
|
||||
{
|
||||
token += digits.substr(0, point) + (point < digits.size() ? "." + digits.substr(point) : "");
|
||||
}
|
||||
// an exponent that keeps the value between about 1e-320 and 1e300
|
||||
const int exponent = static_cast<int>(draw(600)) - 300 - static_cast<int>(point);
|
||||
if (draw(4) != 0)
|
||||
{
|
||||
token += (draw(2) == 0 ? "e" : "E") + std::string(exponent >= 0 && draw(2) == 0 ? "+" : "") + std::to_string(exponent);
|
||||
}
|
||||
else if (point == digits.size())
|
||||
{
|
||||
token += ".0"; // (a float, not an integer)
|
||||
}
|
||||
many_tokens += (i != 0 ? "," : "") + token;
|
||||
}
|
||||
many_tokens += ']';
|
||||
CHECK(json_document::parse(many_tokens).root().dump() == json::parse(many_tokens).dump());
|
||||
}
|
||||
|
||||
// random doubles, written as parse() and dump() would
|
||||
std::mt19937_64 rng(1170); // NOLINT(cert-msc32-c,cert-msc51-cpp,bugprone-random-generator-seed)
|
||||
|
||||
@@ -26,6 +26,7 @@ using ptr_t = ordered_json::json_pointer;
|
||||
#include <functional>
|
||||
#include <iterator>
|
||||
#include <limits>
|
||||
#include <map>
|
||||
#include <random>
|
||||
#include <string>
|
||||
#include <vector>
|
||||
@@ -481,6 +482,19 @@ TEST_CASE("json_view edits: views and values")
|
||||
CHECK(d.root().materialize().dump() == json::parse(R"([1.5, 100.0, 0.1, null, null, 18446744073709551615, -9223372036854775808])").dump());
|
||||
}
|
||||
|
||||
SECTION("numbers of other float types")
|
||||
{
|
||||
// doubles have their own path to the output; other float types are
|
||||
// written as basic_json writes them, non-finite values as null
|
||||
using json_float = nlohmann::basic_json<std::map, std::vector, std::string, bool, std::int64_t, std::uint64_t, float>;
|
||||
using document_float = nlohmann::basic_json_document<json_float, true>;
|
||||
document_float d = document_float::parse("[1.5]");
|
||||
d.push_back(d.root(), std::numeric_limits<float>::quiet_NaN());
|
||||
d.push_back(d.root(), -std::numeric_limits<float>::infinity());
|
||||
CHECK(d.root().dump() == "[1.5,null,null]");
|
||||
CHECK(d.root().dump(2) == json_float::parse("[1.5, null, null]").dump(2));
|
||||
}
|
||||
|
||||
SECTION("nulls become containers, and the root can be replaced")
|
||||
{
|
||||
json_editable_document d = json_editable_document::parse("[null, null]");
|
||||
|
||||
Reference in new issue
Block a user