One of the fastest JSON libraries in the world. Glaze reads and writes from object memory, simplifying interfaces and offering incredible performance.
Glaze also supports:
- Read/write aggregate initializable structs without writing any metadata or macros!
- See example on Compiler Explorer
-
Pure, compile time reflection for structs
-
Standard C++ library support
-
Header only
-
Direct to memory serialization/deserialization
-
Compile time maps with constant time lookups and perfect hashing
-
Nearly zero intermediate allocations
-
Powerful wrappers to modify read/write behavior (Wrappers)
-
Use your own custom read/write functions (Custom Read/Write)
-
Handle unknown keys in a fast and flexible manner
-
Direct memory access through JSON pointer syntax
-
Binary data through the same API for maximum performance
-
No exceptions (compiles with
-fno-exceptions
)- If you desire helpers that throw for cleaner syntax see Glaze Exceptions
-
No runtime type information necessary (compiles with
-fno-rtti
) -
Rapid error handling with short circuiting
See DOCS for more documentation.
Library | Roundtrip Time (s) | Write (MB/s) | Read (MB/s) |
---|---|---|---|
Glaze | 1.20 | 1078 | 1081 |
simdjson (on demand) | N/A | N/A | 1198 |
yyjson | 1.22 | 1007 | 1109 |
daw_json_link | 2.88 | 366 | 560 |
RapidJSON | 3.70 | 289 | 441 |
Boost.JSON (direct) | 4.78 | 198 | 441 |
json_struct | 5.49 | 178 | 336 |
nlohmann | 15.56 | 84 | 82 |
Performance test code available here
Performance caveats: simdjson and yyjson are great, but they experience major performance losses when the data is not in the expected sequence or any keys are missing (the problem grows as the file size increases, as they must re-iterate through the document).
Also, simdjson and yyjson do not support automatic escaped string handling, so if any of the currently non-escaped strings in this benchmark were to contain an escape, the escapes would not be handled.
ABC Test shows how simdjson has poor performance when keys are not in the expected sequence:
Library | Read (MB/s) |
---|---|
Glaze | 988 |
simdjson (on demand) | 110 |
Tagged binary specification: BEVE
Metric | Roundtrip Time (s) | Write (MB/s) | Read (MB/s) |
---|---|---|---|
Raw performance | 0.44 | 3168 | 2350 |
Equivalent JSON data* | 0.44 | 3474 | 2577 |
JSON size: 670 bytes
BEVE size: 611 bytes
*BEVE packs more efficiently than JSON, so transporting the same data is even faster.
Your struct will automatically get reflected! No metadata is required by the user.
struct my_struct
{
int i = 287;
double d = 3.14;
std::string hello = "Hello World";
std::array<uint64_t, 3> arr = { 1, 2, 3 };
std::map<std::string, int> map{{"one", 1}, {"two", 2}};
};
JSON (prettified)
{
"i": 287,
"d": 3.14,
"hello": "Hello World",
"arr": [
1,
2,
3
],
"map": {
"one": 1,
"two": 2
}
}
Write JSON
my_struct s{};
std::string buffer = glz::write_json(s);
or
my_struct s{};
std::string buffer{};
glz::write_json(s, buffer);
Read JSON
std::string buffer = R"({"i":287,"d":3.14,"hello":"Hello World","arr":[1,2,3],"map":{"one":1,"two":2}})";
auto s = glz::read_json<my_struct>(buffer);
if (s) // check std::expected
{
s.value(); // s.value() is a my_struct populated from buffer
}
or
std::string buffer = R"({"i":287,"d":3.14,"hello":"Hello World","arr":[1,2,3],"map":{"one":1,"two":2}})";
my_struct s{};
auto ec = glz::read_json(s, buffer); // populates s from buffer
if (ec) {
// handle error
}
auto ec = glz::read_file_json(obj, "./obj.json", std::string{});
auto ec = glz::write_file_json(obj, "./obj.json", std::string{});
- Requires C++20
- Only designed and tested for 64bit little-endian systems
Actions build and test with Clang (15+), MSVC (2022), and GCC (11+) on apple, windows, and linux.
include(FetchContent)
FetchContent_Declare(
glaze
GIT_REPOSITORY https://github.com/stephenberry/glaze.git
GIT_TAG main
GIT_SHALLOW TRUE
)
FetchContent_MakeAvailable(glaze)
target_link_libraries(${PROJECT_NAME} PRIVATE glaze::glaze)
- Included in Conan Center
find_package(glaze REQUIRED)
target_link_libraries(main PRIVATE glaze::glaze)
See this Example Repository for how to use Glaze in a new project
See FAQ for Frequently Asked Questions
If you want to specialize your reflection then you can optionally write the code below:
This metadata is also necessary for non-aggregate initializable structs.
template <>
struct glz::meta<my_struct> {
using T = my_struct;
static constexpr auto value = object(
&T::i,
&T::d,
&T::hello,
&T::arr,
&T::map
);
};
Glaze also supports metadata provided within its associated class:
struct my_struct
{
int i = 287;
double d = 3.14;
std::string hello = "Hello World";
std::array<uint64_t, 3> arr = { 1, 2, 3 };
std::map<std::string, int> map{{"one", 1}, {"two", 2}};
struct glaze {
using T = my_struct;
static constexpr auto value = glz::object(
&T::i,
&T::d,
&T::hello,
&T::arr,
&T::map
);
};
};
When you define Glaze metadata, objects will automatically reflect the names of your member object pointers. However, if you want custom names or you register lambda functions or wrappers that do not provide names for your fields, you can optionally add field names in your metadata.
Example of custom names:
template <>
struct glz::meta<my_struct> {
using T = my_struct;
static constexpr auto value = object(
"integer", &T::i,
"double", &T::d,
"string", &T::hello,
"array", &T::arr,
"my map", &T::map
);
};
Each of these strings is optional and can be removed for individual fields if you want the name to be reflected.
Names are required for:
- Wrappers
- Lambda functions
Custom reading and writing can be achieved through the powerful to_json
/from_json
specialization approach, which is described here: custom-serialization.md. However, this only works for user defined types.
For common use cases or cases where a specific member variable should have special reading and writing, you can use glz::custom
to register read/write member functions, std::functions, or lambda functions.
See an example:
struct custom_encoding
{
uint64_t x{};
std::string y{};
std::array<uint32_t, 3> z{};
void read_x(const std::string& s) {
x = std::stoi(s);
}
uint64_t write_x() {
return x;
}
void read_y(const std::string& s) {
y = "hello" + s;
}
auto& write_z() {
z[0] = 5;
return z;
}
};
template <>
struct glz::meta<custom_encoding>
{
using T = custom_encoding;
static constexpr auto value = object("x", custom<&T::read_x, &T::write_x>, //
"y", custom<&T::read_y, &T::y>, //
"z", custom<&T::z, &T::write_z>);
};
suite custom_encoding_test = [] {
"custom_reading"_test = [] {
custom_encoding obj{};
std::string s = R"({"x":"3","y":"world","z":[1,2,3]})";
expect(!glz::read_json(obj, s));
expect(obj.x == 3);
expect(obj.y == "helloworld");
expect(obj.z == std::array<uint32_t, 3>{1, 2, 3});
};
"custom_writing"_test = [] {
custom_encoding obj{};
std::string s = R"({"x":"3","y":"world","z":[1,2,3]})";
expect(!glz::read_json(obj, s));
std::string out{};
glz::write_json(obj, out);
expect(out == R"({"x":3,"y":"helloworld","z":[5,2,3]})");
};
};
When using member pointers (e.g. &T::a
) the C++ class structures must match the JSON interface. It may be desirable to map C++ classes with differing layouts to the same object interface. This is accomplished through registering lambda functions instead of member pointers.
template <>
struct glz::meta<Thing> {
static constexpr auto value = object(
"i", [](auto&& self) -> auto& { return self.subclass.i; }
);
};
The value self
passed to the lambda function will be a Thing
object, and the lambda function allows us to make the subclass invisible to the object interface.
Lambda functions by default copy returns, therefore the auto&
return type is typically required in order for glaze to write to memory.
Note that remapping can also be achieved through pointers/references, as glaze treats values, pointers, and references in the same manner when writing/reading.
A class can be treated as an underlying value as follows:
struct S {
int x{};
};
template <>
struct glz::meta<S> {
static constexpr auto value{ &S::x };
};
or using a lambda:
template <>
struct glz::meta<S> {
static constexpr auto value = [](auto& self) -> auto& { return self.x; };
};
Glaze is safe to use with untrusted messages. Errors are returned as error codes, typically within a glz::expected
, which behaves just like a std::expected
.
Glaze works to short circuit error handling, which means the parsing exits very rapidly if an error is encountered.
To generate more helpful error messages, call format_error
:
auto pe = glz::read_json(obj, buffer);
if (pe) {
std::string descriptive_error = glz::format_error(pe, s);
}
This test case:
{"Hello":"World"x, "color": "red"}
Produces this error:
1:17: syntax_error
{"Hello":"World"x, "color": "red"}
^
Denoting that x is invalid here.
Array types logically convert to JSON array values. Concepts are used to allow various containers and even user containers if they match standard library interfaces.
glz::array
(compile time mixed types)std::tuple
std::array
std::vector
std::deque
std::list
std::forward_list
std::span
std::set
std::unordered_set
Object types logically convert to JSON object values, such as maps. Like JSON, Glaze treats object definitions as unordered maps. Therefore the order of an object layout does not have to match the same binary sequence in C++.
glz::object
(compile time mixed types)std::map
std::unordered_map
std::variant
See Variant Handling for more information.
std::unique_ptr
std::shared_ptr
std::optional
Nullable types may be allocated by valid input or nullified by the null
keyword.
std::unique_ptr<int> ptr{};
std::string buffer{};
glz::write_json(ptr, buffer);
expect(buffer == "null");
glz::read_json(ptr, "5");
expect(*ptr == 5);
buffer.clear();
glz::write_json(ptr, buffer);
expect(buffer == "5");
glz::read_json(ptr, "null");
expect(!bool(ptr));
By default enums will be written and read in integer form. No glz::meta
is necessary if this is the desired behavior.
However, if you prefer to use enums as strings in JSON, they can be registered in the glz::meta
as follows:
enum class Color { Red, Green, Blue };
template <>
struct glz::meta<Color> {
using enum Color;
static constexpr auto value = enumerate(Red,
Green,
Blue
);
};
In use:
Color color = Color::Red;
std::string buffer{};
glz::write_json(color, buffer);
expect(buffer == "\"Red\"");
Comments are supported with the specification defined here: JSONC
Comments may also be included in the glz::meta
description for your types. These comments can be written out to provide a description of your JSON interface. Calling write_jsonc
as opposed to write_json
will write out any comments included in the meta
description.
struct thing {
double x{5.0};
int y{7};
};
template <>
struct glz::meta<thing> {
using T = thing;
static constexpr auto value = object(
&T::x, "x is a double"_c,
&T::y, "y is an int"_c
);
};
Prettified output:
{
"x": 5 /*x is a double*/,
"y": 7 /*y is an int*/
}
The
_c
is necessary if member object pointer names are reflected. You can also writecomment("x is a double")
Formatted JSON can be written out directly via a compile time option:
glz::write<glz::opts{.prettify = true}>(obj, buffer);
Or, JSON text can be formatted with the glz::prettify_json
function:
std::string buffer = R"({"i":287,"d":3.14,"hello":"Hello World","arr":[1,2,3]})");
auto beautiful = glz::prettify_json(buffer);
beautiful
is now:
{
"i": 287,
"d": 3.14,
"hello": "Hello World",
"arr": [
1,
2,
3
]
}
To minify JSON:
glz::write<glz::opts{.prettify = true}>(obj, buffer);
std::string minified = glz::minify_json(buffer);
Glaze supports registering a set of boolean flags that behave as an array of string options:
struct flags_t {
bool x{ true };
bool y{};
bool z{ true };
};
template <>
struct glz::meta<flags_t> {
using T = flags_t;
static constexpr auto value = flags("x", &T::x, "y", &T::y, "z", &T::z);
};
Example:
flags_t s{};
expect(glz::write_json(s) == R"(["x","z"])");
Only "x"
and "z"
are written out, because they are true. Reading in the buffer will set the appropriate booleans.
When writing BEVE,
flags
only use one bit per boolean (byte aligned).
Sometimes you just want to write out JSON structures on the fly as efficiently as possible. Glaze provides tuple-like structures that allow you to stack allocate structures to write out JSON with high speed. These structures are named glz::obj
for objects and glz::arr
for arrays.
Below is an example of building an object, which also contains an array, and writing it out.
auto obj = glz::obj{"pi", 3.14, "happy", true, "name", "Stephen", "arr", glz::arr{"Hello", "World", 2}};
std::string s{};
glz::write_json(obj, s);
expect(s == R"({"pi":3.14,"happy":true,"name":"Stephen","arr":["Hello","World",2]})");
This approach is significantly faster than
glz::json_t
for generic JSON. But, may not be suitable for all contexts.
glz::merge
allows the user to merge multiple JSON object types into a single object.
glz::obj o{"pi", 3.141};
std::map<std::string_view, int> map = {{"a", 1}, {"b", 2}, {"c", 3}};
auto merged = glz::merge{o, map};
std::string s{};
glz::write_json(merged, s); // will write out a single, merged object
// s is now: {"pi":3.141,"a":0,"b":2,"c":3}
glz::merge
stores references to lvalues to avoid copies
See Generic JSON for glz::json_t
.
glz::json_t json{};
std::string buffer = R"([5,"Hello World",{"pi":3.14}])";
glz::read_json(json, buffer);
assert(json[2]["pi"].get<double>() == 3.14);
Glaze is just about as fast writing to a std::string
as it is writing to a raw char buffer. If you have sufficiently allocated space in your buffer you can write to the raw buffer, as shown below, but it is not recommended.
glz::read_json(obj, buffer);
const auto n = glz::write_json(obj, buffer.data());
buffer.resize(n);
The glz::opts
struct defines compile time optional settings for reading/writing.
Instead of calling glz::read_json(...)
, you can call glz::read<glz::opts{}>(...)
and customize the options.
For example: glz::read<glz::opts{.error_on_unknown_keys = false}>(...)
will turn off erroring on unknown keys and simple skip the items.
glz::opts
can also switch between formats:
glz::read<glz::opts{.format = glz::binary}>(...)
->glz::read_binary(...)
glz::read<glz::opts{.format = glz::json}>(...)
->glz::read_json(...)
The struct below shows the available options and the default behavior.
struct opts {
uint32_t format = json;
bool comments = false; // Write out comments
bool error_on_unknown_keys = true; // Error when an unknown key is encountered
bool skip_null_members = true; // Skip writing out params in an object if the value is null
bool use_hash_comparison = true; // Will replace some string equality checks with hash checks
bool prettify = false; // Write out prettified JSON
char indentation_char = ' '; // Prettified JSON indentation char
uint8_t indentation_width = 3; // Prettified JSON indentation size
bool shrink_to_fit = false; // Shrinks dynamic containers to new size to save memory
bool write_type_info = true; // Write type info for meta objects in variants
bool force_conformance = false; // Do not allow invalid json normally accepted such as comments, nan, inf.
bool error_on_missing_keys = false; // Require all non nullable keys to be present in the object. Use
// skip_null_members = false to require nullable members
bool error_on_const_read = false; // Error if attempt is made to read into a const value, by
// default the value is skipped without error
uint32_t layout = rowwise; // CSV row wise output/input
bool quoted_num = false; // treat numbers as quoted or array-like types as having quoted numbers
bool number = false; // read numbers as strings and write these string as numbers
bool raw = false; // write out string like values without quotes
bool raw_string = false; // do not decode/encode escaped characters for strings (improves read/write performance)
};
Many of these compile time options have wrappers to apply the option to only a single field. See Wrappers for more details.
It can be useful to acknowledge a keys existence in an object to prevent errors, and yet the value may not be needed or exist in C++. These cases are handled by registering a glz::skip
type with the meta data.
struct S {
int i{};
};
template <>
struct glz::meta<S> {
static constexpr auto value = object("key_to_skip", skip{}, &S::i);
};
std::string buffer = R"({"key_to_skip": [1,2,3], "i": 7})";
S s{};
glz::read_json(s, buffer);
// The value [1,2,3] will be skipped
expect(s.i == 7); // only the value i will be read into
Glaze is designed to help with building generic APIs. Sometimes a value needs to be exposed to the API, but it is not desirable to read in or write out the value in JSON. This is the use case for glz::hide
.
glz::hide
hides the value from JSON output while still allowing API (and JSON pointer) access.
struct hide_struct {
int i = 287;
double d = 3.14;
std::string hello = "Hello World";
};
template <>
struct glz::meta<hide_struct> {
using T = hide_struct;
static constexpr auto value = object(&T::i, //
&T::d, //
"hello", hide{&T::hello});
};
hide_struct s{};
auto b = glz::write_json(s);
expect(b == R"({"i":287,"d":3.14})"); // notice that "hello" is hidden from the output
You can parse quoted JSON numbers directly to types like double
, int
, etc. by utilizing the glz::quoted
wrapper.
struct A {
double x;
std::vector<uint32_t> y;
};
template <>
struct glz::meta<A> {
static constexpr auto value = object("x", glz::quoted_num<&A::x>, "y", glz::quoted_num<&A::y>;
};
{
"x": "3.14",
"y": ["1", "2", "3"]
}
The quoted JSON numbers will be parsed directly into the double
and std::vector<uint32_t>
. The glz::quoted
function works for nested objects and arrays as well.
Glaze supports Newline Delimited JSON for array-like types (e.g. std::vector
and std::tuple
).
std::vector<std::string> x = { "Hello", "World", "Ice", "Cream" };
std::string s = glz::write_ndjson(x);
glz::read_ndjson(x, s);
- Output performance profiles to JSON and visualize using Perfetto
See the ext
directory for extensions.
Glaze is distributed under the MIT license with an exception for embedded forms:
--- Optional exception to the license ---
As an exception, if, as a result of your compiling your source code, portions of this Software are embedded into a machine-executable object form of such source code, you may redistribute such embedded portions in such object form without including the copyright and permission notices.