ARDUINOJSON_USE_DOUBLE

Description

ARDUINOJSON_USE_DOUBLE determines the type used to store floating point values in JsonVariant:

  • If ARDUINOJSON_USE_DOUBLE == 0, then JsonVariant stores a float
  • If ARDUINOJSON_USE_DOUBLE == 1, then JsonVariant stores a double

The default is 0 on embedded systems, 1 otherwise.

Can’t use the Assistant

The ArduinoJson Assistant assumes that your program uses the default value for ARDUINOJSON_USE_DOUBLE.

If you change from the default, the ArduinoJson Assistant will return incorrect values.

Several .ino or .cpp files?

Be careful if several compilation units compose your program, i.e., if your project contains several .ino or .cpp files.

It’s crucial that you use the same value of ARDUINOJSON_USE_DOUBLE in each compilation unit; otherwise, the executable will contain a mix of both versions, and will inevitably crash.

Examples

Example with 32-bit precision

The following program uses 32-bit precision:

#define ARDUINOJSON_USE_DOUBLE 0
#include <ArduinoJson.h>

void example_float()
{
  StaticJsonBuffer<200> jsonBuffer;
  JsonObject& root = jsonBuffer.createObject();
  root["pi"] = 3.14159265359;

  root.printTo(Serial);
}

It produces the following output:

{"pi":3.141593}

See online demo on wandbox.org

Example with 64-bit precision

The following program uses 64-bit precision:

#define ARDUINOJSON_USE_DOUBLE 1
#include <ArduinoJson.h>

void example_double()
{
  StaticJsonBuffer<200> jsonBuffer;
  JsonObject& root = jsonBuffer.createObject();
  root["pi"] = 3.14159265359;

  root.printTo(Serial);
}

It produces the following output:

{"pi":3.141592654}

See online demo on wandbox.org

See also

Global warming stripes by Professor Ed Hawkins (University of Reading)