Python — orjson

A fast JSON library for Python

orjson is a JSON library, which can quickly and accurately complete the mutual conversion between Python objects and JSON formats. Compared with Python’s native JSON library and other third-party JSON libraries, orjson has richer functions and higher efficiency. Source code and more information about orjson : https://github.com/ijl/orjson

Introduction

First of all, let’s understand the advantages and disadvantages of orjson:

• Datetime, date and time instances can be serialized to RFC 3339 format, for example: “2022–06–12T00:00:00+00:00”
• Serializing numpy.ndarray instances is 4–12 times faster than other libraries, but uses less memory, about 1/3 of the other libraries
• The output speed is 10 to 20 times faster than the standard library
• The result of serialization is bytes, not str
• When serializing str, unicode is not escaped to ASCII
• Serialize float 10 times faster than other libraries and deserialize twice as fast as other libraries
• Subclasses that can serialize str, int, list and dict directly
• The load( ) and dump( ) methods are not provided. In the native JSON library, the load( ) method can convert…