JSON

There are also other programming languages ​​than Python.

Other languages ​​can not work with Python code. If you would like to "talk" with such programs - pass them some processing information or to get results from them - you have to pass the information in a simplified form.

Types

Most programming languages ​​have some numbers, some sort of lists, a variety of strings and some variation of dictionaries (or several ways to create dictionaries). And they have a way how to write True, False and None.

These basic types are usually sufficient for information handover in a legible form, although there are not the exact equivalents in all languages (Python has two basic types of numbers - int andfloat). So we will focus on them.

Data encoding

Another problem is data transfer: so for you to be able to write data on disk or transfer via the Internet, it has to be converted to a sequence of bytes (numbers from 0 to 255). Simplified: you have to convert it to a string.

There are plenty of ways to encode data into text. Each way is trying to find the right balance between legibility for people/computers, length of record, security, options and extensibility. We already know the syntax for Python:

{
    'name': 'Anna',
    'city': 'Prague',
    'languages': ['Czech', 'English', 'Python'],
    'age': 26,
}

Another way to write data is YAML:

name: Anna
city: Prague
languages:
   - Czech
   - English
   - Python
age: 26

Finally, there is also JSON (Javascript Object Notation), which, for its simplicity, has expanded the most:

{
  "Name": "Anna",
  "City": "Prague",
  "Languages": ["Czech", "English", "Python"],
  "Age": 26
}

JSON in Python

Object encoding in JSON is simple: there is a json module, whose loads method retrieves data from the string:

import json

json_string = """
    {
      "name": "Anna",
      "city": "Brno",
      "languages": ["Czech", "English", "Python"],
      "age": 26
    }
"""

data = json.loads(json_string)
print(data)
print(data['city'])

And then there is the dumps method, which decodes the given data and returns a string.

The string that dumps(data) returns is suitable for computer treatment. If you want to read it, it is better to set ensure_ascii = False (so that accented letters are not encoded with\) and indent = 2 (indent with two spaces).

>>> print(json.dumps(data, ensure_ascii = False, indent = 2))
{
  "name": "Anna",
  "city": "Brno",
  "languages": [
    "Czech",
    "English",
    "Python"
  ],
  "age": 26
}

Exercise

Write code to print the value of salary of Emma from following JSON string.

Try not to depend on the fact that her entry is first in the list)

sampleJson = """{
  "company":{
    "employees":[
      {
        "name":"emma",
        "payable":{
          "salary":7000,
          "bonus":800
        }
      },
      {
        "name":"anna",
        "payable":{
          "salary":5500,
          "bonus":1000
        }
      }
    ]
  }
}"""

import json

sampleJson = """{
  "company":{
    "employees":[
      {
        "name":"emma",
        "payable":{
          "salary":7000,
          "bonus":800
        }
      },
      {
        "name":"anna",
        "payable":{
          "salary":5500,
          "bonus":1000
        }
      }
    ]
  }
}"""
content = json.loads(sampleJson)
employees = content.get("company", {}).get("employees", [])
for employee in employees:
    name = employee.get('name')
    if name == "emma":
      payable = employee.get("payable", {})
      total = payable.get("salary", 0) + payable.get("bonus", 0)
      print(f"{name} has a total salary with bonus: {total}")

A complete description of json module - including write/read functions directly to/from files - is in the documentation.