Python JSONPath Examples

Filed Under: Python
Python Jsonpath

What is JSONPath?

JSONPath is an expression language to parse JSON data. It’s very similar to the XPath expression language to parse XML data.

The idea is to parse the JSON data and get the value you want. This is more memory efficient because we don’t need to read the complete JSON data.

Python JSONPath Libraries

There are many JSONPath libraries in Python.

  1. jsonpath: It’s a port of the Perl, and JavaScript versions of JSONPath.
  2. jsonpath-rw: The complete Python implementation of the JSONPath expressions. The JSONPath expressions are first class objects, easy to analyze, transform, parse, print, and extend. The jsonpath-rw-ext module provides some additional extensions to extend its functionalities.
  3. jsonpath-ng: The final implementation of the JSONPath that aims to be standard compliant, including arithmetic and binary comparison operators. This library merges jsonpath-rw and jsonpath-rw-ext modules and further enhances it.

Which Python JSONPath Library to Use?

The jsonpath-ng module is the most comprehensive and written purely in Python. It supports both Python 2 and Python 3. So, we will use this module for Python JSONPath examples.

Installing jsonpath-ng Module

We can install jsonpath-ng module using PIP.

$ pip3.7 install jsonpath-ng
Python Jsonpath Ng Install

Python jsonpath-ng Install

Parsing a Simple JSON Data using JSONPath

Let’s look at a simple example to parse the JSON data and get the required attribute value.

import json

from jsonpath_ng import jsonpath, parse

json_string = '{"id":1, "name":"Pankaj"}'
json_data = json.loads(json_string)

jsonpath_expression = parse('$.id')

match = jsonpath_expression.find(json_data)

print("id value is", match[0].value)


[DatumInContext(value=1, path=Fields('id'), context=DatumInContext(value={'id': 1, 'name': 'Pankaj'}, path=Root(), context=None))]
id value is 1

We are using json module to convert the JSON string to a dictionary.

Parsing a List using JSONPath Expression

The JSON key can contain a list of values. We can use JSONPath expression to parse the list and get the list of values.

Let’s say we have a JSON file “db.json” with the following contents.

  "employees": [
      "id": 1,
      "name": "Pankaj",
      "salary": "10000"
      "name": "David",
      "salary": "5000",
      "id": 2

We want to parse this JSON file and get the list of employee ids. We can use JSONPath expressions to get this data very easily.

import json
from jsonpath_ng import jsonpath, parse

with open("db.json", 'r') as json_file:
    json_data = json.load(json_file)


jsonpath_expression = parse('employees[*].id')

for match in jsonpath_expression.find(json_data):
    print(f'Employee id: {match.value}')


{'employees': [{'id': 1, 'name': 'Pankaj', 'salary': '10000'}, {'name': 'David', 'salary': '5000', 'id': 2}]}
Employee id: 1
Employee id: 2

If you want to get the data into a list, you can use Python list comprehension.

emp_ids_list = [match.value for match in jsonpath_expression.find(json_data)]
print(emp_ids_list)  # [1, 2]


JSONPath provides us an easy way to parse the JSON data and extract specific values. It’s very useful when the JSON data is huge and we are interested in only few of the values.



  1. Sri says:

    Hi Pankaj,
    Thanks for your nice introductory tutorial on search/parsing a json object. How would you go about extracting a specific set of k, v pair, – given the key. Note that there will be multiple occurrences of the key, with different values. For example, I want to extract all the ‘Statement: ‘ in this roles.json (

    a =
    datums = parse(‘$..Statement’).find(a)
    len(datums) = 5. ===> this is not what I want. In fact, there are many more statements in roles.json

    Your help is very much appreciated.

    1. Alok says:

      I am not sure if i correctly understand your query , I am assuming you want to extract all the ‘Sid’ values under Statement array – If so is the case you can use below JSONPath – ” $.roles..policy.Statement[*].Sid “, Alternatively you can visit [] and write jsonpath according to your requirements

  2. Jie Liu says:

    What if I only want to read info for employee 1, not for employee 2?

Leave a Reply

Your email address will not be published. Required fields are marked *

Generic selectors
Exact matches only
Search in title
Search in content