[Blog](https://scrapfly.io/blog) / [data-parsing](https://scrapfly.io/blog/tag/data-parsing) / [Quick Intro to Parsing JSON with JMESPath in Python](https://scrapfly.io/blog/posts/parse-json-jmespath-python) # Quick Intro to Parsing JSON with JMESPath in Python by [Bernardas Alisauskas](https://scrapfly.io/blog/author/bernardas) Apr 10, 2026 9 min read [\#data-parsing](https://scrapfly.io/blog/tag/data-parsing) [\#python](https://scrapfly.io/blog/tag/python) [ ](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fscrapfly.io%2Fblog%2Fposts%2Fparse-json-jmespath-python "Share on LinkedIn") [JMESPath](https://jmespath.org/) is a popular JSON query language used for parsing JSON datasets. It has gained popularity in web scraping as JSON is becoming the most popular data structure in this medium. Many popular web scraping targets contain [hidden JSON data](https://scrapfly.io/blog/posts/how-to-scrape-hidden-web-data#what-is-hidden-web-data) that can be extracted directly. Unfortunately, these datasets are huge and contain loads of useless data. This makes JSON parsing an important part of the modern web scraping process. In this Jmespath tutorial, we'll take a quick overview of this path language in web scraping and Python. We'll cover setup, the most used features and do a quick real-life example project by scraping [Realtor.com](https://www.realtor.com/). ## Key Takeaways Master JMESPath for parsing JSON datasets in web scraping using Python, handling complex data extraction from hidden JSON data with powerful query expressions. - Use jmespath python as the JSON equivalent of XPath/CSS selectors for extracting data from large JSON datasets - Parse hidden JSON data embedded in websites more efficiently than manual dictionary navigation - Apply JMESPath query expressions for complex data filtering, projections, and transformations - Handle nested JSON structures and arrays with powerful path navigation and selection capabilities - Implement JMESPath in web scraping workflows to process API responses and dynamic content data - Use JMESPath functions for data manipulation, filtering, and aggregation during JSON parsing tasks **Get web scraping tips in your inbox**Trusted by 100K+ developers and 30K+ enterprises. Unsubscribe anytime. ## What is JMESPath? JMESPath is a path language for parsing JSON datasets. In short, it allows writing path rules for selecting specific data fields in JSON. When web scraping, JMESPath is similar to XPath or CSS selectors we use to parse HTML - but for JSON. This makes JMESPath a brilliant addition to our web scraping toolset as HTML and JSON are the most common data formats in this niche. ## JMESPath Setup JMESPath is implemented in many different languages: | Language | Implementation | |---|---| | Python | [jmespath.py](https://github.com/jmespath/jmespath.py) | | PHP | [jmespath.php](https://github.com/jmespath/jmespath.php) | | Javascript | [jmespath.js](https://github.com/jmespath/jmespath.js) | | Ruby | [jmespath.rb](https://github.com/jmespath/jmespath.rb) | | Lua | [jmespath.lua](https://github.com/jmespath/jmespath.lua) | | Go | [go-jmespath](https://github.com/jmespath/go-jmespath) | | java | [jmespath-java](https://github.com/burtcorp/jmespath-java) | | Rust | [jmespath.rs](https://github.com/jmespath/jmespath.rs) | | DotNet | [jmespath.net](https://github.com/jdevillard/JmesPath.Net) | In this tutorial, we'll be using Python though other languages should be very similar. To install jmespath in Python we can use `pip install` terminal command: shell```shell $ pip install jmespath ``` ## Jmespath Usage Tutorial You're probably familiar with dictionary/hashtable dot-based path selectors like `data.address.zipcode` - this dot notation is the foundation of JMESPath but it can do much more! Just like with Python's lists we can also **slice and index** jmespath arrays: python```python import jmespath data = { "people": [ {"address": ["123 Main St", "California" ,"US"]}, ] } jmespath.search("people[0].address[:2]", data) ['123 Main St', 'California'] ``` Further, we can apply projections that apply rules for **each list element**. This is being done through the `[]` syntax: python```python data = { "people": [ {"address": ["123 Main St", "California" ,"US"]}, {"address": ["345 Alt St", "Alaska" ,"US"]}, ] } jmespath.search("people[].address[:2]", data) [ ['123 Main St', 'California'], ['345 Alt St', 'Alaska'] ] ``` Just like with lists we can also apply similar projections to objects (dictionaries). For this `*` is used: python```python data = { "people": { "foo": {"email": "foo@email.com"}, "bar": {"email": "bar@email.com"}, } } jmespath.search("people.*.email", data) [ "foo@email.com", "bar@email.com", ] ``` The most interesting feature of JMESPath for web scraping has to be **data reshaping**. Using the `.[]` and `.{}` syntax we can completely reshape lists and objects: python```python data = { "people": [ { "name": "foo", "age": 33, "addresses": [ "123 Main St", "California", "US" ], "primary_email": "foo@email.com", "secondary_email": "bar@email.com", } ] } jmespath.search(""" people[].{ first_name: name, age_in_years: age, address: addresses[0], state: addresses[1], country: addresses[2], emails: [primary_email, secondary_email] } """, data) [ { 'address': '123 Main St', 'age_in_years': 33, 'country': 'US', 'emails': ['foo@email.com', 'bar@email.com'], 'first_name': 'foo', 'state': 'California' } ] ``` As you can see, using JMESPath we can easily parse complex datasets into something more digestible which is especially useful when web scraping JSON datasets. ## Web Scraper Example Let's explore a real-life JMESPath python example by taking a look at how it would be used in web scraping. In this example project, we'll be scraping real estate property data on [realtor.com](https://www.realtor.com/) which is a popular US portal for renting and selling properties. We'll be using a few Python packages: - [httpx](https://pypi.org/project/httpx/) - HTTP client library which will let us communicate with Realtor.com's servers - [parsel](https://pypi.org/project/parsel/) - HTML parsing library which will help us to parse our web scraped HTML files. And of course `jmespath` for parsing JSON. All of these can be installed using `pip install` command: shell```shell $ pip install jmespath httpx parsel ``` Realtor.com is using [hidden web data](https://scrapfly.io/blog/posts/how-to-scrape-hidden-web-data#what-is-hidden-web-data) to render its property pages which means instead of parsing HTML we can find the whole JSON dataset hidden in the HTML code. Let's take a look at any random example property like [this one](https://www.realtor.com/realestateandhomes-detail/335-30th-Ave_San-Francisco_CA_94121_M17833-49194) If we take a look at the page source we can see the JSON data set hidden in a `