Skip to main content

A tool to convert HAR files into a locustfile.

Project description


A tool to convert web browser sessions (HAR files) into Locust load testing scenarios (locustfiles).

Use it when you want to replay HAR files (containing recordings of interactions with your website) in load tests with Locust.


Install from PyPI with pip:

pip install har-transformer



$ transformer my_scenarios_dir/


import transformer

with open("", "w") as f:
    transformer.dump(f, ["my_scenarios_dir/"])

Example HAR files are included in the examples directory for you to try out.


Create a HAR file

Using Chrome Developer Tools


  1. Prepare your scenario by thinking through the steps you want to execute
  2. Open Chrome in either Guest or Incognito mode (it's important to have no cookies prior to starting)
  3. Open the Developer Tools
  4. Open the Network panel
  5. Select Disable cache and Preserve log
  6. Clear the existing log by clicking the Clear :no_entry_sign: button
  7. Ensure recording is enabled, the Record button should be red :red_circle: (click to toggle)
  8. Navigate to your target site by entering the URL in the address bar e.g.,, etc.
  9. Perform your scenario by clicking through the pages, filling in forms, clicking buttons etc.

N.B. It's recommend that after each click/action, and before the next one, you wait until the network panel stops showing activity. This is to ensure that all requests are properly recorded.


Once you have finished performing your scenario:

  1. End recording by clicking the Record :red_circle: button
  2. Right-click on any of the file names listed in the bottom pane of the Network panel
  3. Select Save as HAR with content
  4. Save the file on your machine

ℹ️ You can view the details of the recorded HAR by simply drag-and-dropping it into the Chrome Developer Tools Network panel.

Scenario weights

By default, all scenarios have the same weight 1, which means they should all be executed as often as the others.

You can specify a different weight for each HAR file by creating an associated weight file. Weights must be positive, non-null integers due to how Locust works. Consider the following scenario group:

$ ls

$ cat scenario2.weight

Here are the weights Transformer will specify in the corresponding locustfile:

Scenario Weight
scenario1.har 1
scenario2.har 3

This means that Locust will run scenario2.har 3 times as much as scenario1.har.

Hierarchical scenarios

It may be the case that all your scenarios cannot be treated the same:

  • some only apply to a part of your load test target,
  • some should run more often than others.

For instance, the Zalando website works differently depending on its top-level domain: has an additional step before payment compared to This requires using different scenarios (i.e. different HAR files) depending on the tested domain. Moreover, to execute a realistic load test, we need to produce much more traffic targeting certain countries (i.e. groups of scenarios) than others.

To accommodate this way of working, our HAR files are organized in TLD-specific directories, each of which is potentially associated to a specific "weight" according to the relative amount of traffic expected:

├── Germany/
│   ├── scenario_1.har
│   ├── scenario_2.har
│   └── scenario_2.weight
├── Germany.weight
├── Switzerland/
│   └── scenario_1.har
└── Switzerland.weight

The "weight file" of directories (Germany.weight, Switzerland.weight) are similar to scenario weight files but apply to the whole directory (relatively to other directories of the same level). Thus, in the previous example, if Germany.weight is 6 and Switzerland.weight is 2, then Germany scenarios will be executed (in total) 3 times more (6 / 2) than Switzerland scenarios. A directory without a weight file has a weight of 1.

Scenarios can be arbitrarily nested, allowing you to organize and weight them as you want:

├── Germany/
│   ├── beauty/
│   │   └── checkout.har
│   ├── kids/
│   │   └── frontpage.har
│   ├── kids.weight
│   └── scenario_1.har
├── Germany.weight
├── Switzerland/
│   └── scenario_1.har
└── Switzerland.weight

To represent this, Transformer will produce nested Locust TaskSets in the resulting locustfile.

Blacklist specific URLs

By default, Transformer converts all requests found in input HAR files. You can blacklist, i.e. ignore, certain URLs by creating a .urlignore file in the directory in which Transformer is called.


A .urlignore containing:


... would ignore requests with URLs such as,, and https://mosaic01-abc.js (amongst others).

An example blacklist file called .urlignore_example exists that you can use as a base.

Use plugins

In case you need some customization of the Locustfile generated by Transformer, you can use its plugin mechanisms to achieve it. Please see plugins documentation for more details.


Please read for details on our process for submitting pull requests to us, and please ensure you follow the

The only required dependency for local development is Poetry.

With Poetry, our Makefile installs all necessary dependencies and run all tests and linters:



We use SemVer for versioning. For the versions available, see the tags on this repository.


See also the list of contributors who participated in this project.


This project is licensed under the MIT License - see the file for details.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for har-transformer, version 1.0.2.dev20190221094105
Filename, size File type Python version Upload date Hashes
Filename, size har_transformer-1.0.2.dev20190221094105-py3-none-any.whl (173.7 kB) File type Wheel Python version py3 Upload date Hashes View hashes
Filename, size har-transformer-1.0.2.dev20190221094105.tar.gz (48.0 kB) File type Source Python version None Upload date Hashes View hashes

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page