Paramétrage avec des données CSV

When performance testing, we often need to simulate many users doing similar, but not identical, actions. For example, logging in with different credentials or searching for different items.

This is where parameterization comes in! Instead of hardcoding values, we use variables to inject dynamic data into our tests. This lesson focuses on using CSV files for this purpose in JMeter.

Parameterization is the process of replacing hardcoded values in your test script with variables. These variables then get their values from an external source, like a data file.

• It makes your tests more realistic by simulating diverse user inputs.
• It helps prevent caching issues on the server side, as each request can be unique.
• It allows you to test different scenarios without modifying the test script itself.

In JMeter, the primary element for parameterization using external files is the CSV Data Set Config. This element reads data from a plain text file, typically with comma-separated values (CSV).

It works by reading a new line of data for each iteration of your test or for each virtual user, making the requests unique.

Your CSV file is just a simple text file. Each line represents a set of data for one iteration or one user. Values within a line are separated by a delimiter, usually a comma (`,`).

• Ensure the file is accessible to JMeter (e.g., in the same directory as your .jmx script).
• You don't need a header row in the CSV file itself; you'll define variable names in JMeter.

Here's what a simple CSV file might look like, containing usernames and passwords:

user1,pass1
user2,pass2
user3,pass3

Each line provides a unique pair of username and password that can be used by a different virtual user or in different test iterations.

To add this element in JMeter:

1. Right-click on your Test Plan or Thread Group.
2. Go to Add > Config Element > CSV Data Set Config.

Placing it under the Test Plan makes it available globally. Placing it under a Thread Group limits its scope to that specific group.

The CSV Data Set Config has several important fields to configure:

• Filename: Path to your CSV file.
• Variable Names: Comma-separated list of names for variables (e.g., username,password).
• Delimiter: Character separating values (default is comma).
• Recycle on EOF?: If True, JMeter restarts from the beginning of the file when it reaches the end.
• Stop Thread on EOF?: If True, the thread stops when it runs out of data.
• Sharing Mode: How data is shared among threads (e.g., All threads, Current thread group).

Once you've defined your variable names in the CSV Data Set Config (e.g., username and password), you can use them in your JMeter Samplers.

The syntax for referencing a variable is ${variableName}.

For instance, in an HTTP Request, you might use ${username} in the username field and ${password} in the password field.

Let's say you're testing a login page that accepts user and pass as parameters. With a CSV file like user1,pass1, you'd configure your HTTP Request as follows:

Method: POST
Path: /login
Parameters:
  user: ${username}
  pass: ${password}

Each time a virtual user makes this request, JMeter will pull the next available username and password from your CSV file.

You've successfully set up a CSV Data Set Config. Which of the following are crucial settings you must define for it to work correctly?

Great job! You've learned how to parameterize your JMeter tests using the CSV Data Set Config.

• Parameterization makes tests more realistic.
• The CSV Data Set Config reads data from a CSV file.
• You define variable names in the Config element to map to your CSV columns.
• Use the ${variableName} syntax to reference data in your samplers.
• Key settings include Filename, Variable Names, Delimiter, Recycle on EOF, Stop Thread on EOF, and Sharing Mode.

Next, we'll explore handling dynamic values directly from server responses!