WJ test for assignment task

[filewen]

Added a new e2e test under smoke. (wj.test.ts)
Added a new test framework and a workflow. Here is the README of this framework:

WJ Test Framework

1. Core Design

a. Data-Driven Testing

Customize test data in CSV files to quickly expand the range of test scenarios.

b. Test Case Management

• Define test files to flexibly select test cases for execution and unify the management of test cases to run (see Test Execution via Test File Management for details).

• Execute test cases through the TestRunner.

c. CI Integration

d. Test Reporting

Generate HTML reports automatically after test execution.

e. To-Be-Expanded Features

• Send test result reports via email.

• Extend test files: Filter and execute test cases by test case levels and tags.

• Organize and design UI test cases using the POM (Page Object Model) pattern.

2. Test Data Files

a. Each row in the data file drives one test case. For example, the CSV file below will make the TestRunner generate three test cases in the same group:

id,test\_case,url,method,headers,body,expected\_status

1,GET from petstore - 200,https://petstore.swagger.io/v2/pet/100,GET,,,200

2,GET from petstore - 404,https://petstore.swagger.io/v2/pet/12345678,GET,,,404

3,httpbin - GET request,https://httpbin.org/get,GET,"User-Agent: Mozilla/5.0

b. Write test case logic according to standard Playwright test file conventions, and reference fields in the data file via the record parameter.

c. The TestRunner automatically matches the data file with the same name as the test file. For example:

• Test file: http_request.test.ts

• Corresponding data file: http_request_data.csv

d. If no data file is provided, the TestRunner will execute the test case directly.

3. Test Case Execution

a. Direct Test Execution

• Run all test cases:

npm run test:build -w wj-test -- --project=WJTest

• Run a single test case:

npm run test:build -w wj-test -- --project=WJTest http\_request

You can also run tests in the workspace root directory (packages/wj-test):

• Run all test cases:

npx playwright test --project WJTest

• Run a single test case:

npx playwright test --project WJTest -- http\_request

b. Test Execution via Test Files

• Execute after compilation:

npm run start:csv -w wj-test

• Directly execute compiled files:

npm run test:csv -w wj-test

Test File Description

The framework supports controlling test cases to execute via a CSV test file:

• File name: test-config.csv

• CSV file format:

run,path,filename <--- Field names (fixed)
yes,tests,example <--- Test case 1 to execute
yes,tests, <--- Test case 2 to execute

Field Description

• run: Specifies whether to execute the test case. Value options: yes|no (default: yes).

• path: The directory where the test cases are stored. If only path is specified (without filename), it means all test cases under the directory will be executed.

• filename: The name of the test case (without the .test.ts suffix).

4. CI Integration (GitHub Action)

I created a new workflow for wj-test on GitHub:

• Workflow name: WJ Tests of Insomnia App

• Workflow file name: wj-test.yml

After the "Install packages" step in this workflow, the test cases under wj-test will be executed. Once the tests are completed, an HTML test report will be generated, which can be viewed here:

https://filewen.github.io/insomnia/

5. File Structure

Root path: packages/wj-test

packages/wj-test/
├── package.json          # Node.js configuration
├── playwright.config.ts  # Playwright project configuration
├── tsconfig.json         # TypeScript compilation configuration
├── test-config.csv       # Test case execution control file
├── common/
│   ├── app.ts            # Simplified App startup logic
│   ├── paths.ts          # Path retrieval file
│   └── test-runner.ts    # Test case execution control
├── scripts/
│   └── run-csv-tests.ts  # CSV-based test case execution control
└── tests/
    ├── example.test.ts       # A simple Playwright test case (for framework validation)
    ├── simple_requests.test.ts  # Creates a simple HTTP request (no dependency on data files)
    ├── http_request.test.ts     # Data-driven HTTP request test case (matches CSV files, sends requests, and verifies responses).
    └── http_request_data.csv    # Data file for `http_request.test.ts`.
/.github/workflows/
└── wj-test.yml       # GitHub Action workflow control file.

6. Additional Notes

a. For testing convenience, I simplified some components when starting the tests:

• Insomnia user information is not loaded: Test cases start from "Local Scratch" to simulate a new user scenario.

• Local Webserver is not started: Requests point to external public APIs.

b. In the action workflow, due to a lack of permission to access @kong/insomnia-plugin-external-vault in my forked repository, I removed npm run verify-bundle-plugins -w insomnia from the postinstall script in package.json. This should be added back for official use.

[CLAassistant]

All committers have signed the CLA.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}