# Cloud Device Test (Beta)

To help developers detect bugs in a Mini Program and evaluate the quality of a Mini Program, a cloud environment for testing on physical devices is provided for free together with a set of design schemes. Before delivering the Mini Program to users, you can send it to our cloud platform for testing in a way that simulates the real use. A test report is automatically generated when the test is completed.

# Features

Bug detection: JS exceptions thrown by the Mini Program are automatically detected during the test.
Performance analysis: Performance metrics, including load time, CPU usage, and memory usage, are collected during the test.
Model coverage: The Mini Program is randomly assigned to run on 4 to 8 terminal models (depending on the device resources) for testing, allowing you to observe its performance on different models
Other: Weak network testing, multi-scenario testing, and more features are under development.

# Start Testing

Just click the Test button and check your test report later. It's as simple as that!

# Final Step

Once submitted for testing, the current version of the Mini Program will be sent to the cloud test queue for testing. Under normal circumstances, the result will be available within 24 hours. If the test report is still not accessible the next day, please provide feedback in the developer forum, and our engineers will get on it as soon as possible.

There is still much to improve in our system at present, and we are trying our best to make it better. Providing developers with more valuable tools has always been our goal.

If you have any suggestions, needs, or questions, please feel free to leave your feedback in the developer forum. (#Cloud Test)

# Q & A

# Are the cloud devices physical devices?

Yes. The devices are deployed and maintained by WeTest.

# Why was an obvious JS error in my Mini Program not detected in the cloud test?

At present, the proxy algorithm used in cloud testing is an exhaustive randomized algorithm aiming at high page coverage, but with weak defect detection capability. It uses different execution paths each time and is subject to time limits. Therefore, the path where the bug is located may not have been tested yet.

# Can text errors in my Mini Program be detected in the cloud test?

No. As the proxy algorithm is not intelligent (although we are adding more intelligent behaviors to it), errors that are obvious to human may not be detected, especially business-related errors. Semantic understanding is required to discover such text errors, which is beyond the algorithm's capability.

# Will poor test results affect the review and release of the Mini Program?

No. The test report is just a DevTools for comparison and analysis. How you use the results is up to you.

A feature will be provided in the next iteration that will allow you to configure a test account and password for the proxy algorithm to automatically log in and traverse more pages.