Low‑Cost Mobile H5 Automated Testing Framework Using Appium for Multi‑Device Compatibility and Regression

Introduction

The previous H5 automation sharing highlighted Selenium + Chrome headless for lightweight regression, but Chrome does not reflect real device environments, leading to style and resolution issues that cannot be fully exposed by that approach.

Considerations

To address these gaps, the goal is to perform H5 testing on real devices with fast regression, rapid external compatibility checks, and online monitoring for more precise and efficient automation.

Solution Overview

The solution maintains high efficiency, low operation and maintenance cost, and is built around the Appium framework to run H5 tests on various mobile devices and apps.

Functional Framework Design

Key capabilities include simultaneous testing on multiple devices and terminals, split‑screen testing for long H5 pages, scheduled online monitoring, and real‑time email reporting of anomalies.

Module Implementation Details

1. Test Task Initialization – Users create a task on the platform, specifying URL, target devices, click density, screen partitions, etc.; the platform writes these parameters to a config file that the script reads.

2. H5 Page Initialization – The URL is opened on the selected app (e.g., via QR code or in‑app feedback) to mimic real user behavior.

3. Pre‑set Position Click & Image Capture – The page is divided into regions based on click density; each region is clicked, and a screenshot is captured. This method avoids element‑specific logic but may generate invalid or duplicate clicks.

4. Platform Visual Data Management – Integrated into the UI automation platform with features for task creation, baseline image review, and result reporting, supporting batch device selection, parameter configuration, and one‑click execution.

5. Test Reporting – After each phase the platform generates detailed result reports and email notifications, including device info, URL, and any detected issues.

6. Regression Test Diff – Captured baseline images from compatibility testing are compared with new runs; significant diffs trigger email alerts and are added to the result report.

Future Plans

Machine‑learning based automatic anomaly detection will be added to recognize issues such as overlapping text, style distortion, or occlusion. Additionally, a test‑case image‑set feature will allow users to upload screenshots of click positions, enabling template‑matching to filter out ineffective clicks.