Introducing Godeyes: A Static Analysis Tool for Detecting Android App Crashes

Godeyes is a dedicated static scanning tool designed to identify crash‑prone code in wireless Android and iOS applications. It runs on Windows, Linux, and macOS, with scanning rules derived from typical crash case analyses, achieving over 90% detection accuracy and helping maintain code quality during rapid app iteration.

The tool offers three integration options for Android projects: an Eclipse plugin, an Android Studio plugin, and a command‑line version for Jenkins CI pipelines. Installation involves placing the plugin in the appropriate directory or installing it via the IDE, while the Jenkins version runs on a Linux server and can be scheduled for daily scans.

Godeyes can detect a wide range of potential problems, including:

System API compatibility issues

Large image parsing leading to OOM

Uncaught number‑formatting exceptions

ArrayList usage without null checks

String.split results not validated for length

Adding Fragments without checking isAdded

Division or modulo operations without divisor validation

String.substring usage without length checks

HashMap retrievals without null checks

ArrayList get calls without index bounds checks

Dynamic UI element loading without existence verification

IO streams not closed, causing OOM

Database operation exceptions not caught

Thrown exceptions without handling

dismiss() calls without isShowing checks

Methods returning null followed by immediate calls

Database cursors not closed, leading to OOM

Dialog destruction without isShowing verification

Overridden lifecycle methods without super calls

Array index out‑of‑bounds errors

Advanced configuration allows developers to select specific rules for scanning and to define whitelist entries for false‑positive results, filtering out classes or line numbers that should be ignored.

Scan results are presented as an HTML report with clickable details, code examples, and recommended fixes, making analysis and remediation straightforward; the article notes that over 30 issues were resolved in their project using Godeyes.