We built a tool to analyze Java files. We are doing a static analysis using JavaParser, and we are visualizing our results using a VS-Code extension. The visualization component includes:
- Syntax highlighting: We underline areas of code which we have identified as problematic. The colors for the highlighted sections correspond to the associated "risk level"
- Error messaging and suggestions: When hovering the cursor over underlined sections of code, we will display a message describing what the error is, as well as a suggestion on how to address the issue
- Custom configuration of risk levels and colors: The user is able to provide custom configurations on the risk levels of each type of identified error, as well as provide custom coloring for the syntax highlighting
Our static analyzer identifies the following exception-related issues:
- Empty catch exception blocks
- Expensive Exception Control Flow: identify try-catch blocks inside loops
- Unclosed Resources
- Unhandled Exceptions
- Undeclared Exceptions
- System Error in Catch: identify catch blocks that use System.err or printStackTrace
- Always Triggered Catch
- Redundant Try-Catch: find cases where conditional checks can replace exception handling
- Exception Propagation: count number of functions affected by exceptions
We implemented our java analysis using JavaParser in Java and parsed analysis results to a JSON object using Gson. The source code for the analysis can be found in java-analyzer. We build this into a JAR file, which can then be used in our VS-code extension.
The visualization component is implemented as a VSCode extension in TypeScript. It gets analysis results from JSON that java-analyzer provides. Instructions on how to install and use the extension are included in vscode-extension/README.md