pytest-rca-report 0.1.2

Interactive RCA report generator for pytest runs, with AI-based analysis and visual dashboard

Pytest-RCA-Report

An interactive Root Cause Analysis (RCA) dashboard for any Pytest project.
Quickly visualize, triage, and investigate test failures with actionable LLM-powered insights.

Features

• Visual summary of all test results (pass/fail/skip), anomalies, and execution times.
• Root Cause Analysis: Highlights failing modules/scripts.
• Interactive test table with search, status filtering, and drilldown for each test case.
• Classification of failures: Distinguishes between likely real bugs and probable test/infrastructure issues.
• Actionable recommendations generated by LLM (e.g. GPT-4).
• Downloadable results in JSON/CSV.
• Summary graphs (pie/bar) and execution time distribution.
• Responsive UI – optimized for both desktop and mobile.
• Works with any Pytest project: Supports unit, integration, and end-to-end tests, as long as JUnit XML is generated.

---

Demo

Installation

pip install pytest-rca-report

Usage

1. Run your tests with JUnit XML output:

   pytest --junitxml=pytest-results.xml
   

2. Generate the RCA report:

   pytest-rca-report
   

3. View the dashboard:

   python -m http.server --directory dashboard
   

   Then open http://localhost:8000/index.html in your browser.

Usage & Navigation

• Results Summary: Top summary of all tests.
• Summary Graphs: Pie & bar charts of pass/fail/skip.
• Log Anomalies: Aggregated errors and warnings across all tests.
• Root Cause Analysis: Failure counts per module/script.
• Execution Time: Histogram & slowest tests listed.
• Actionable Recommendations: Next steps as suggested by the LLM.
• All Testcases Table:
  • Search any test by name.
  • Filter by status (Passed/Failed/Skipped).
  • Click any row for a detailed modal with properties and failure reason.
  • Download all results as JSON or CSV.
• Failure Classification: Clear separation between likely real bugs and possible test automation issues.

---

Requirements

• Python 3.9+
• OpenAI API key (set as OPENAI_API_KEY in your environment)

Troubleshooting

• If the dashboard doesn't load data, always serve the dashboard folder with a local web server (not by opening the HTML file directly).

Roadmap / Suggestions

• Regression trend analysis between runs
• "Mark as Known Issue" UI
• Integration with defect tracking (Jira etc.)
• Team comments on test failures