qa-run-bundler

Create a self-contained QA run bundle (ZIP + index.html viewer) from automation artifacts.

• PyPI: qa-run-bundler
• CLI: qa-bundle
• Includes a pytest plugin (AUTO-ON by default)

What it does today (v0.2.3)

This release adds Playwright auto-capture (pytest-playwright) + root-level report output + a paths json for Slack/Email/Jira/Allure.

It:

• determines the report base directory:
  • if inside a git repo → uses git root (git rev-parse --show-toplevel)
  • else → uses the current working directory (where pytest was run)
• creates a timestamped run folder under:
  • <base>/test_report/<project>/<env>/.../runs/<run_id>/
• writes a standalone viewer:
  • <run_root>/qa_bundle/index.html (works on file:// by default)
• generates:
  • <run_root>/qa_bundle/bundle.zip (size limit 1GB)
• writes a machine-readable paths file (for Slack/Email/Jira/Allure):
  • <run_root>/qa_bundle/qa_bundle_paths.json
• records a basic test summary (nodeid/status/duration + phases)

Auto-captured (Playwright)

Enabled by default (can be disabled via env/CLI flags):

• ✅ Video recording (Playwright page video)
• ✅ HAR network recording (network/network.har, with content=embed)
• ✅ Console logs capture (tests/<test>/logs/console.log)
• ✅ Trace capture (tests/<test>/traces/trace.zip)
• ✅ Screenshot on failure (tests/<test>/screenshot_failed.png)

Disable via .env

QA_BUNDLE=0
# or individual toggles
QA_VIDEO=0
QA_HAR=0
QA_TRACE=0
QA_CONSOLE=0
QA_SCREENSHOT_ON_FAIL=0

Requirements / Scope

• Python + Playwright only (pytest-based).
• This package is intended for Playwright UI automation runs and will not capture Selenium/Appium by default.

Install

pip install qa-run-bundler

CLI usage (non-pytest users)

qa-bundle build --root test_report/<project>/<env>/runs/<run_id>

Output:

• <root>/qa_bundle/index.html
• <root>/qa_bundle/bundle.zip

Pytest plugin (AUTO-ON)

This plugin runs even when you click PyCharm Play.

Disable via .env / env var

QA_BUNDLE=0

or

pytest --no-qa-bundle

Recommend env vars (to control output path)

PROJECT_NAME=consumer_website
ENV=dev
TEST_ENV=qa

Output paths

• Prod: test_report/<project>/prod/runs/<run_id>/... (no TEST_ENV segment)
• Non-prod:
  • with TEST_ENV: test_report/<project>/<env>/<test_env>/runs/<run_id>/...
  • without TEST_ENV: test_report/<project>/<env>/runs/<run_id>/...

Planned features (recommended next)

1. Viewer upgrades:
   • filters (failed-only / search / slow tests)
   • per-test attachments UI (open video/trace/logs directly)
   • network table view with cURL copy
2. Network improvements:
   • per-test HAR split option
   • size caps + redaction rules (tokens/cookies)
3. Console improvements:
   • include request correlation ids in logs when possible
4. Allure helper (optional):
   • attach bundle.zip automatically if allure is installed