9pm runs tests in any language without the bloat. No frameworks to learn, no exotic dependencies - just reliable test execution and clear results.
- Universal: Runs any executable - shell scripts, Python, Perl, compiled binaries
- TAP Protocol: Handles results from anything that outputs Test Anything Protocol
- Crash Detection: Detects and handles tests that crash mid-execution
- Flexible Execution: Run individual tests from command line or organized test suites
- Nested Structure: Supports complex hierarchical suites with custom options and names
- Ready Environment: Provides temporary directories, logging paths, and clean isolation
Make complex test setups easy while keeping simple tests simple.
- Clone the repository:
git clone <repository-url> cd 9pm
- Install core dependencies:
# Debian/Ubuntu sudo apt install python3-yaml # RHEL/Fedora/CentOS sudo dnf install python3-pyyaml
Note
This is all you need to run tests and export JSON results. For report generation, see the Report Generation section below.
Run the framework with:
./9pm.py [OPTIONS] SUITE | TEST ...| Option | Description |
|---|---|
-a, --abort |
Stop execution after the first failure. |
-v, --verbose |
Enable verbose output. |
-p, --proj |
Specify an explicit project configuration. |
| Option | Description |
|---|---|
-c, --config |
Test Case config. |
-d, --debug |
Enable test case debug. |
-o, --option |
Test case options (repeatable). |
Example:
./9pm.py -o "ssh" suites/main.yaml cases/cleanup.shTest suites are YAML files that organize individual test cases or reference nested suites. Test cases must be executable files.
- case: "tests/smoke.sh"
- case: "tests/hammer.pl"
- case: "tests/cleanup.py"- case: "tests/smoke.sh"
- case: "tests/unit_test.tcl"
- suite: "suites/integration.yaml"
name: "Integration-Tests"
- suite: "suites/regression.yaml"
name: "Regression-Tests"Name can be used effectivly when passing different options to the same test case.
- case: "tests/scp.sh"
opts:
- "192.168.1.1"
- "bootloader.bin"
name: "upload-bootloader"
- case: "tests/scp.sh"
opts:
- "192.168.1.1"
- "linux.sqfs"
name: "upload-os"Resulting in the following:
o Execution
|-- o 0001-upload-bootloader
`-- o 0002-upload-os.sh
9pm exports test results as JSON by default. Additional report formats can be generated using the report.py tool.
Every test run generates a comprehensive result.json file containing:
- Test metadata and timestamps
- Complete hierarchical test structure
- Embedded test output logs
- Summary statistics
Install additional dependencies for report generation:
# For PDF reports
gem install --user-install asciidoctor-pdf rouge
# On Debian/Ubuntu
sudo apt install ruby-asciidoctor-pdf ruby-rougeGenerate reports from JSON:
# Generate specific format
python3 report.py result.json github # -> result-gh.md
python3 report.py result.json markdown # -> result.md
python3 report.py result.json asciidoc # -> report.adoc
# Generate all formats
python3 report.py result.json all
# Generate PDF from AsciiDoc
make report # Generates report.pdf- GitHub Markdown (
result-gh.md): GitHub-compatible with emoji status indicators - Plain Markdown (
result.md): Simple markdown format - AsciiDoc (
report.adoc): Comprehensive report with embedded test logs - PDF (
report.pdf): Professional report generated from AsciiDoc
| Status | Emoji |
|---|---|
| Passed | :white_check_mark: |
| Failed | :red_circle: |
| Skipped | :large_orange_diamond: |
| Masked Failure | :o: |
| Masked Skip | :small_orange_diamond: |
- Logs: Stored in timestamped directories under the configured
LOG_PATH. A symlinklastpoints to the latest log directory. - Scratch Directory: Temporary directories are created for each test run and cleaned up automatically.
- Temporary Database File: A temporary database file is used during the test lifecycle.
Tests must output results in TAP (Test Anything Protocol) format.
#!/bin/bash
echo "1..1"
echo "ok 1 - Sample test case"This project is licensed under GPL-2.0 license. For contributions or issues, open a pull request or issue on GitHub.