pulsar-scheduler 0.2.0

Lightweight Python job scheduler with cron, DuckDB, web UI, and CLI

⚡ Pulsar

A lightweight Python process manager with DuckDB persistence, cron scheduling, a web UI, and a CLI.

Features

• Cron scheduling — register Python scripts with standard 5-field cron expressions
• DuckDB storage — all jobs and run history persisted in a single .duckdb file
• Web UI — real-time dashboard at http://localhost:8844 (auto-refreshes every 5s)
• REST API — full CRUD on jobs and runs at /api/* (Swagger docs at /docs)
• CLI — add, list, trigger, remove jobs and view history from the terminal
• Manual triggers — run any job on-demand from the UI or CLI
• Log capture — stdout/stderr captured per run and viewable in the UI
• Process tracking — PIDs tracked, stale runs marked as crashed on restart

Quick Start

# 1. Install
pip install pulsar-scheduler

# 2. Start the server
pulsar serve

# 3. Open http://localhost:8844 in your browser

Project Structure

Pulsar/
├── main.py           # Entry point — CLI + server bootstrap
├── db.py             # DuckDB schema & thread-safe queries
├── executor.py       # Subprocess execution in background threads
├── scheduler.py      # APScheduler cron wrapper
├── server.py         # FastAPI app + embedded HTML UI
├── requirements.txt
├── example_job.py    # Sample job (succeeds)
└── example_fail.py   # Sample job (fails — for testing)

CLI Usage

Start the server

python main.py serve [--host 0.0.0.0] [--port 8844] [--db Pulsar.duckdb]

Register a job

# Run example_job.py every 5 minutes
python main.py add my-job ./example_job.py "*/5 * * * *"

# With extra arguments
python main.py add etl-daily ./etl.py "0 2 * * *" --args "--env prod --verbose"

List jobs

python main.py list

Trigger a job manually

python main.py trigger <job_id>

Remove a job

python main.py remove <job_id>

View run history

python main.py history
python main.py history --job-id <id> --limit 10

REST API

Method	Endpoint	Description
GET	/api/status	Server status & uptime
GET	/api/jobs	List all jobs
POST	/api/jobs	Register a new job
DELETE	/api/jobs/{id}	Remove a job
POST	/api/jobs/{id}/trigger	Trigger a job manually
POST	/api/jobs/{id}/toggle	Enable/disable a job
GET	/api/runs?job_id=&limit=50	List runs
POST	/api/runs/{id}/cancel	Cancel a running job
POST	/api/reload	Reload jobs from DB

Example: add a job via API

curl -X POST http://localhost:8844/api/jobs \
  -H "Content-Type: application/json" \
  -d '{"name":"my-job","script_path":"./example_job.py","cron_expression":"*/5 * * * *"}'

Cron Syntax

Standard 5-field cron expressions:

┌───────────── minute (0–59)
│ ┌───────────── hour (0–23)
│ │ ┌───────────── day of month (1–31)
│ │ │ ┌───────────── month (1–12)
│ │ │ │ ┌───────────── day of week (0–6, Mon=1)
│ │ │ │ │
* * * * *

Expression	Meaning
*/5 * * * *	Every 5 minutes
0 9 * * *	Daily at 09:00
0 */2 * * *	Every 2 hours
0 9 * * 1	Every Monday at 09:00
30 8 1 * *	1st of every month at 08:30

Architecture

                ┌──────────┐
  CLI ─────────►│          │
                │  DuckDB  │◄──── state persistence
  Web UI ──────►│          │
                └────┬─────┘
                     │
              ┌──────┴──────┐
              │   FastAPI    │◄──── REST API + HTML
              │   Server     │
              └──────┬──────┘
                     │
              ┌──────┴──────┐
              │ APScheduler  │◄──── cron triggers
              └──────┬──────┘
                     │
              ┌──────┴──────┐
              │  Executor    │◄──── subprocess.Popen per job
              │  (threads)   │
              └─────────────┘

Notes

• The server uses sys.executable to run scripts, so jobs use the same Python interpreter.
• Stdout/stderr are captured after the process finishes (not streamed).
• Logs are truncated to 50 KB per run to keep the database lean.
• On server restart, any runs still marked as running are set to crashed.
• The DuckDB file is portable — just copy Pulsar.duckdb to move your setup.