Skip to main content

A multi-platform cron-like job runner in Python

Project description

Pycroner

Pycroner

Pycroner is a lightweight cron style job runner implemented in Python. Jobs are configured via a YAML file and executed by the runner once their cron schedule matches the current time.

Why Pycroner?

This started as a tool I built for myself.

I was working on a system that had to run both on Windows and Linux, and keeping cron jobs in sync with Windows Task Scheduler was a constant headache. I wanted something dead simple, predictable, and cross-platform.

And I didn’t want to write the same job schedules twice in two different formats.

So I built Pycroner.

It runs scheduled jobs from a single YAML file, no matter what OS you’re on. You just write your jobs once and they work everywhere.

Along the way, I added a few extra things:

  • Fanout support — Run the same job multiple times with different args or in parallel.
  • Hot reload — Update your config and it just picks it up live.
  • Hooks — Jobs can run with specific scheduling hooks that are not possible with regular cron patterns, like on_start and on_exit.
  • Multi-schedule configs - You can define multiple schedules per config and they will be merged as one unified set of rules for the job, hook and cron jobs don't get merged, but still work together.

If you're building automation or ETL flows, or just want a sane way to run time-based jobs in a Python project, this might save you from the time and pain I went through managing a project on both Windows and Linux.

Features

  • Parses standard five field cron expressions (minute, hour, day, month, weekday) using a small built in parser.
  • Jobs can optionally be fanned out into multiple processes. Fanout may be an integer (repeat the job N times) or a list of argument strings that will be appended to the base command.
  • Configuration lives in pycroner.yml by default. The exact format is described in pycroner/spec.md.

Installation

From PyPI:

pip install pycroner

Usage

From code

  1. Create a pycroner.yml file describing your jobs. A simple example is shown below.
  2. Run the job runner from a Python script:
from pycroner.runner import Runner

Runner("pycroner.yml").run()

The runner checks schedules every minute and spawns each job as a subprocess when its cron expression matches the current time.

From CLI

You can also invoke the runner directly from the command line using the pycroner command. By default it looks for pycroner.yml in the current directory:

pycroner

Specify an alternative working directory with --at or a specific configuration file with --config:

pycroner --at /path/to/project
pycroner --config custom.yml

Example Configuration

jobs:
  - id: "index_articles"
    schedule: "*/15 * * * *"
    command: "python index.py"
    fanout: 4

  - id: "daily_etl"
    schedule: "0 2 * * *"
    command: "python etl.py"
    fanout:
      - "--source=internal --mode=full"
      - "--source=external --mode=delta"

  - id: "ping"
    schedule: "* * * * *"
    command: "python ping.py"

  - id: "startup"
    schedule: "on_start"
    command: "python startup.py"

  - id: "cleanup"
    schedule: "on_exit"
    command: "python cleanup.py"

  - id: "multi-conf-job"
    schedule: 
      - "on_start"
      - "*/2 * * * *"
      - "*/3 * * * *"
    command: "echo 'Hello usefulness'" 

Jobs run independently, and any output or error handling is left to your commands. For full details see pycroner/spec.md.

If the configuration file changes while the runner is active, it will be reloaded automatically so updates take effect without restarting.

Output from each job is streamed with a colored prefix containing the job id, and if fanned out, the fanout numeric id is attached.

Hooks

Hooks allow specific schedule executions that are not possible with regular cron expressions

List of available hooks:

  • on_start
  • on_exit

More hooks are being considered, and if you find a use case for a new hook you may open a PR or a discussion.

Startup and Shutdown Hooks

Jobs scheduled with on_start run once immediately when the runner boots.

Jobs scheduled with on_exit run once when the process is shutting down. The runner registers handlers for SIGINT and SIGTERM and also uses atexit to ensure shutdown hooks are executed on normal program termination.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pycroner-0.1.8.tar.gz (25.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

pycroner-0.1.8-py3-none-any.whl (23.5 kB view details)

Uploaded Python 3

File details

Details for the file pycroner-0.1.8.tar.gz.

File metadata

  • Download URL: pycroner-0.1.8.tar.gz
  • Upload date:
  • Size: 25.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.3

File hashes

Hashes for pycroner-0.1.8.tar.gz
Algorithm Hash digest
SHA256 aecfb1baaf05cc4345350db2622d269d28da75b703e7d85911b3b1600a9d01af
MD5 a755c763828d97ee59d683826744abac
BLAKE2b-256 31e917544302d0b5ef460baa028f3dc51eb5ab0641896681031e39fb1fd254e0

See more details on using hashes here.

File details

Details for the file pycroner-0.1.8-py3-none-any.whl.

File metadata

  • Download URL: pycroner-0.1.8-py3-none-any.whl
  • Upload date:
  • Size: 23.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.3

File hashes

Hashes for pycroner-0.1.8-py3-none-any.whl
Algorithm Hash digest
SHA256 a68fbca90774fa38b794e2abd497fc4ae5fd291900a61298ddeb0ed00c43c599
MD5 17812afc3a37a5500a2a1b024c61483e
BLAKE2b-256 0db28e6e7f2447944a92581532f7f62723aca0f0b93bc8c8fd37ccc8a70418f5

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page