Skip to content

More than DAGs in R... An R package to visualise, log, and monitor DAG execution/progression using special comments and drop in functions

License

Unknown, MIT licenses found

Licenses found

Unknown
LICENSE
MIT
LICENSE.md

isaac-florence/sword

Repository files navigation

sword

Lifecycle: experimental CRAN status R-CMD-check

Inspired by Apache Airflow and prompted by Eli Lily’s {targets}, sword is a flexible package which interprets special, {roxygen2}-style comments to create interactive DAGs of pipelines or processes in R.

Designed to provide a non-disruptive, drop-in DAG building capacity that is simplistic and easy to implement. You can also use drop-in functions to monitor pipeline progression in an external, auxiliary dataset for real-time monitoring.

The reason for sword’s conception was to provide real-time logging and monitoring of frequently run pipelines or processes which use similar but new/updated data or contexts, attempting to emulate the UI used in Apache Airflow. Naturally sword does not have the same functionality for pipeline/process execution, it is purely for visualisation and monitoring/logging.

What about {targets}? {targets} is a very clever package which at the outset may appear similar to sword but in reality the two are dissimilar in functionality and purpose. sword is much more simplistic and doesn’t attempt to be as clever as {targets}, but in return shouldn’t require any changes to code or style, and isn’t as restrictive. I recommend having a look at {targets} to see if its more extensive and structured functionality is a better fit for your project.

Installation

sword is not yet on CRAN. You can install the development version using {remotes}

remotes::install_git("isaac-florence/sword")

Development

sword is in its primordial phase of development and does not have the full suite of intended first release functionality. Please check back for announcements on the state of monitoring/logging capabilities as these develop.

Planned first release functionality Status
Create DAGs using roxygen2-style script comments in testing
Drop-in functions for logging pipeline progression to be developed
Visualise pipeline progression logs in shiny app to be developed

About

More than DAGs in R... An R package to visualise, log, and monitor DAG execution/progression using special comments and drop in functions

Topics

Resources

License

Unknown, MIT licenses found

Licenses found

Unknown
LICENSE
MIT
LICENSE.md

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages