Skip to content

Latest commit

 

History

History
70 lines (45 loc) · 4.12 KB

README.md

File metadata and controls

70 lines (45 loc) · 4.12 KB

Repeatr

repeatr run same_thing_in > same_thing_out

Repeatr is a tool for running processes repeatedly. Repeatr is designed to make task definition precise, environment setup portable, and results reproducible.

Some of Repeatr's key features and goals include:

  • Zero-ambiguity environment: Repeatr is developed on the principle of "precise-by-default". All files in your environment are managed by content-addressible storage (think: pinned as if by a git commit hash).
  • Deep-time reproducibility: Repeatr represents a commitment to reproducible results today, tomorrow, next week, next year, and... you get the picture. Repeatr configuration explicitly enforces a split between << data identity >> and << data location >>. The former never changes; the latter is explicitly variable.
  • Communicable results: Repeatr describes processes in a Formula. Communicating a Formula -- via email, gist, pastebin, whatever -- should be enough for anyone to repeat your work.
  • Control over data flow: Pull input files from multiple systems; explicitly declare sections of filesystem that are useful results to pass along. Granular control lets you build pipelines that are clean, explicit, and fast.
  • Variation builds on precision: Repeatr designs for systems like automatic updates and matrix tests on environmental variations by building them on top of Formulas. This allows clear identification of each version/test/etc, making it possible to clearly report what's been covered and what needs to be finished. Other tools can generate and consume Formulas as an API, plotting complex pipelines and checking reproducibility of results however they see fit.

Repeatr is not a build tool; think of it more as a workspace manager. It's important to have a clean workspace, fill it with good tools, and keep the materials going both in and out of your workspace well-inventoried. You can use make, cake, rake, bake, or whatever's popular this month inside Repeatr; Repeatr gives you a framework to make sure everyone plays nice.

More documentation

Repeatr is just one part of an ecosystem of software called the Timeless Stack.

The Timeless Stack documentation has its own repo: https://github.com/polydawn/timeless

Much of the documentation is published in html book form: https://repeatr.io/

Other closely related projects include Rio -- this provides the filesystem snapshot and transport for Repeatr -- and Stellar -- which provides bigger-picture pipelining tools to drive around Repeatr in interesting (and more user-friendly) ways.

⚠️ Alpha Warning ⚠️

Repeatr is in "version 0.x" ranges. That means (technically) we are not promising backwards compatablity yet; releases may make API changes.

Despite this warning, we consider the repeatr API fairly stable, and are happy to recommend building with it. We made one set of breaking changes out of the last 2.5 years; and we expect this API to last at least twice as long as the previous one.

Binary releases are available as a bundle in the github releases page for the Timeless Stack. It's also easy to build from source if you want the absolute latest bleeding-edge features.

Building from Source

Git-clone, then in the repo dir:

fling init          # fetch libraries
fling install-deps  # build rio component (used to fetch other plugins)
fling fetch-plugins # fetch plugins
fling               # build & test

Future incremental builds are just fling -- the rest of that was all first-time setup.

Binaries go into the bin/ dir; add it to your $PATH.

Libraries are handled via git submodules. You can run fling init again at any time to re-sync them. Plugins are handled via rio, another part of the Timeless Stack that Repeatr builds upon.

You can use fling -h to see other individual build and test command options. For example fling test will only run tests; fling install will not test, just build binares in bin/.