Liking cljdoc? Tell your friends :D

Orzo

CircleCI Clojars Project cljdoc badge MIT License Beta Project

Orzo lets you manage the versioning strategy of any project from one single place.

Table of Contents

Getting Started

Add orzo as a dependency in your deps.edn file (possibly as an alias as shown below):

{:aliases
 {:orzo {:extra-deps
         {:orzo {:mvn/version "0.1.5"}}}}}

orzo has a series of composable functions specifically designed to implement versioning strategy. You can see other example in Usage below or refer to the API docs to compose your own strategy.

For the sake of this getting started, lets assume you want a semver strategy that composes the major and minor from a text file with a patch that is the job number of your build server.

Assumption one: you have a file resources/base_version.txt that contains:

1.2

Assumption two: your build server places the build number on an environment variable called BUILD_NUM.

Assumption three: you want to save the calculated version into a file at resources/version.txt so that your users are able to identify this built.

Assumption four: if everything goes fine (assuming your build server runs tests as well) you want to tag your repo with the version number prefaced with a v (as in v0.4.542).

In order to accomplish these you'll first need to create a script called gen_version.clj on your classpath:

(ns gen-version
  (:require [orzo.core :as orzo]))

(defn -main [& args]
  (try
    (println (-> (orzo/read-file "resources/base_version.txt")
                 (orzo/extract-base-semver)
                 (orzo/set-semver {:patch (orzo/env "BUILD_NUM")})
                 (orzo/save-file "resources/version.txt")
                 (orzo/stage)))
    (System/exit 0)
    (catch Exception e
      (println e)
      (System/exit 1))))

orzo works with composable functions that can be easily piped with a thread-first macro.

The first function orzo/read-file reads the first line of resources/base_version.txt and returns its contents.

orzo/extract-base-semver will take that version just read and return a semver object that can be manipulated.

For the next line we use orzo/set-semver to set the :patch of the semver with whatever is the content of the environment variable BUILD_NUM (read here with another orzo function orzo/env).

We then save the generated version with orzo/save-file so that it can be read by users of this packaged later.

orzo also offers a convenient function for "staging" the version so that it can be used at a later point. The function here is orzo/stage and it's usually the last one in the script.

To run the script above:

$ clojure -A:orzo -m gen-version

If your base version file is 1.2 and your build number is 451 you should see:

1.2.451

Tip: if you are testing it locally and do not have an environment version BUILD_NUM set, simply run:

$ BUILD_NUM=451 clojure -A:orzo -m gen-version

Staging the version is important for the next script. We will call it tag_and_push_version.clj:

(ns tag-and-push-version
  (:require [orzo.core :as orzo]
            [orzo.git :as git]))

(defn -main [& args]
  (try
    (println (-> (orzo/unstage)
                 (orzo/prepend "v")
                 (git/tag)
                 (git/push-tag)))
    (System/exit 0)
    (catch Exception e
      (println e)
      (System/exit 1))))

This script is recovering the version that was staged by the previous script with orzo/unstage. Then it prepends it with a v, creates a tag (orzo/tag) and pushes the tag back to your origin (orzo/push-tag).

Of course you will only run this script if the build was successful. In such case, run it with:

$ clojure -A:orzo -m tag-and-push-version

Motivation

Versioning is a concern in every single project. orzo is a simple, configurable and extensible tool to implement any versioning strategy your project requires.

Usage

All functions withing orzo are composable via thread-first macros. Such a practice allows for very short and easy-to-read scripts.

For a detailed insight over each function, do refer to the API docs.

The functions are grouped by informational functions, transformation functions, validation functions and persistence functions.

Informational functions read some information from the system into the stream. Examples are orzo.core/env or orzo.core/read-file that read an environment variable or a file, respectively. Many other functions fall into that category, for instance orzo.git/last-tag that extracts the last tag from git.

Transformation functions alter the version in the pipeline. Examples are orzo.core/append orzo.core/extract-base-semver or orzo.core/bump-semver. These let you append a value to the version, transform the version into semver or bump certain semver indicators.

Validation functions will throw if something is wrong. For instance orzo.git/assert-clean? will throw if the repo is not clean locally.

Persistence functions are usually at the very end of orzo scripts. They will have side effects and therefore must be used carefully. orzo.core/save-file is a great example where a file is saved with the version passed to it. Other examples are orzo.git/tag and orzo.git/push. Both functions have implications on your git repo (locally and remotely, respectively).

Bugs

If you find a bug, submit a GitHub issue.

Help

This project is looking for team members who can help this project succeed! If you are interested in becoming a team member please open an issue.

License

Copyright © 2019 Tiago Luchini

Distributed under the MIT License.

Can you improve this documentation?Edit on GitHub

cljdoc is a website building & hosting documentation for Clojure/Script libraries

× close