Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/ludat/conferer

Configuration managment for haskell
https://github.com/ludat/conferer

configuration hackage hacktoberfest haskell

Last synced: 5 days ago
JSON representation

Configuration managment for haskell

Awesome Lists containing this project

README 

        
# Conferer



[![Hackage](https://img.shields.io/hackage/v/conferer)](https://hackage.haskell.org/package/conferer)

[![GitHub Workflow Status](https://img.shields.io/github/actions/workflow/status/ludat/conferer/haskell.yml?branch=master)](https://github.com/ludat/conferer/actions)

[![Hackage-Deps](https://img.shields.io/hackage-deps/v/conferer)](https://packdeps.haskellers.com/feed?needle=conferer)

[![License](https://img.shields.io/github/license/ludat/conferer)](https://www.mozilla.org/en-US/MPL/2.0/)



Most user oriented information you can find in the website: [conferer.ludat.io](https://conferer.ludat.io/)



## Testing and CI



We have about 50 tests, mostly for the conferer package since it's the one with most of the logic

both sources and FromConfig packages also have some tests but they are not tested as throughtly.



To run them do:



```shell

stack test # for everything

stack test conferer # to run only core library tests

stack test conferer-source-json # to run json source tests

stack test conferer-warp # to run warp fromConfig tests

```



Our ci is pretty intense as well we test with the last 5 versions of ghc (oldest being 8.0.2)

and with the nightly provided by stack and on every platform (windows, linux and macOS).



So I'm pretty confident that if someone introduces a platform dependent bug we'll catch it.



## Licensing



This library is released under the Mozilla Public License 2.0 which is a weak copyleft license.



As usual you can use this library for anything you want, the only difference is that **if you modify

the source and distribute** you must publish your modifications.



You can use it for privative software, GPL'd code, in house development and distribute

it as much as you like.



## Contributing



It's pretty simple, if you want to add a feature that already has a ticket just go ahead and create a PR

referencing the ticket.



If you want to add a new feature that doesn't have a ticket then you can add a ticket first to validate

that the feature makes sense and then create a PR (you can create the PR without the ticket if you feel

like it).



## Devlopment



Most of the time I use ghcid to run tests:



```bash

ghcid -c='stack repl --test --main-is conferer-aeson:test:specs' --run=':main' --warnings

```



There you can replace `conferer-aeson` with the right package and that will run those tests, and also

reload if any of the other packages change.



## The website



The website is based on [docusaurus](https://docusaurus.io/) so it's mostly generated from the

markdown stored in the [/docs](/docs) directory and built and deployed using

[netlify](https://www.netlify.com/).



## Publishing new versions



Publishing a new version is done using stack and parallel, beware that packages in the

examples directory shouldn't be published.



But sadly versions need to be bumped by hand following PVP.



So to actually publish a new version of packages:



```shell

find -name '*.cabal' | grep -v example | parallel --tty stack upload {//}

```



## Future maybe things



* Interpolate keys with other keys: `{a: "db", b: "${a}_thing"}`, getting `b`

  will give `"db_thing"` (maybe) even in different levels of configuration

* A LOT of sources

* A LOT of `FromConfig` implementations