netR

Installation

You can install netR from GitHub:

install.packages("devtools")
devtools::install_github("BenaroyaResearch/netR")

Repository

Repository uses Github Actions and codecov. See badges above.

renv

renv is an R package for dependency management. renv helps keep same versions of packages used for all collaborators.
It is encouraged to keep netR dependencies as few as possible (0 dependency and light-weight data.table for data manipulation is recommended).

GitHub Actions

There are 2 workflows in GH Actions: R-CMD-check and test-coverage.

After each push to the remote repository, automated checks are performed. R CMD Check runs all package unit tests and checks if the package can be installed on most popular platforms - MacOS, Windows and Linux.

Code coverage reports are generated for main branch only.

Code coverage

To measure test coverage of netR codebase, covr package is used (See workflows in GitHub Actions).

Readme

This Readme is generated from README.Rmd. To update this file, modify .Rmd file in the package and rebuild using:
devtools::build_readme() or knit README.Rmd - Cmd + Shift + Enter (Windows Ctrl + Shift + Enter).

Development workflow

1. Clone the repository git clone https://github.com/BenaroyaResearch/netR.git

2. Create a development branch with name indicating what changes the branch is going to introduce - git checkout -b <branch_name>

3. Open netR.Rproj file and call renv::restore(). This updates packages versions to the ones that are specified in renv.lock file. If needed, packages will be installed. This applies only to the netR project on your machine. If you need to install and use another package, use install.packages() and then update renv.lock file using renv::snapshot() (so the next person can renv::resotre()). If you face troubles installing packages when using renv, you might need change the download method with Sys.setenv(RENV_DOWNLOAD_METHOD = "libcurl") for libcurl or wininet.

4. Add .R file with a new function(s) or add to an existing file

5. Add unit tests (recommended)

6. Run devtools::document() to update functions’ man files and NAMESPACE,
   devtools::test() to run your unit test and check if they all pass and devtools::check() to run R CMD check locally. If unit tests and cmd-check is not run locally, it will be run by GitHub anyway.

7. Increment version in DESCRIPTION file and add new line to ver_log.md file. If this is your first contribution to the package, remember to add yourself to Author field in DESCRIPTION file.

8. Push your code to remote: git push -u origin <branch_name>

9. Open a Pull Request for your branch. After passing R-CMD-check and getting approval from other collaborator, merge your development branch into main

10. Switch to main branch with git checkout main and sync your local with remote branch using git pull origin main

Merging into main branch

main is a protected branch, meaning that you cannot push directly to it. To merge code you should create a pull request from your development branch. At least one reviewer needs to approve your PR to be able to merge into main.
All required GitHub Actions workflows must pass (R-CMD-check). This prevents merging not working/incompatible code into main.

Unit tests

Package uses testthat for unit testing.
Tested functions are encouraged - when working with networks it is sometimes easy to introduce a bug. Assertions of what we think a function does and what actually happens is easier when using unit tests.

Versioning

Semantic versioning is used. In short:
MAJOR.MINOR.PATCH (e.g. 0.2.1)
MAJOR - for incompatible changes that possibly break existing code
MINOR - adding new functions to the package
PATCH - bug fixes and minor changes

Code of conduct

1. All contributors are welcome
2. We are kind