01-setup.Rmd

# (PART) Setup {-}

# Setup

## R
Download and install the [latest version of R](https://cran.r-project.org/).


<!-- When migrating to a new version of R, this line might help updating packages.  -->
<!-- update.packages(checkBuilt = TRUE, ask = FALSE) -->


## RStudio

Download and install the [latest version of RStudio](https://www.rstudio.com/products/rstudio/download/#download).


There are a couple of useful settings to change in RStudio. 


* Tools / Global Options / Workspace: Restore .Rdata at startup = OFF
* Tools / Global Options / Workspace: Save workspace to .Rdate on exit = Never


See [Efficient R / Rstudio options](https://csgillespie.github.io/efficientR/set-up.html#rstudio-options) for more information. 


## System dependencies
Windows users: install [RTools](https://cran.r-project.org/bin/windows/Rtools/). 

Mac users: install [Xcode](https://developer.apple.com/xcode/resources/) and [homebrew](https://brew.sh/).

### Spatial packages
To use spatial packages (eg.: `sf`), we need to install some system dependencies. The `sf` [README](https://github.com/r-spatial/sf#installing) has great instructions. Make sure to install these dependencies before trying to install spatial packages.

You may also find it easier to install using the R-universe version of packages, for example: 

```{r, eval = FALSE}
# Enable this universe
options(repos = c(
  	rspatial = 'https://r-spatial.r-universe.dev',
    r_spatial = 'https://rspatial.r-universe.dev',
    CRAN = 'https://cloud.r-project.org'))

# Install sf and terra packages
install.packages(c('sf', 'terra'))
```


### R Markdown {#rmd-setup}

Install packages

```{r, eval = FALSE}
install.packages(c('rmarkdown', 'bookdown', 'knitr'))
```


and [`tinytex`](https://yihui.org/tinytex/) helps us get TeX easily installed on your computer. 


```{r, eval = FALSE}
install.packages('tinytex')
tinytex::install_tinytex()
```

After installing `tinytex`, make sure to restart RStudio. 


## Git {#git-setup}
Quick setup instructions here, but see the chapter on [Git](#git) for more details. 

### Install git

Check out the great instructions in  [Happy Git With R](https://happygitwithr.com/install-git.html). Be sure to check if `git` is installed first! 

### Hosting services

Historically, the lab primarily uses [GitLab](https://gitlab.com) for private
and group projects and [GitHub](https://github.com) for mirroring published
projects or packages. This was largely dictated by GitHub's lack of free, private 
repositories, and GitLab's great continuous integration tools. Nowadays, the 
line is less clear - GitHub offers free private repositories and GitHub Actions. 
GitHub is more commonly used, especially in R package development. Take your pick
or use both!


1. Make an account on [GitHub](https://github.com/signup?user_email=&source=form-home-signup)
or [GitLab](https://gitlab.com/users/sign_in) (or both), and message Alec your usernames. 
1. Setup global git user name and email. See [here](https://happygitwithr.com/hello-git.html). 
1. Setup a local folder on your computer to store all of your git repositories. I use a folder called 'Local-git' under Documents (`/home/alecr/Documents/Local-git`). 

You can alternatively clone your repositories to different places on your hard
drive, but I find it more straightforward to keep everything together. 



### Git clients

You may opt to use a git client to work with git on a regular basis. Again, see
a great discussion in the Happy Git With R book
[here](https://happygitwithr.com/git-client.html). We've tried using cross
platform clients in the past to make it easy for folks to help each other.
Recently, it has been [GitAhead](https://gitahead.github.io/gitahead.com/) - but
development on it has recently halted... Next step TBD.