This page describes the steps for a project administrator to follow to set up a project environment. Collaborators can skip to next page.

Set up web services

Set up a slack workspace and share with collaborators.
Set up an overleaf project and share with collaborators.
Set up a notion workspace and share with collaborators.

Set up an RStudio project

In RStudio, create an RcppEigen project by following File > New Project > New Directory > R Package with RcppEigen.
Suppose that this project is named Economics.
Once you create an RStudio project, a folder of the project name is created.

You can start the project by double-clicking Economics.Rproj file or by opening it from the RStudio project menu on the top right corner of RStudio.
In RStudio, you will find a terminal tab next to the R console. In the following instruction, when you are asked to use a terminal, you should use the terminal in RStudio. If you start the RStudio project, the current directory is set at the project directory.
When you close the RStudio and reopen it, it automatically reload the previous project. If you want start without specifying a project, untick "Tools" > "Global Options" > "General" > "Restore most recently opened project at startup".
In the following, we assume that both R console and terminal set the project folder as the current directory when you run commands.
We slightly modify the project files.
1. The NAMESAPACE originally looks like
```
useDynLib(Economics)
import(RcppEigen)
importFrom(Rcpp, evalCpp)
export("Rcpp.fake.fun")
```
Replace the last line as:
```
useDynLib(Economics)
import(RcppEigen)
importFrom(Rcpp, evalCpp)
exportPattern("^[[:alpha:]]+")
```
exportPattern specifies the C/C++ functions to be automatically exported as R functions. By setting the pattern as ^[[:alpha:]]+, C++ functions with a comment //[[Rcpp::export]] are automatically exported as R functions by Rcpp. With export("Rcpp.fake.fun"), only a function named Rcpp.fake.fun will be exported.
1. In src/Makevars and src/Makevars.win, add the following line:
```
PKG_CPPFLAGS = -w 
```
src/Makevars controls the compile option for macOS and linux. src/Makevars.win controls them for Windows. Without -w option, compiling a project with Eigen generate many warnings. You can add as many options as you prefer.
1. Remove Read-and-delete-me.
Check whether the package can be built and installed to the library by choosing Build > Clean and Rebuild. If you did not apply the above modifications, you are likely to fail the build. If the build succeeds, in the build pane of RStudio, you will see something like:

==> Rcpp::compileAttributes()

Updated R/RcppExports.R

==> Rcmd.exe INSTALL --preclean --no-multiarch --with-keep.source Economics

installing to library 'C:/Users/kohei/Documents/R/win-library/4.0'