# packrat overview for R working group (Stephanie Labou, 02-22-2017)

###### What is packrat? ######

# Packrat solves a number of problems, including portability and reproducibility.
# Packrat makes a "private package library" that is associated with a particular script (or project).
# This means that you can work with different version of packages for different projects.
# (This is very useful!!)

# Additionally, you can send a "snapshot" (scripts plus private library) to somebody else,
# who can then use exactly what you are using. 
# This is great for collborations! Especially in cases where everyone may have various package versions -
# packrat let's you make sure that everyone is on the same page and your work is truly reproducible.

# packrat is (potentially) VERY POWERFUL. We won't cover it all it all in this demo,
# so I highly recommend you check out the main packrat site: packrat: https://rstudio.github.io/packrat/

###### packrat demo ######

# First, you need to install packrat
install.packages("packrat")

# You'll also need to make sure your machine is able to build packages from source
# For Windows, this means making sure you have Rtools: https://cran.rstudio.com/bin/windows/Rtools/
# See here for other operating systems: https://support.rstudio.com/hc/en-us/articles/200486498-Package-Development-Prerequisites

# load packrat library
library(packrat)

# Set up a packrat project:
# This will make your current working directory a packrat project.
# I've named my folder "packrat_demo" and it will contains all the scripts, data, and other files for this pojrect
# It will also contain the private packrat library associated with this project.

packrat::init(".")

# We are now in packrat mode, and everything we load, package-wise, will be placed in our private project library
# Any packages you install from inside a packrat project are only available to that project
# and packages you install outside of the project are not available to the project.

library(ggplot2) #packrat is a blank slate, so I need to install packages I want from scratch

install.packages("ggplot2")

# Take a snapshot to save the changes in packrat

packrat::snapshot()

# Note packrat/packrat.lock
# This lists the precise package versions that were used to satisfy dependencies, including dependencies of dependencies. 
# (This file should never be edited by hand!)

# There is a lot more that you can do when building a packrat project from scratch.
# See the walkthrough guide: https://rstudio.github.io/packrat/walkthrough.html

# This is all well and good for keeping your own projects separate - no need to worry about package version conflicts!

# What I think is most useful about package is sharing projects with "bundle"
# Works like this: I bundle my project, I send it to a collaborator, then my collaborator unbundles it
# and is working in *the same environment* I am.
# No more worries about conflicting package versions from person to person or computer to computer.

packrat::bundle()

# I have written my project to a tar.gz file

# So let's say I open a fresh R studio session with a fresh script in a separate folder: