Managing Environments

Where we are

How do we keep this all running smoothly?


Some surprises are good


But this is production, not showbiz.

Improving & Protecting your Workflow

  • Manage your environment
  • Use a dev-test-prod workflow

Environments

Think of your data science workbench as a kitchen:

  • A repository is the grocery store, a central place where everyone gets their packages.
  • A library is the pantry, where you keep your own private set of packages.
  • A Package is the food!

Environments

👿 Sometimes things go wrong

  • Revisited an old R script, and it didn’t work.
    • “I swear that {insert function from package} worked before…”
  • Shared R code with a colleague, but they can’t get it to run.
    • “Why does this run on my computer, but not on Sue’s?”
    • “John pulled in my project from GitHub but can’t get it to work!”
  • You modified something in one project, and borked something in another project.
    • “I upgraded a package in one project, and now my other projects don’t work!”

Enter renv

The renv package helps you create reproducible environments for your R projects. Use renv to make your R projects more:

  • Isolated: Installing a new or updated package for one project won’t break your other projects, and vice versa. That’s because renv gives each project its own private package library.

  • Portable: Easily transport your projects from one computer to another, even across different platforms. renv makes it easy to install the packages your project depends on.

  • Reproducible: renv records the exact package versions you depend on, and ensures those exact versions are the ones that get installed wherever you go.

Enter renv

renv.lock

  • renv is designed to keep track of your PROJECT’S PACKAGE DEPENDENCIES
  • As your install, update, or delete packages from your project, you want to update your renv.lock file using renv::snapshot()
{
  "R": {
    "Version": "4.3.1",
    "Repositories": [
      {
        "Name": "RSPM",
        "URL": "https://packagemanager.posit.co/cran/latest"
      }
    ]
  },
  "Packages": {
    "BH": {
      "Package": "BH",
      "Version": "1.81.0-1",
      "Source": "Repository",
      "Repository": "RSPM",
      "Hash": "68122010f01c4dcfbe58ce7112f2433d"
    },
    "DBI": {
      "Package": "DBI",
      "Version": "1.1.3",
      "Source": "Repository",
      "Repository": "RSPM",
      "Requirements": [
        "R",
        "methods"
      ],
      "Hash": "b2866e62bab9378c3cc9476a1954226b"
    },
    "DT": {
      "Package": "DT",
      "Version": "0.28",
      "Source": "Repository",
      "Repository": "RSPM",
      "Requirements": [
        "crosstalk",
        "htmltools",
        "htmlwidgets",
        "jquerylib",
        "jsonlite",
        "magrittr",
        "promises"
      ],
      "Hash": "ab745834dfae7eaf71dd0b90f3b66759"
    }

renv.lock

  • renv is designed to keep track of your PROJECT’S PACKAGE DEPENDENCIES
  • As your install, update, or delete packages from your project, you want to update your renv.lock file using renv::snapshot()
  • Restore a project’s package dependencies using renv::restore()



Follow along with me - Getting started with renv

Dev-Test-Prod

Ensure your data products behave as expected and there are no surprises

Deployment Methods to Posit Connect

  1. Push-button
  2. Git-backed
  3. Programmatic

📣 Push-button is good for rapid prototyping. Beyond that, you really should be using git-backed or programmatic deployment via CI/CD pipeline to keep your code in sync with your deployment.

Git Backed Deployment

Connect will deploy directly from your git repository

Ref:

Dev-Test-Prod Methods // Git-backed Deployment

  • Deploy production and test branches

  • Automatically update production with a git merge

  • Stakeholders keep the same stable URL throughout

Dev-Test-Prod Methods // CI/CD Deployment

More Resources

That’s a wrap!
🤪 🥴 🫨 😵

Recap of Resources

https://posit-conf-2024.github.io/ds-workflows-r/resources.html

Now what?

Posit Team

Workshop Feedback

Please go to pos.it/conf-workshop-survey. Your feedback is crucial! Data from the survey informs curriculum and format decisions for future conf workshops, and we really appreciate you taking the time to provide it.

Click to go back to Data Science Workflows with Posit Tools - R Focus website ↩︎