Introduction | A word of caution | Installing old packages archived on CRAN | Installing packages from GitHub | Installing local archives | Installing packages from private repositories | Installing Python packages | Installing Python packages not available via nixpkgs (impure) | Installing Julia packages | Converting from an renv.lock file | A complete example | Package installation issues

Part 1: R Packages with | Setting up the Development Environment | Initializing the Package | Workflow Basics | Using Your Package in a Nix Environment | Part 2: Python Packages with uv | Configuring pyproject.toml | Development Workflow | Local Development Loop

Introduction | Why rix and Nix? | Installing Nix | Windows prerequisites | Using the Determinate Systems installer | If you installed Nix another way | Recommended: setup_cachix() | Alternative: the cachix client | Optional: add Mischko Heming's binary cache | NixOS users | Using Mischko Heming's binary cache | A note on disk space | Installing rix | Case 1: you don't have R installed and want to install it via Nix | Case 2: you already have R installed via your OS's package manager | Configuring an IDE | Other "Nix"es

Introduction | Why rix and Nix? | Installing Nix | Recommended: setup_cachix() | Alternative: the cachix client | Installing rix | Case 1: you don't have R installed and want to install it via Nix | Case 2: you already have R installed | More macOS specificities | R support for Apple Silicon in nixpkgs | Shared libraries issue | RStudio and other development interfaces on macOS | Why do we need all these special tweaks? | Path of the Nix installation not in PATH | Configuring an IDE | Other "Nix"es

Introduction | A word of caution | Installing old packages archived on CRAN | Installing packages from GitHub | Installing local archives | Installing packages from private repositories | Installing Python packages | Installing Python packages not available via nixpkgs (impure) | Installing Julia packages | Converting from an renv.lock file | A complete example | Package installation issues

Part 1: R Packages with | Setting up the Development Environment | Initializing the Package | Workflow Basics | Using Your Package in a Nix Environment | Part 2: Python Packages with uv | Configuring pyproject.toml | Development Workflow | Local Development Loop

Introduction | Why rix and Nix? | Installing Nix | Windows prerequisites | Using the Determinate Systems installer | If you installed Nix another way | Recommended: setup_cachix() | Alternative: the cachix client | Optional: add Mischko Heming's binary cache | NixOS users | Using Mischko Heming's binary cache | A note on disk space | Installing rix | Case 1: you don't have R installed and want to install it via Nix | Case 2: you already have R installed via your OS's package manager | Configuring an IDE | Other "Nix"es

Introduction | Why rix and Nix? | Installing Nix | Recommended: setup_cachix() | Alternative: the cachix client | Installing rix | Case 1: you don't have R installed and want to install it via Nix | Case 2: you already have R installed | More macOS specificities | R support for Apple Silicon in nixpkgs | Shared libraries issue | RStudio and other development interfaces on macOS | Why do we need all these special tweaks? | Path of the Nix installation not in PATH | Configuring an IDE | Other "Nix"es

Non-technical intro to Nix | The Nix package manager | rix workflow | Converting from an renv.lock file | Using default.nix files

Non-technical intro to Nix | The Nix package manager | rix workflow | Converting from an renv.lock file | Using default.nix files

Introduction | Setting up Git | Setting up your environment | Using your environment | Using GitHub Actions | For a project using | For a project that doesn't use | How to update or reproduce your project

Introduction | Setting up Git | Setting up your environment | Using your environment | Using GitHub Actions | For a project using | For a project that doesn't use | How to update or reproduce your project

Introduction | Build Logs: The Key to Your Pipeline's History | Cleaning Up Artifacts with rxp_gc() | 1. Targeted Deletion: Cleaning Up Old Builds | 2. Full Garbage Collection | Conclusion

Large Pipelines Become Unwieldy | Basic Usage | Visualising Sub-Pipelines | Switching Between Colour Modes | How It Works Internally | Best Practices | Conclusion

Introduction | Build Logs: The Key to Your Pipeline's History | Cleaning Up Artifacts with rxp_gc() | 1. Targeted Deletion: Cleaning Up Old Builds | 2. Full Garbage Collection | Conclusion

Large Pipelines Become Unwieldy | Basic Usage | Visualising Sub-Pipelines | Switching Between Colour Modes | How It Works Internally | Best Practices | Conclusion

Introduction | Starting with the basics (simple but not entirely reproducible) | 100% reproducible literate programming | Conclusion

Introduction | Setting up | Where to find packages to fix | Case 1: packages that need dependencies to build | Case 2: packages that need a home, X, or simple patching | Case 3: packages that require their attributes to be overridden | Case 4: packages that need a dependency that must be overridden | Case 5: darwin-specific fixes | Case 6: an R packages requires software not in nixpkgs (or outdated in nixpkgs) | Study! Study! Study! Study! | Broken packages | Reviewing PRs

Introduction | Converting an historical renv project | Starting a new project | Caveats | Package versions are not exactly the same between the renv.lock and default.nix files | Don’t use the same folder for your Nix and {renv} projects | Mind the R version

Why bring {rix} into your package workflow? | Declaring your package environment | Committing the environment to your repository | Keeping the snapshot fresh | Using {rix} in GitHub Actions | Summary

Introduction | The {lookup} package | Caveats | Fixed commits | Duplicated packages | Authenticating to GitHub | Consider using dependencies that made it to CRAN | Custom Git hosting platforms | Conclusion

Introduction | Adding tools to an environment | Installing TexLive packages | Installing IDEs

Introduction | Using GitHub Actions to build the environment | Using your cache

Introduction | R packages available through Nix | Using bleeding and frozen edge package for your environments | Making your own cache

Introduction | Our recommendation | Recommended setup on macOS | Recommended setup on Windows | Recommended setup on Linux | RStudio | RStudio on macOS | RStudio on Linux or Windows | Positron | Configuring native Positron to interact with Nix shells | Letting Nix manage Positron on any operating system | VS Code/Codium | Configuring native VS Code/Codium to interact with Nix shells | Letting Nix manage VS Code/Codium | Emacs | Configuring native Emacs to interact with Nix shells | Letting Nix manage Emacs | Making a launcher to start your editor

Introduction | A generic Dockerfile | Dockerizing a Shiny application | NixOS

Project-specific Nix environments | Which r_ver should you choose? | Running old projects with | Running programs from an environment | Running single functions in a subshell | Nix environments are not completely isolated from your system

Introduction | Starting with the basics (simple but not entirely reproducible) | 100% reproducible literate programming | Conclusion

Introduction | Setting up | Where to find packages to fix | Case 1: packages that need dependencies to build | Case 2: packages that need a home, X, or simple patching | Case 3: packages that require their attributes to be overridden | Case 4: packages that need a dependency that must be overridden | Case 5: darwin-specific fixes | Case 6: an R packages requires software not in nixpkgs (or outdated in nixpkgs) | Study! Study! Study! Study! | Broken packages | Reviewing PRs

Introduction | Converting an historical renv project | Starting a new project | Caveats | Package versions are not exactly the same between the renv.lock and default.nix files | Don’t use the same folder for your Nix and {renv} projects | Mind the R version

Why bring {rix} into your package workflow? | Declaring your package environment | Committing the environment to your repository | Keeping the snapshot fresh | Using {rix} in GitHub Actions | Summary

Introduction | The {lookup} package | Caveats | Fixed commits | Duplicated packages | Authenticating to GitHub | Consider using dependencies that made it to CRAN | Custom Git hosting platforms | Conclusion

Introduction | Adding tools to an environment | Installing TexLive packages | Installing IDEs

Introduction | Using GitHub Actions to build the environment | Using your cache

Introduction | R packages available through Nix | Using bleeding and frozen edge package for your environments | Making your own cache

Introduction | Our recommendation | Recommended setup on macOS | Recommended setup on Windows | Recommended setup on Linux | RStudio | RStudio on macOS | RStudio on Linux or Windows | Positron | Configuring native Positron to interact with Nix shells | Letting Nix manage Positron on any operating system | VS Code/Codium | Configuring native VS Code/Codium to interact with Nix shells | Letting Nix manage VS Code/Codium | Emacs | Configuring native Emacs to interact with Nix shells | Letting Nix manage Emacs | Making a launcher to start your editor

Introduction | A generic Dockerfile | Dockerizing a Shiny application | NixOS

Project-specific Nix environments | Which r_ver should you choose? | Running old projects with | Running programs from an environment | Running single functions in a subshell | Nix environments are not completely isolated from your system

Getting data into the pipeline | Declaring build steps | Generating the pipeline | Defining a Reproducible Shell for Execution | Optional steps before building the pipeline | Graphical representation of the pipeline's DAG | Tracing the lineage of derivations | Building and inspecting outputs | No-op builds for individual derivations | Caveats | Conclusion

Analysing the mtcars dataset using R and Python | Building a Quarto or R Markdown document

Installing the required software | Bootstrapping a project

Getting data into the pipeline | Declaring build steps | Generating the pipeline | Defining a Reproducible Shell for Execution | Optional steps before building the pipeline | Graphical representation of the pipeline's DAG | Tracing the lineage of derivations | Building and inspecting outputs | No-op builds for individual derivations | Caveats | Conclusion

Analysing the mtcars dataset using R and Python | Building a Quarto or R Markdown document

Installing the required software | Bootstrapping a project

The Problem: Silent Failures | What is {chronicler}? | The Challenge with Nix Builds | The Solution: Automatic Chronicle Checking | Three-State Status System | Complete Example | Automatic Warnings | Best Practices | When to Use Chronicler | Further Reading

The Problem: Silent Failures | What is {chronicler}? | The Challenge with Nix Builds | The Solution: Automatic Chronicle Checking | Three-State Status System | Complete Example | Automatic Warnings | Best Practices | When to Use Chronicler | Further Reading

Running pipelines on GitHub Actions

Running pipelines on GitHub Actions

Introduction | The First Response to a Failed Build: rxp_inspect() | Investigating Structural Issues with rxp_trace() | A Proactive Strategy: Isolating Derivations with noop_build | Historical Debugging: Going Back in Time | Conclusion

Introduction | Custom Encoding and Decoding in R | Cross-Language Data Transfer: R ↔ Python | Summary

Importing a single local file | Importing a single file from the internet | Importing many files from a directory | Importing files with dependencies (e.g., Shapefiles) | Conclusion

Definitions | Derivations | Dependencies of derivations | The Nix store and hermetic builds | Summary and conclusion

Generating waveshaders data using Julia and plotting it using R | Setting up the environment | Creating the pipeline | Helper functions | Data transfer between R and Julia | Conclusion

Overview | What rixpress is | What rixpress is not | Primary audience | In-scope features (current) | Out-of-scope features (not planned) | Roadmap | Near-term (next minor releases) | Maybe later | Not planned | How to propose new features | Related projects and inspiration | Links

Setting up the Environment | Setting up the pipeline | Compiling and Sampling the Model | Custom Serialisation | Summary

Introduction | The First Response to a Failed Build: rxp_inspect() | Investigating Structural Issues with rxp_trace() | A Proactive Strategy: Isolating Derivations with noop_build | Historical Debugging: Going Back in Time | Conclusion

Introduction | Custom Encoding and Decoding in R | Cross-Language Data Transfer: R ↔ Python | Summary

Importing a single local file | Importing a single file from the internet | Importing many files from a directory | Importing files with dependencies (e.g., Shapefiles) | Conclusion

Definitions | Derivations | Dependencies of derivations | The Nix store and hermetic builds | Summary and conclusion

Generating waveshaders data using Julia and plotting it using R | Setting up the environment | Creating the pipeline | Helper functions | Data transfer between R and Julia | Conclusion

Overview | What rixpress is | What rixpress is not | Primary audience | In-scope features (current) | Out-of-scope features (not planned) | Roadmap | Near-term (next minor releases) | Maybe later | Not planned | How to propose new features | Related projects and inspiration | Links

Setting up the Environment | Setting up the pipeline | Compiling and Sampling the Model | Custom Serialisation | Summary

Introduction | An example of a reproducible analytical pipeline using Nix

Testing code in evolving software dependency environments with confidence | Two operational modes of computations in environments: 'System-to-Nix' and 'Nix-to-Nix' | Case study 1: Evolution of base R | Setting up the (R) software environment with Nix | Defining and interactively testing custom R code with function(s) | Run functioned up code and investigate results produced in pure Nix Rsoftware environments | Syntax option for specifying function in expr argument of with_nix() | Comparing as.vector.data.frame() for both R versions 4.1.3 and 4.2.0 from Nixpkgs | Case study 2: Breaking changes in {stringr} 1.5.0 | Case study 3: Using a subshell to get hard to install dependencies

Introduction | An example of a reproducible analytical pipeline using Nix

Testing code in evolving software dependency environments with confidence | Two operational modes of computations in environments: 'System-to-Nix' and 'Nix-to-Nix' | Case study 1: Evolution of base R | Setting up the (R) software environment with Nix | Defining and interactively testing custom R code with function(s) | Run functioned up code and investigate results produced in pure Nix Rsoftware environments | Syntax option for specifying function in expr argument of with_nix() | Comparing as.vector.data.frame() for both R versions 4.1.3 and 4.2.0 from Nixpkgs | Case study 2: Breaking changes in {stringr} 1.5.0 | Case study 3: Using a subshell to get hard to install dependencies

Introduction | Monadic laws | First law | Second law | Third law | flatmap() for chronicle objects

Introduction | Basic Recording and Inspection | Handling Failures Gracefully | Capturing Errors | Controlling Strictness | Conclusion