prefix-dev/pixi

Features

• Conda Environment Reproducibility - Creates reproducible, lock-file-backed conda environments with cross-platform support and multi-language dependency resolution.
• Task Definitions - Defines cross-platform commands in the manifest with arguments, dependencies, working directory, and environment variables.
• Multi-Source Dependency Resolutions - Installs and coordinates packages from conda-forge and PyPI for Python, C++, Rust, Node.js, and other languages in a single project.
• Cross-Platform Task Pipelines - Defines and runs multi-step command pipelines that work identically on Linux, macOS, and Windows inside the project environment.
• Lock-File-Based Cache Keys - Generates cache keys from lock-file content to automatically save and restore environments in CI.
• Environment-Aware Execution - Executes defined tasks or arbitrary commands inside the current workspace environment.
• Manifest-Defined Tasks - Creates cross-platform, repeatable commands or chains of commands in the manifest to share and run consistently.
• Cross-Platform Development Tools - Runs identically on Linux, macOS, and Windows, with environment definitions that work across all operating systems.
• Global Tool Isolation - Installs command-line tools into their own isolated environments, making them system-wide available.
• Python Package Managers - Adds Python dependencies from PyPI alongside conda packages using uv integration.
• Multi-Source Package Managers - Coordinates dependency resolution and installation from both conda-forge and PyPI simultaneously.
• Environment Activation - Makes all binaries and environment variables from a pixi environment available to subsequent workflow steps.
• CI - Makes all pixi-installed binaries and environment variables available to subsequent workflow steps.
• Environment Caching Utilities - Caches project environments using lock-file hashes to skip reinstallation on cache hits in CI.
• Lock-File-Based Environment Caching - Stores and restores project and global pixi environments using a lock-file-based hash to skip reinstallation.
• Multi-Environment Manifests - Defines multiple isolated environments within a single manifest, each with its own dependency set.
• Environment Managers - Composes several distinct dependency sets in a single manifest file for different workflows.
• Conda - Creates reproducible, lock-file-backed conda environments with cross-platform support and multi-language dependency resolution.
• Multi-Environment Composition - Defines multiple isolated environments within a single manifest, each with its own dependency set.
• Isolated Execution Environments - Executes commands or opens interactive shells within the project's isolated environment.
• Package Dependency Management - Adds packages from conda or PyPI to the workspace manifest, resolving all dependencies and installing them.
• Package Dependency Managers - Installs and coordinates packages from conda-forge and PyPI for multiple programming languages.
• Multi-Manifest Workspaces - Manages multiple sub-projects with independent manifests and dependencies within a single repository.
• Project Task Execution - Executes user-defined scripts or commands inside the project's isolated environment, managing complex pipelines.
• Cross-Platform Lock Files - Produces a single lock file that resolves packages for all platforms, ensuring reproducible installs across operating systems and architectures.
• Cross-Platform - Executes user-defined commands and multi-step pipelines inside isolated project environments on Linux, macOS, and Windows.
• Dependency Lock Files - Installs dependencies with frozen or locked flags to ensure reproducible environments from the lock file.
• Conda Installers - Pulls packages from conda-forge or custom channels into the project environment.
• Reproducible Environments - Creates isolated, lock-file-backed environments that can be exactly recreated on any machine.
• Lock-File Pinning - Pins every transitive dependency to exact versions in a cross-platform lock file for reproducibility.
• Automatic Lock-File Regenerations - Automatically regenerates the lock file when the manifest or installed environment falls out of sync.
• Multi-Source Dependency Resolution - Resolves and installs packages from both conda-forge and PyPI simultaneously, coordinating their dependency graphs.
• Task Execution Engines - Runs user-defined multi-step command pipelines as named tasks with dependencies and environment variables.
• Manifest-Defined Pipelines - Defines and runs complex sequences of commands as named tasks within a project manifest.
• Global Installation Managers - Installs command-line tools in isolated environments so they are available system-wide without conflicts.
• Multi-Environment Installers - Installs and caches multiple environments defined in the project manifest, each with its own set of dependencies.
• Lock File Updates - Modifies the lock file to reflect manifest changes without altering the installed environment.
• Workspace Scaffolding - Creates new project directories with a manifest file and directory structure from a single command.
• Monorepo Managers - Manages multiple sub-projects with their own manifests and dependencies within a single repository from a root directory.
• Multi-Source Dependency Additions - Adds packages from PyPI alongside conda dependencies, preventing conflicts between the two sources.
• Registry Authentication - Authenticates against package registries using tokens, username and password, conda tokens, S3 keys, or keyring.
• Shell Environment Managers - Opens interactive shell sessions configured with the workspace's environment variables and paths.
• Shell Execution Environments - Opens interactive shell sessions inside the project's isolated environment for direct command execution.
• Environment-Aware Command Executions - Executes shell commands or scripts inside the pixi environment using a custom shell wrapper without prefixing each command.
• Project Initializers - Creates new project directories with a manifest file and directory structure from a single command.
• Subdirectory Configurations - Runs pixi commands from a specified subdirectory in monorepos where the project is not at the repository root.
• Workspace Initialization Templates - Creates a new project directory with a manifest file ready for managing dependencies and environments.
• Environment Caching - Ships a dedicated GitHub Actions action that caches and restores conda environments using lock-file hashes.
• Environment Caching in CI - Provides a GitHub Actions action that caches and restores conda environments using lock-file hashes to speed up CI workflows.
• Package Installations - Installs conda packages into their own isolated environment, making them available system-wide.
• Isolated Global Installations - Installs conda packages into their own isolated environment, making commands available system-wide.
• Manifest-Lock Synchronization - Installs only when the lock file is fully up-to-date with the manifest, failing otherwise.
• Environment Caching Actions - Provides a GitHub Actions action that caches and restores conda environments using lock-file hashes to speed up CI workflows.
• Subdirectory Configurations - Specifies a subdirectory within a monorepo where the tool should look for manifest files and run commands.

Star history