Project Settings

The [project] and [source] sections define project metadata and control which files Batuta processes.

[project] Section

[project]
name = "my-project"
description = "A Python ML pipeline migrated to Rust"
primary_language = "Python"
authors = ["Alice <alice@example.com>", "Bob <bob@example.com>"]
license = "MIT"

Key	Type	Default	Description
`name`	string	`"untitled"`	Project name used in generated Cargo.toml and reports
`description`	string	(none)	Optional project description
`primary_language`	string	(none)	Primary source language (`Python`, `C`, `Shell`, `Rust`)
`authors`	array	`[]`	List of author strings
`license`	string	`"MIT"`	SPDX license identifier

When you run batuta init, the name is inferred from the directory name and primary_language is detected by file extension analysis.

[source] Section

[source]
path = "."
exclude = [".git", "target", "build", "dist", "node_modules", "__pycache__", "*.pyc", ".venv", "venv"]
include = []

Key	Type	Default	Description
`path`	string	`"."`	Root directory for source analysis (relative to config file)
`exclude`	array	See below	Glob patterns for files and directories to skip
`include`	array	`[]`	Glob patterns that override exclude rules

Default Exclude Patterns

The following patterns are excluded by default to skip build artifacts, virtual environments, and version control metadata:

.git, target, build, dist
node_modules, __pycache__, *.pyc
.venv, venv

Include Overrides

The include array takes precedence over exclude. Use it to pull specific files back into scope.

[source]
exclude = ["tests"]
include = ["tests/integration"]  # Keep integration tests, skip unit tests

Workspace Configuration

For monorepo or multi-crate projects, set path to the workspace root and use exclude to skip directories that should not be transpiled.

[source]
path = "."
exclude = [".git", "target", "docs", "scripts", "infra"]

Batuta traverses the source tree recursively from path, respecting the exclude and include filters at every level.

Navigate: Table of Contents