pyre_tools

pyre_tools is a Python dependency analysis tool that works both as a CLI and as an importable module.

Install

pip install pyre_tools

CLI

Analyze a project:

pyre_tools path/to/project

Analyze and install missing dependencies:

pyre_tools path/to/project --install

Generate requirements.txt and pyproject.toml:

pyre_tools path/to/project --write-files

The legacy pyre command is still available as a compatibility alias.

pyre_tools -h also shows how to call the package as a module from other Python code. If you hit Ubuntu/Debian user-site pip environment issues, run:

pyre --repair-hints

The CLI defaults to English for terminal compatibility. You can switch languages with:

pyre --lang zh path/to/project
pyre --lang en path/to/project

Python API

import pyre_tools; pyre_tools.shield()

Put that line at the top of your entry script. pyre_tools will infer the current project, install missing dependencies, and restart the process once if it had to patch the environment.

shield() now does a fast preflight on the current script first:

• local project modules like utils.py or utils/ are treated as already satisfied
• already importable dependencies are accepted immediately
• only real missing modules fall back to full project analysis and installation
• in REPL / python -c mode, it falls back to the current working directory instead of trying to parse <stdin> or <string>

Public API

• analyze_path(path, include_stdlib=False)
• bootstrap(path=".", auto_install=True, output_dir=None, write_files=False, include_stdlib=False, upgrade=False)
• check_and_install_dependencies(path=".", auto_install=True, output_dir=None, write_files=False, include_stdlib=False, upgrade=False)
• shield(path=None, auto_install=True, output_dir=None, write_files=False, include_stdlib=False, upgrade=False, restart_on_install=True)
• find_missing_requirements(report_or_requirements)
• install_requirements(requirements, python_executable=None, upgrade=False)
• export_requirements(requirements, output_path)
• export_pyproject_toml(requirements, output_path, project_name)
• get_install_hint(package_name="pyre_tools")
• format_module_not_found_hint(module_name="pyre_tools", package_name="pyre_tools")

Notes

• The importable module name is pyre_tools.
• The primary console command is also pyre_tools.
• This makes missing-module guidance straightforward: if users see ModuleNotFoundError: No module named 'pyre_tools', the install command is pip install pyre_tools.
• The shortest module integration is import pyre_tools; pyre_tools.shield().