IDEALIT Passport
Menu

Passport search

Find a mission or fix

Type at least two characters.

    Browse help by symptomBrowse all references
    How to start
    Curriculum previewThis is not your assigned passport. No identity, answers, or completion progress are stored here.How to start your passport
    Handbook / labs

    labs

    Lab: Reproducible Python Environment

    Goal: prove that a project can be installed and checked from a newly created environment using declared dependencies.

    View exact sourceReturn to my dashboard

    Goal: prove that a project can be installed and checked from a newly created environment using declared dependencies.

    Prerequisites: Python setup and a small Python training/project repository.

    Execution context: run on your local computer in Windows PowerShell, macOS zsh, or Linux Bash. Do not install Conda in Euler project/work storage.

    1. Inspect The Project Contract#

    Identify which files declare dependencies and commands, for example:

    • pyproject.toml;
    • environment.yml;
    • requirements.txt;
    • README.md;
    • Makefile or task configuration.

    Ask an agent to explain them without editing:

    Goal: explain how to recreate and verify this Python project.
Context: inspect README and dependency/build configuration.
Constraints: do not install packages or edit files.
Verification: identify the exact test and lint commands from repository evidence.
Output: environment creation, installation, and verification steps, with any
uncertainty labelled.

    2. Create A Fresh Environment#

    Use this fixed training-environment name so the commands work in PowerShell, zsh, and Bash without shell-specific variables:

    conda create -n passport-python python=3.11 -y
conda activate passport-python
python --version
python -m pip --version

    Install the project exactly as its README specifies. For a package with development extras, this may be:

    python -m pip install -e '.[dev]'

    Do not copy a command from this example when the repository declares a different method.

    3. Verify From Declared Commands#

    Typical checks are:

    pytest
ruff check .

    Use only commands actually supported by the repository.

    4. Check Repository Cleanliness#

    git status --short

    The environment, caches, downloaded data, and generated results must not appear as untracked project content. Update .gitignore only for project-appropriate patterns, not to hide source files you do not understand.

    Expected Result#

    • The selected Python interpreter belongs to the fresh environment.
    • Installation succeeds from declared project files.
    • Tests/checks run with recorded output.
    • Creating and using the environment does not dirty the repository.

    Common Failures And Safe Recovery#

    • Wrong Python selected: deactivate, reactivate the intended environment, and recheck python --version and python -m pip --version.
    • Works only in an old environment: compare declared dependencies instead of copying the old environment wholesale.
    • GPU package fails locally: reproduce CPU functionality first; platform GPU installation is a separate requirement.

    Understand Before Accepting AI Output#

    • I know which file declares each required dependency.
    • I verified the interpreter and pip belong to the same environment.
    • I ran the checks myself.
    • I can recreate the environment without relying on chat history.

    Evidence#

    Provide environment creation, installation, and verification commands plus their concise outcomes. Do not include package-registry tokens or environment variables containing secrets.

    Ask For Help When#

    Dependency declarations conflict, installation requires unpublished/private packages without documented access, or platform constraints are undocumented.