code-trainer 1.0.0

Systeme d'entrainement

Trainer — CLI Exercise Trainer

Trainer is a command-line exercise training tool to manage, run and validate programming exercises. It supports importing repositories, installing exercises into a working directory, running an interactive sandboxed session, and validating solutions.

Installation

Install the package in editable/development mode:

pip install -e .

This will install the package and its runtime dependencies (for example PyYAML). After installation the trainer command (the CLI) is available.

Available Commands

The CLI supports the following subcommands (see trainer <cmd> -h for full options):

• import <git-url> — Import an exercises repository into the local trainer base directory.

  • Options: --base-dir (default: ~/.trainer)

• ls — List exercises available in the base directory.

  • Options: --base-dir, --tag, -d/--done (show already completed), -r/--repo (filter by repository name)

• install <exs> — Install an exercise into a (usually empty) target directory.

  • Options: --base-dir (where exercises are stored), --user-dir

• exec <exs> — Create a temporary working directory for the exercise, install it there and launch an interactive shell (with timer support).

  • Options: --base-dir

• check — Run the exercise validation command defined in the exercise config.yml in the current or provided directory.

  • Options: --base-dir, --user-dir

• update — Pull updates for repositories in the base directory (uses git).

  • Options: --base-dir, --repo (optional specific repo)

• stats — Show history statistics (done/failed exercises).

  • Options: --filter

• recommend <reference_alias> — Recommend similar exercises based on tags and difficulty.

  • Options: --base-dir, --top-k, --include-done, --allow-lower-difficulty, --min-common-tags

Note: some internal code paths reference a time action for reporting timer state, but time is not defined as a top-level subcommand in the CLI argument parser. Use check and the timer mechanisms that run with exec and install workflows.

Typical workflow

1. Import repositories that contain exercises:

trainer import https://github.com/your-org/exercises.git

2. List available exercises:

trainer ls --base-dir ~/.trainer

3. Run an exercise in a sandboxed shell (creates a temporary working directory):

trainer exec min_1

4. Work on the exercise in the interactive shell and then validate the solution from within the working directory (or from your user dir):

trainer check --user-dir /path/to/your/workdir

5. Optionally install an exercise into a user directory (useful for sharing or persistent workspaces):

trainer install min_1 --user-dir ~/workspace/min_1

Configuration

Exercises may include a config.yml with fields such as commands, timeout, tags, and description. The check command runs the commands.validate command (if defined) from the merged configuration (repository config.yml + exercise config.yml).

The trainer stores an index file .trainer_index.yml in the base directory (default ~/.trainer) and history in ~/.trainer/history.yml.

Help

General help:

trainer -h

Help for a specific command, e.g.:

trainer import -h
trainer ls -h
trainer install -h
trainer exec -h
trainer check -h

Repository

Source repository :

https://github.com/RomainGueninchault/pfa_systeme