Skip to content

AGENTS.md

Latest commit

 

History

History
81 lines (59 loc) · 4.05 KB

AGENTS.md

File metadata and controls

81 lines (59 loc) · 4.05 KB

AGENTS.md

Project objective

Leapp-repository contains the actors, models, and libraries for RHEL in-place upgrades (IPU). It runs on top of the Leapp framework, which provides the actor runtime, workflow execution, messaging, and core CLI behavior. It's important to notice, that framework-level mechanics belong to leapp, while upgrade-path logic belong to leapp-repository.

Architecture

Leapp IPU content is actor-based and message-driven:

  • Actors exchange typed messages (consumes / produces) through workflow phases.
  • Messages only flow forward: a producer must run before its consumer. Within the same phase, the framework resolves ordering automatically.
  • Prefer scanner/checker design: collect facts in one actor, evaluate in another.
  • Keep actor process() thin; place implementation logic in actor libraries for testability.

Phase workflow (always apply)

  • FactsCollectionPhase: collect system facts and produce messages
  • ChecksPhase: consume previously collected facts, report/inhibit compatibility issues
  • If modification is required, do it in later phases (ApplicationsPhase or ThirdPartyApplicationsPhase).
  • FinalPhase: to perform actions after the upgraded system is booted.

For full phase details if ambiguous :

Repository Layout

├── system_upgrade/
    ├── common/                      # common repository/content used and accessible for all system upgrades
    │   ├── actors/<name>/
    │   │   ├── actor.py             # Actor class definition
    │   │   ├── libraries/           # Actor-private logic
    │   │   └── tests/               # Actor tests (unit_test_*.py, test_*.py)
    │   ├── models/                  # Data models (consumed/produced by actors)
    │   ├── topics/                  # Message topics
    │   └── libraries/               # Shared libraries (importable via leapp.libraries.common)
    ├── el8toel9/                     # content specific only for  RHEL 8 -> 9 upgrades
    └── el9toel10/                    # content specific only for  RHEL 9 -> 10 upgrades

General rules (always apply)

  • Follow leapp specific guidelines: docs/source/contributing/coding-guidelines.md and project conventions first.
  • Follow Python coding guidelines: https://leapp.readthedocs.io/en/stable/contributing.html
  • Analyze and plan before introducing code.
  • Ask, don't assume. Be verbose when user input is unclear or can be interpreted in many ways.
  • Explain your decisions briefly; expand detail only when requested.
  • If multiple approaches or solution exists, pick simpler one.
  • Be ready to undo your changes.
  • Before introducing new code, check if it's not already solved in codebase. Avoid code duplication.
  • Try to write as simple code as possible. New code should easy to read and understand even for non-experienced developers.
  • Don't change code, which is not directly involved in given task. Don't try to improve unrelated code.
  • Don't perform git commit, git push, or create PRs without explicit user request.

Common Commands

Quick reference (details in skills below):

make lint                                    # pylint + flake8 + isort
TEST_CONTAINER=el9 make test_container       # full CI-equivalent test
ACTOR=<name> make dev_test_no_lint           # single-actor fast test

Skills

Load the relevant skill for the task at hand:

  • Run tests / lint → skills/leapp-test-runner/
  • Write or fix unit tests → skills/leapp-unit-tests/
  • Develop or modify actors → skills/leapp-actor-dev/

Definition of Done

  • Scope is clear and limited to the requested change.
  • New or changed code is covered by unit tests.
  • Relevant checks were run (make lint and/or targeted tests) and results are reported.
  • Risks, assumptions, and follow-up work are explicitly called out