Parse git-log Output Github Action

Thanks claude-3-opus-20240229 and LMSYS ChatBot Arena for first draft.

App skeleton to be used as github template repo.

Status

[DRAFT] [WIP] --> not fully implemented

Quickstart

• Quickstart

TOC

• Usage
• Install
• Purpose
• Reason
• Paradigms
• App Structure
• App Details
• TODO
• Inspirations
• Rescources

Usage ↑

• TODO Usage

Install ↑

• TODO Install

Purpose ↑

• TODO Purpose

Reason ↑

• TODO Reason

Paradigms ↑

• Keep to low branching factor and single outcomes
• Export complex functions into modules
• Aims for coding approach
  • Behavior Driven Design (What should it do?)
  • Test Driven Design (Does it do?)
• Aims for code quality
  • Works, i.e. passes tests which were written before
  • Modular and cohesive
  • Separation of concerns and appropriate coupling
  • Abstraction and information hiding
• Aims for CI/CD
  • Unit Testing
  • Acceptance Testing
  • Performance Testing
  • Static Analysis
  • Sign-offs
  • Security Testing
  • Scalability Testing
  • Realeasable Outcome

App Structure ↑

sh / ├─ sub1/ │ ├─ sub1sub1/ │ │ ├─ │ │ └─ │ └─ file ├─ sub2/ │ └─ file ├─ file └─ file

App Details ↑

• Details

TODO ↑

• Structure
  • [x] Test Pipfile
  • Adopted as proposed successor of requirements.txt
  • Several advantages like auto-venv and combined prod/dev in one TOML
  • pipenv with Pipfile & Pipfile.lock
  • pipenv install -e for editable mode, i.e. 'dependency resolution can be performed with an up to date copy of the repository each time it is performed'
  • [x] Use Makefile instead of self-implemented imparative setup.sh
  • Test successful and functional
  • [ ] Dynamically create a hierarchical configuration with hydra
  • [ ] Implement basic CI/CD-Skeleton
  • [ ] Create /docs with sphinx gh-action
  • [ ] Have a look at ReadTheDocs
• Coding
  • [x] Try dataclass and field from dataclasses
  • Used to auto add special classes like __init__, __str__, __repr__
  • Suitable for more complex classes and projects
  • [ ] Have a look at PyTest to prepare for TDD and BDD
  • [ ] Consistent typing and type hinting
  • [ ] Use pydantic or traitlets for type hinting or strong typing
  • [ ] Consistent usage of pydoc for /docs with pandoc
  • [ ] Decouple concerns into separate containers, e.g. avoid big container because of torch
  • Uses type hinting and decorators
  • [ ] Consistent usage of if or try for features and catches
  • [ ] Try logging instead of print()
• API
  • [ ] Try arparse
  • [ ] Implement basic API, e.g. with gunicorn or FastAPI
• Dependency tracking and app sourcing
  • [x] Test __init__.py for packing
  • Only needed for regular packages Python 3.2 and earlier
  • Python 3.3 and above uses namespace packages, see PEP 420 - Implicit Namespace Packages
  • [ ] Test conda creating an environment from an environment.yml file
  • conda env create -f environment.yml
  • [ ] Provide package as single source app version with setup.py
  • [ ] Experiment with pyproject.toml to build app wheel
• Best Practices
  • [x] Adopt SemVer
  • Using MAJOR.MINOR.PATCH (Breaking.Feature.Patch)
  • [x] Adopt CHANGELOG.md
  • Using Added, Removed, Changed and Unreleased
  • Also recommended: Deprecated, Fixed and Security
  • [x] Adopt Semantic commit messages
  • Purposful add human and machine readable meaning to commit messages
  • [ ] Adhere to Docker BP
  • [ ] Adhere to BP from The Hitchhiker's Guide to Python!

Inspirations ↑

• While-True-Do.io Repo template
• Arc-Project Pydantic
• Mala-Project Pydoc to Sphinx
• xformers Conda env file
• Jupyter Notebook structure

Resources ↑

• Development
  • Dave Farley: Test Driven Development vs Behavior Driven Development
  • Dave Farley: How to Build a DEPLOYMENT PIPELINE? (Continuous Delivery)
  • Martin Fowler: