Create package_management.md

Author: isedwards

python/package_management.md

@@ -0,0 +1,56 @@
+1. **`distutils`**:
+   - Relevance: Deprecated.
+   - Purpose: Originally the standard for packaging Python modules. It's now outdated and has been officially deprecated in Python 3.10.
+   - Action: Avoid using it in new projects. 
+
+2. **`setuptools`**:
+   - Relevance: Very Relevant.
+   - Purpose: An enhancement and extension of `distutils`. It's currently the most widely used tool for packaging and distributing Python packages.
+   - Action: Use for traditional packaging tasks, especially if you're maintaining existing projects.
+
+3. **`twine`**:
+   - Relevance: Very Relevant.
+   - Purpose: Utility for publishing Python packages on PyPI. It provides a safer way to upload packages compared to `setup.py upload`.
+   - Action: Use when you want to release/distribute your package on PyPI.
+
+4. **`pip`**:
+   - Relevance: Very Relevant.
+   - Purpose: The package installer for Python. It's the standard way to install Python packages.
+   - Action: Use regularly for installing Python packages.
+
+5. **`poetry`** (mentioned twice in your list):
+   - Relevance: Very Relevant.
+   - Purpose: Dependency management and packaging tool. It offers a more holistic and integrated approach to package creation, dependency management, and publishing.
+   - Action: Consider for new projects, especially if you want integrated dependency management and packaging.
+
+6. **`flit`**:
+   - Relevance: Relevant.
+   - Purpose: Simplified packaging, especially for pure-Python libraries. It aims to be more straightforward than `setuptools`.
+   - Action: Consider for simpler projects, especially those without complex build processes or dependencies.
+
+7. **`easy_install`**:
+   - Relevance: Deprecated.
+   - Purpose: Was a package installer associated with `setuptools`.
+   - Action: Avoid using. Always prefer `pip`.
+
+8. **`pipenv`**:
+   - Relevance: Relevant.
+   - Purpose: Offers dependency management for Python projects, aiming to bring the best of all packaging worlds to the Python world. It automatically creates and manages a virtual environment for your projects.
+   - Action: Use if you prefer its workflow and need both package management and virtual environment management integrated.
+
+9. **`conda`**:
+   - Relevance: Very Relevant (especially in data science and scientific computing).
+   - Purpose: An open-source package management system and environment management system. It's not limited to Python; it can package any software. It's particularly popular with data scientists using the Anaconda distribution.
+   - Action: Use if you're working in scientific computing, need non-Python packages, or are using the Anaconda distribution.
+
+To summarize:
+
+- For **general package management and installation**: Use `pip`.
+  
+- For **traditional package creation and distribution**: Use `setuptools` and `twine`.
+  
+- For **modern package creation, dependency management, and integrated workflows**: Consider `poetry` or `pipenv`.
+  
+- For **scientific computing or non-Python package management**: Consider `conda`.
+
+- Avoid deprecated tools like `distutils` and `easy_install`.