project_manager

A conda poetry project manager - replacing .sh/.bat scripts with Python

A collection of actions that allow you to use subprocess to create/manage conda envs, poetry projects, register ipykernels, and export Sublime Text 3 Python Build configuration settings.

Things are actively being refactored and haven't been documented and annotated. As such, method names are long and verbose for the meantime.

This package was created in order to practice replacing shell/.bat scripts with Python, and to allow me to use conda to install packages (via poetry) not available for computers with a Mac M1 chip. The Sublime Text 3 build config generator was created in order to be able to execute Python code with little performance issues I've been experiencing with my IDE.

For some methods found in this module to work you need to have the following:

• Required

  • Anaconda/Conda
  • Poetry

• Optional

  • gy
    • Used for creating .gitignore files

Note

Some method names might undergo name changes as documentation is added.

Note many of these commands were run on a Mac. Do modify the snippets as needed to best suit your case, and raise any issues

Installation Using Poetry

poetry add git+https://github.com/joeld1/project-manager.git#main

Note: If you're using poetry 1.2.0a1, make sure to set the following poetry config setting

experimental.new-installer = true

Examples

Create a Conda Environment and Register Kernel

Running the following command:

import sys

from project_manager import CondaEnvManager as CEM

# Don't add illegal characters or spaces
conda_env_name = "hello_world"

# Version of python for conda env
python_version = "3.9"

CEM.create_and_init_conda_env(conda_env_name, python_version)

sys.exit()

Execute poetry init --no-interaction

Running the following command activates a conda env and runs poetry init --no-interaction:

import sys

from project_manager import PoetryProjectManager as PPM

# Don't add illegal characters or spaces
# the poetry project will be the same python version as the conda env specified
conda_env_name = "hello_world"

# you can specify a directory or leave it as None for current directory
poetry_proj_dir = None

PPM.execute_poetry_init(conda_env_name,poetry_proj_dir)

sys.exit()

Catch a ModuleNotFoundError and add missing module to pypoetry.toml

The following lines of code can help you add a missing module to your pyproject.toml file. This only works if the py script you are running is already part of a poetry project folder, and if a poetry.toml file is located in the directory of any parents of the .py file containing the below block of code. If a pyproject.toml isn't found in any parents it'll fail.

import sys

from project_manager import PoetryProjectManager as PPM

try:
    import some_local_module_with_missing_dependencies_not_found_in_your_pypoetry_toml
except Exception as e:
    PPM.add_poetry_package_from_exception(__file__,e,ignore_version=True)

sys.exit()

Uninstall a conda env

import sys

from project_manager import CondaEnvManager as CEM

conda_env_name = "hello_world"
rc = CEM.uninstall_conda_env(conda_env_name)

assert rc == 0

sys.exit()

Uninstall a jupyter kernel

import sys

from project_manager import CondaEnvManager as CEM

kernel_name = "hello_world"
rc = CEM.uninstall_kernel(kernel_name)

assert rc == 0

sys.exit()

Uninstall a jupyter kernel and conda env in 1 go (different names too)

import sys

from project_manager import CondaEnvManager as CEM

conda_env_name = "hello_world"
kernel_name = "hello_world"

# specify 1 name
# The default assumes that conda_env_name == kernel_name
CEM.uninstall_conda_and_kernel(conda_env_name)

# specify 2 names if both are different
CEM.uninstall_conda_and_kernel(conda_env_name="hello_world",kernel_name="some_other_name")

sys.exit()

Executing git init

import sys

from project_manager import GitProjectManager as GPM

# None for current directory
path_to_dir = None

# Otherwise specify the folder path
path_to_dir = "path/to_some/folder"

GPM.init_dir(path_to_dir)

sys.exit()

(optional) Use gy to generate a .gitignore file (requires gy)

This adds the following gitignore files to the directory passed in

git_ignores_to_add = ['dropbox', 'jetbrains', 'jupyternotebooks', 'macos', 'microsoftoffice', 'python','sublimetext', 'virtualenv', 'visualstudio', 'visualstudiocode']

import sys

from project_manager import ProjectManager as PM

path_to_proj_dir = "path/to_some/folder"
PM.add_git_ignore_to_project(path_to_proj_dir)

sys.exit()

Create a Conda Env, Register Jupyter Kernel, and Init Current Directory as a Poetry Project

Running the following command, creates a Python 3.9 conda env, registers the jupyter kernel, and inits the current directory as a poetry project:

import sys

from project_manager import LocalProjectManager as LPM

env_name = "hello_world"
python_version = "3.9"

LPM.init_current_dir_as_a_poetry_conda_project(env_name, python_version)

sys.exit()

Register Python Interpreters with Sublime Text 3

Registering Any Python Interpreter with Sublime Text 3

If you're interested in registering any Python interpreter with Sublime Text 3, running the following block should help you do that.

import sys

from project_manager import SublimeBuildConfigGenerator as SBCG

path_to_python_bin = r"/opt/homebrew/anaconda3/envs/hello_world/bin"
build_config_name = "hello_world"

SBCG.export_sublime_text_build_config(path_to_python_bin, build_config_name)

sys.exit()

Registering a Conda Python Interpreter with Sublime Text 3

If you want to register a Conda env with Sublime Text 3 without looking up the path to it, then the following block should suffice.

import sys

from project_manager import SublimeBuildConfigGenerator as SBCG

conda_env_name = "hello_world"
SBCG.generate_sublime_text_3_build_config_from_conda_env(conda_env_name)

sys.exit()

Sublime Text 3 - Build Config Output Path

After generating Sublime Text 3 build config settings. You should now be able to see a .sublime-build config in your application data folder

Mac Users

for Mac users, the config settings will be found in:

Users/{YOUR_USERNAME}/Library/Application Support/Sublime Text 3/Packages/User/{build_config_name}.sublime-build

PC Users

for PC users, the config settings will be found in:

C:\Users\{YOUR_USERNAME}\AppData\Roaming\Sublime Text 3\Packages\User\{build_config_name}.sublime-build