A collection of utility scripts/functions that I use frequently. Includes helper functions/default variables for libraries like loguru, diskcache, and msgpack.
The utilities are broken down into 2 modules:
std: Utilities with no external dependencies, only the Pythonstdlib.ext: Utilities for dependencies likeloguruandmsgpack-
Note: It is generally a good practice to import
extmodules as a whole, instead of importing functions/variables from the module. -
This is because some of the function names I chose may be common (like
get_ts()in theext.time_utilsmodule). -
Example:
from red_utils.ext import time_utils now = time_utils.get_ts()or, with
pendulum:from red_utils.ext.time_utils import pendulum_utils now = pendulum_utils.get_ts()
-
Common code shared by the std and ext modules can be imported from red_utils.core and red_utils.domain. Any code in these modules should be clean of any external dependency. This is because the std module imports from core, and adding non-stdlib functionality in red_utils.core breaks the philosophy of the stdlib module. I may introduce a red_utils.ext.core at some point.
Some domain objects (dataclass or regular Python classes) may be stored in red_utils.domain. As of release v0.2.12, this module is empty, but future releases may bring some utilities in the form of a class.
Custom/common exceptions are stored in red_utils.exc.
The red-utils package makes use of the Python stdlib pkgutil module to control imports. Packages in the ext module are only imported and available in red_utils if the corresponding dependency exists.
For instance, red_utils.ext.msgpack_utils will only be available if this check in src/red_utils/ext passes:
import pkgutil
...
if pkgutil.find_loader("msgpack"):
from . import msgpack_utils
pkgutil.find_loader() is used throughout the app to control imports and ensure red_utils is stable, by keeping uninstalled module's utilities out of the namespace.
This project uses dependencies groups, meaning it can be installed with pip install red-utils for the base package, or with dependency groups like pip install red-utils[all] (to install all packages with a corresponding red-util module), pip install red-utils[http] (to install some helpful packages for HTTP requests, i.e. httpx and diskcache), and more.
- pip
pip install red-utils
- pdm
pdm add red-utils
Note: I will do my best to update this, but to get an accurate view of available dependency groups and the packages that will be installed, check the pyproject.toml file. Look for the dependency lists, i.e. dependencies = [ (the base set of dependencies), all = [, http = [, etc.
[all]: Install all packages that have a corresponding util. This may be a large install, and is generally not recommended.
[arrow]: By default, the pendulum library is used for time_utils. Installing red_utils[arrow] allows for importing arrow functions from red_utils.ext.time_utils.arrow.
[fastapi]: Dependencies for fastapi_utils
[http]: My standard "HTTP toolkit." Comes with a request client (httpx), logging (loguru), caching (diskcache), & more.
- pip:
pip install red-utils[fastapi,http]
- pdm:
pdm add red-utils[fastapi,http]
Check the Github page to see modules in the ext and std modules (or click one of those words to be taken there).
Please see the developing docs for instructions on setting up a dev environment to work on red-utils.