Skip to main content

Another XDG Base Directory Specification utility.

Project description


pipeline Build status codecov pypi

xdgenvpy is yet another Python utility for the XDG Base Directory Specification, but one that provides Pythonic access as well as a CLI utility. xdgenvpy adheres to the XDG Base Directory spec on Unix systems, and also provides similar for Windows-based systems.

How to use


There are three main ways to use xdgenvpy as a Python package,

  1. Retrieve XDG environment variables, or the specification defaults.
  2. Determine package specific directories based on the XDG spec.
  3. Or pedantically create package specific directories before attempting to use the directory.

To use xdgenvpy as a simple XDG base directory getter, simply create a new xdgenvpy.XDG object and use the properties it exposes.

from xdgenvpy import XDG
xdg = XDG()
print(xdg.XDG_DATA_HOME)        # /home/user/.local/share
print(xdg.XDG_CONFIG_HOME)      # /home/user/.config
print(xdg.XDG_CACHE_HOME)       # /home/user/.cache
print(xdg.XDG_RUNTIME_DIR)      # /run/user/1000
print(xdg.XDG_DATA_DIRS)        # /home/user/.local/share:/usr/local/share/:/usr/share/
print(xdg.XDG_CONFIG_DIRS)      # /home/user/.config:/etc/xdg

But sometimes you want to use package specific directories derived from the XDG base directories. This can be done with the xdgenvpy.XDGPackage class.

from xdgenvpy import XDGPackage
xdg = XDGPackage('mypackage')
print(xdg.XDG_DATA_HOME)        # /home/user/.local/share/mypackage
print(xdg.XDG_CONFIG_HOME)      # /home/user/.config/mypackage
print(xdg.XDG_CACHE_HOME)       # /home/user/.cache/mypackage
print(xdg.XDG_RUNTIME_DIR)      # /run/user/1000/mypackage
print(xdg.XDG_DATA_DIRS)        # /home/user/.local/share/mypackage:/usr/local/share/:/usr/share/
print(xdg.XDG_CONFIG_DIRS)      # /home/user/.config/mypackage:/etc/xdg')

Lastly, you could also use xdgenvpy.XDGPedanticPackage to ensure each of the package specific directories exist before the calling code attempts to use the directory. Instances of the xdgenvpy.XDGPedanticPackage class will not create system level directories, only package directories on the DATA, CONFIG, CACHE, and RUNTIME variables.

from xdgenvpy import XDGPedanticPackage
xdg = XDGPedanticPackage('mypackage')
print(xdg.XDG_DATA_HOME)        # /home/user/.local/share/mypackage
print(xdg.XDG_CONFIG_HOME)      # /home/user/.config/mypackage
print(xdg.XDG_CACHE_HOME)       # /home/user/.cache/mypackage
print(xdg.XDG_RUNTIME_DIR)      # /run/user/1000/mypackage
print(xdg.XDG_DATA_DIRS)        # /home/user/.local/share/mypackage:/usr/local/share/:/usr/share/
print(xdg.XDG_CONFIG_DIRS)      # /home/user/.config/mypackage:/etc/xdg


xdgenvpy also includes a runnable module, which is easily accessible via the script xdg-env. Pip will normally install scripts under something like: ~/.local/bin

The installed xdg-env command essentially takes a list of XDG variables, and an optional package name. For each XDG variable specified, xdg-env will print its corresponding value based on the specification. It can optionally take the name of a package and include that into the variable's values.

But can't we just echo the XDG variables like so?


Well, yes. But there is a problem when the variables are not defined. The xdg-env command will always print a value that adheres to the spec. If the environment variable does not exist, then the default value will be returned, as defined by the XDG Base Directory Specification.

Although the Python package supports a pedantic mode, the xdg-env command will not change the file system. Even if a package name is supplied and the directories do not exist, xdg-env will not create any files/directories. This was simply a design decision to keep the shell command as file-system safe as possible.

How to install

Install locally as a normal user:

pip3 install --user xdgenvpy

Or install globally as the all powerful root:

sudo pip3 install xdgenvpy

A Word About Windows

The XDG Base Directory Specification. does not mention how the spec should be implemented on Windows-based platforms. That said, many applications on Windows still follow a very similar convention to the XDG base directory spec. And that is to generally place config files under %APPDATA%/MyPackage/configs.

If we squint, it kind of looks like we can simply replace the POSIX tilde ~ with the Windows %APPDATA% variable. Then there's a directory that is the application's name. And finally, any configs or data files the application needs to save is under that directory.

Generally, xdgenvpy works in this way on Windows-based platforms. Though this is not a perfect solution as Windows applications can put configs and data files under any of the directories Local, LocalLow, and Roaming (where the Roaming directory typically is pointed to by %APPDATA%). Additionally, some XDG variables do not make much sense on Windows-based platforms. XDG_RUNTIME_DIR is one such example. On Unix systems it defaults to /run/user/USERID. There is no close equivalent to a Windows-based directory. As such, xdgenvpy does not do anything fancy other than prepend %APPDATA% to most directories and drop any . prefixes for hidden directories/files.

That said, if you use xdgenvpy extensively on Windows platforms and would like better support, create GitLab issues on the project or submit Merge Requests. Let's all make xdgenvpy as useful as possible, even if it needs to implement XDG base directory spec-like features on non-Unix platforms.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for xdgenvpy, version 2.3.4
Filename, size File type Python version Upload date Hashes
Filename, size xdgenvpy-2.3.4-py3-none-any.whl (13.0 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size xdgenvpy-2.3.4.tar.gz (15.9 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page