cltoolbox is a wrapper around argparse
, and allows you to write complete CLI
applications in seconds using a simple decorator.
Significant portions of this code are based on mando. The primary differences between mando and cltoolbox are:
- cltoolbox supports automatic detection of Sphinx, Google, and Numpy docstring formats by using the docstring_parser library
- cltoolbox supports python 3.7+
If you need support of the mando formatted docstring or python 2 you have to use mando instead of cltoolbox.
pip install cltoolbox
conda install -c conda-forge cltoolbox
The argparse
module that comes with Python requires a programmer to
duplicate information that Python can easily parse from the function signature
and docstring. The cltoolbox
does this for you by using decorators.
from cltoolbox import command, main
@command
def echo(text, capitalize=False):
"""Echo the given text."""
if capitalize:
text = text.upper()
print(text)
if __name__ == "__main__":
main()
Generated help:
$ python example.py -h
usage: example.py [-h] {echo} ...
positional arguments:
{echo}
echo Echo the given text.
optional arguments:
-h, --help show this help message and exit
$ python example.py echo -h
usage: example.py echo [-h] [--capitalize] text
Echo the given text.
positional arguments:
text
optional arguments:
-h, --help show this help message and exit
--capitalize
Actual usage:
$ python example.py echo spam
spam
$ python example.py echo --capitalize spam
SPAM
Something more complex and real-world-ish. The code:
from cltoolbox import command, main
@command
def push(repository, all=False, dry_run=False, force=False, thin=False):
"""Update remote refs along with associated objects.
:param repository: Repository to push to.
:param --all: Push all refs.
:param -n, --dry-run: Dry run.
:param -f, --force: Force updates.
:param --thin: Use thin pack."""
print(
"Pushing to {0}. All: {1}, dry run: {2}, force: {3}, thin: {4}".format(
repository, all, dry_run, force, thin
)
)
if __name__ == "__main__":
main()
cltoolbox understands Sphinx, Google, and Numpy dostrings, from which it can create short options and their help for you.
$ python git.py push -h
usage: git.py push [-h] [--all] [-n] [-f] [--thin] repository
Update remote refs along with associated objects.
positional arguments:
repository Repository to push to.
optional arguments:
-h, --help show this help message and exit
--all Push all refs.
-n, --dry-run Dry run.
-f, --force Force updates.
--thin Use thin pack.
Let's try it!
$ python git.py push --all myrepo
Pushing to myrepo. All: True, dry run: False, force: False, thin: False
$ python git.py push --all -f myrepo
Pushing to myrepo. All: True, dry run: False, force: True, thin: False
$ python git.py push --all -fn myrepo
Pushing to myrepo. All: True, dry run: True, force: True, thin: False
$ python git.py push --thin -fn myrepo
Pushing to myrepo. All: False, dry run: True, force: True, thin: True
$ python git.py push --thin
usage: git.py push [-h] [--all] [-n] [-f] [--thin] repository
git.py push: error: too few arguments
Amazed uh? Yes, cltoolbox got the short options and the help from the docstring!
You can put much more in the docstring, and if that isn't enough, there's an
@arg
decorator to customize the arguments that get passed to argparse.
cltoolbox understands Python 3-style type annotations and will warn the user if the arguments given to a command are of the wrong type.
from cltoolbox import command, main
@command
def duplicate(string, times: int):
"""Duplicate text.
:param string: The text to duplicate.
:param times: How many times to duplicate."""
print(string * times)
if __name__ == "__main__":
main()
$ python3 test.py duplicate "test " 5
test test test test test
$ python3 test.py duplicate "test " foo
usage: test.py duplicate [-h] string times
test.py duplicate: error: argument times: invalid int value: 'foo'
The cltoolbox supports shell autocompletion via the
argcomplete
package and supports custom format classes. For a complete
documentation, visit https://timcera.bibucket.io/cltoolbox/.