GitHub | PyPI | Issues | Changelog
click-loglevel
provides a LogLevel
parameter type for use in Click
programs that wish to let the user set the logging level. It accepts all of
the logging
log level names (CRITICAL
, ERROR
, WARNING
,
INFO
, DEBUG
, and NOTSET
, all case insensitive), and converts them
into their corresponding numeric values. It also accepts integer values and
leaves them as-is. Custom log levels are also supported.
Starting in version 0.4.0, shell completion of log level names (both built-in and custom) is also supported.
click-loglevel
requires Python 3.8 or higher. Just use pip for Python 3 (You have pip, right?) to install it:
python3 -m pip install click-loglevel
myscript.py
:
import logging
import click
from click_loglevel import LogLevel
@click.command()
@click.option(
"-l",
"--log-level",
type=LogLevel(),
default="INFO",
help="Set logging level",
show_default=True,
)
def main(log_level: int) -> None:
logging.basicConfig(
format="[%(levelname)-8s] %(message)s",
level=log_level,
)
logging.log(log_level, "Log level set to %r", log_level)
if __name__ == "__main__":
main()
Running myscript.py
:
$ python3 myscript.py
[INFO ] Log level set to 20
$ python3 myscript.py --log-level DEBUG
[DEBUG ] Log level set to 10
$ python3 myscript.py --log-level error
[ERROR ] Log level set to 40
$ python3 myscript.py --log-level 15
[Level 15] Log level set to 15
Script with custom log levels:
import logging
import click
from click_loglevel import LogLevel
logging.addLevelName(15, "VERBOSE")
logging.addLevelName(25, "NOTICE")
@click.command()
@click.option(
"-l",
"--log-level",
type=LogLevel(extra=["VERBOSE", "NOTICE"]),
default="INFO",
help="Set logging level",
show_default=True,
)
def main(log_level: int) -> None:
logging.basicConfig(
format="[%(levelname)-8s] %(message)s",
level=log_level,
)
logging.log(log_level, "Log level set to %r", log_level)
if __name__ == "__main__":
main()
The click_loglevel
module contains a single class:
A subclass of click.ParamType
that accepts the standard logging level names
(case insensitive) and converts them to their corresponding numeric values. It
also accepts integer values and leaves them as-is.
Custom log levels can be added by passing them as the extra
argument to the
constructor. extra
can be either an iterable of level names (in which case
the levels must have already been defined — typically at the module level — by
calling logging.addLevelName()
) or a mapping from level names to their
corresponding values. All custom log levels will be recognized case
insensitively; if two different level names differ only in case, the result is
undefined.