[default: ] vs. (default: )

[DimitriPapadopoulos]

I see a mixture of [default: ...] and (default: ...) for default values of user options.

• 24 occurrences of [default: ...], for example:

  distutils/distutils/command/install_lib.py

  Lines 41 to 46 in e5b06e1

  	(
  	'optimize=',
  	'O',
  	"also compile with optimization: -O1 for \"python -O\", "
  	"-O2 for \"python -OO\", and -O0 to disable [default: -O0]",
  	),

• 11 occurrences of (default: ...), for example:

  distutils/distutils/command/install_data.py

  Lines 18 to 23 in e5b06e1

  	(
  	'install-dir=',
  	'd',
  	"base directory for installing data files "
  	"(default: installation base dir)",
  	),

Is it on purpose? If so, what meaning does either version convey? If not, would it be OK to standardise?

[DimitriPapadopoulos]

See pypa/setuptools#4421.

[jaraco]

I'm unaware of any intent behind the difference. Feel free to spelunk the blame history to see if the commit messages provide any insight, or don't and we can proceed with standardizing. I have no preference.

[DimitriPapadopoulos]

The blame history of a few examples does not provide any insight.

I suggest using [default: ]. The closest to a "standard" that defines how default values should be printed is perhaps docopt:

[...] and whether that argument has a default value ([default: 10]).