You can use the command
shell built-in to bypass the normal lookup process and run the given command as an external command regardless of any other possibilities (shell built-ins, aliases, etc.). This is often done in scripts which need to be portable across systems, although probably more commonly using the shorthand \
(as in \rm
rather than command rm
or rm
, as especially the latter may be aliased to something not known like rm -i
).
$ time
real 0m0.000s
user 0m0.000s
sys 0m0.000s
$ command time
Usage: time [-apvV] [-f format] [-o file] [--append] [--verbose]
[--portability] [--format=format] [--output=file] [--version]
[--quiet] [--help] command [arg...]
$
This can be used with an alias, like so:
$ alias time='command time'
$ time
Usage: time [-apvV] [-f format] [-o file] [--append] [--verbose]
[--portability] [--format=format] [--output=file] [--version]
[--quiet] [--help] command [arg...]
$
The advantage of this over e.g. alias time=/usr/bin/time
is that you aren't specifying the full path to the time
binary, but instead falling back to the usual path search mechanism.
The alias
command itself can go into e.g. ~/.bashrc or /etc/bash.bashrc (the latter is global for all users on the system).
For the opposite case (forcing use of the shell built-in in case there's an alias defined), you'd use something like builtin time
, which again overrides the usual search process and runs the named shell built-in. The bash man page mentions that this is often used in order to provide custom cd
functionality with a function named cd
, which in turn uses the builtin cd
to do the real thing.
time
command – Gabriel Staples Apr 21 '22 at 02:15