walk

Walk is a lightweight alternative to the find(1) command, it's main purpose is to walk directories recursively and print the name of each line on a separate line. It is very much inspired by the 9front command walk(1).

Usage: walk [ -dfx ] [ -n min,max ] [ -e "fmt" ] [ name ... ] [ ! cmd ]
  !	Run cmd in a sub shell with sh(1) for each match.
  	If an unescaped % occurs in the command list it will
  	be replaced with the file name.
  -d	Print only directories.
  -e string
    	Specifies the output format.
    	The attributes are automatically separated with a space.
    	The following characters are accepted:
    	U	Owner name
    	G	Group name
    	M	name of the last user to modify the file
    	a	last access time
    	m	last modification time
    	n	final path element (name)
    	p	path
    	s	size (bytes)
    	x	permissions (default "p")
  -f	Print only non-directories.
  -n string
    	Sets the inclusive range for depth filtering.
    	The expected format is "min,max" and both are optional.
    	An argument of n with no comma is equivalent to 0,n.
  -x	Print only if the executable bit is set.

Installation

go build

Examples

This example is straight from 9fronts man page and is adjusted to work with this version of walk. It will walk the ~/bin/ directory, list its files sorted by the modification date:

walk -f -e mp ~/bin | sort -n | sed -E 's/^[^ ]+ //'

Here is another example where walk will only print files or directories that are deeper than 2 levels or more:

walk -n 2, ~/bin

License

MIT