The "usability idea" is to leverage user recall of test flags by staying as
close as reasonable to that set. The only real difference is that ft {for
f)ile t)ype} uses h to mean a h)ard link not an alias for -L.
Yes, there are more verbose ways to do this with man 1 find and shell for
loops and more terse ways to do it with Zsh extended globbing (man 1 zshexpn).
I suspect most people are more comfortable with a shell loop that would also be portable, but OTOH, it really is a very small program.
ft [optional-params] [paths: string...]
Batch (in both predicates & targets) `test` / `[` . Emit subset of paths
that pass `expr`. E.g.: `$(ft -eL \*)` =~ Zsh extended glob `\*(@)`. Can
also read stdin as in `find -type f|ft -ew`. (Yes, could cobble together
with GNU `find -files0-from` less tersely & with more quoting concerns.) Maybe
counter-intuitively, exit with status = match count (0=NONE).
-f=, --file= string "" optional input ( `"-"` | !tty = ``stdin`` )
-d=, --delim= char '\n' input file delimiter; `\\0` -> NUL
-t=, --term= char '\n' output path terminator
-p=, --pattern= string "$1" emit a \$1-using pattern; E.g. "match:\$1"
-q, --quiet bool false Do not emit; Just count as exit status
-e=, --expr= string "e" Concatenated extended one-letter test(1) codes
e (e)xists in any way
b is (b)lock special
c is (c)haracter special
d is a (d)irectory
f is a regular (f)ile
l|L is a symbolic (l)ink; NOTE: h differs!
p is a named (p)ipe {aka FIFO}
S is a (S)ocket;CASE differs from ls/find
s has a (s)ize greater than zero
h is a (h)ard link; Link count > 1
N (N)ew; modify time > access time
k has its stic(k)y bit set
u its set-(u)ser-ID bit is set
g is set-(g)roup-ID
O is (O)wned by the effective user ID
G is owned by effective (G)roup ID
r|R|A user|World|Group can (r)ead
w|W|I user|World|Group can (w)rite
x|X|E user|World|Group can e(x)ecute|travers
In all cases a file must exist for 'true'
Codes are logically ANDed; '^' prefix => NOT