Dave Close wrote on Mon, Aug 22, 2022 at 03:01:11PM -0700: > $ grep CALL ./* | wc > 16325 61444 950870 > $ grep CALL $( ls ./* ) | wc > 16375 61601 953451 > > So I almost have a solution. Except the extra 50 lines found in the > second command are disturbing. Thus far, it appears that those lines > are duplicates and some of the duplicated lines are from files with > names that do not contain any special characters.
That is because this is not correct either. Just don't use ls for thing, `find -print0` is just about the only thing that can handle arbitrary filenames (well, direct shell expansions also mostly work with care; but I'll assume your example is oversimplified otherwise you wouldn't be trying to use ls... right?) It's probably easier to understand with a concrete example: $ echo a > one $ echo a > one\ one $ grep a $(ls ./*) ./one:a ./one:a one:a See here how the arguments got split as the space, so the 'one' file got listed three times instead of handling both files once. As for quoting: $ grep a $(ls --quoting-style=shell-escape-always ./*) grep: './one': No such file or directory grep: './one: No such file or directory grep: one': No such file or directory As you can see here the quotes from shell escaping are treated as literals, you'd get something that appears to be what you want with an extra eval: $ eval grep a $(ls --quoting-style=shell-escape-always ./*) ./one:a ./one one:a but it's hardly something I would recommend. Anyway, here's an example for find -print0: $ find . -type f -print0 | xargs -0 grep a -- ./one one:a ./one:a -- Dominique
