On Sat, Jul 8, 2023 at 1:58 PM Pádraig Brady <p...@draigbrady.com> wrote: > On 08/07/2023 19:52, Jim Meyering wrote: > > On Sat, Jul 8, 2023 at 7:28 AM Pádraig Brady <p...@draigbrady.com> wrote: > >> It would be nice to be able to support users who want to replace all > >> the *sum utilities with a script that did: > >> > >> exec cksum -a <algo> --untagged "$@" > >> > >> To that end, it would be good to support all the options of the older > >> utils. > >> This is a bit awkward with --base64 added in the last few months, as that > >> also added the short -b option. However it's only a few months released, > >> so it's probably not too onerous to change. > >> > >> The attached makes the change, which I'll augment with NEWS and tests/ > >> upon agreement. > > > > Good idea. I agree that the added value outweighs the cost of > > divergence with OpenBSD cksum's -b (https://man.openbsd.org/cksum.1) > > Oh right. Though all OpenBSD's cksum options are not supported > so we don't have to worry too much about direct compat here. > > > One nit: the in-comment reference to -b needs to say --base64, too (in > > tests/cksum/cksum-base64.pl): > > > > # Ensure that each of the above works with -b: > > (map {my ($h,$v)= @$_; my $o=fmt $h,$v; > > - [$h, "-ba $h", {IN=>{f=>''}}, {OUT=>"$o\n"}]} @pairs), > > + [$h, "--base64 -a $h", {IN=>{f=>''}}, {OUT=>"$o\n"}]} @pairs), > > Latest patch attached.
Nice. I like seeing so many ifdefs being deleted. One nit is in NEWS. I'd change this: + 'cksum -b' will no longer print base64-encoded checksums. Rather that to this: + 'cksum -b' no longer prints base64-encoded checksums. Rather that oh, and also simplify this: + printf -- '%s' "$pmode" | grep '<mode>' > /dev/null && pmode="$tmode" to this: + case $pmode in *'<mode>') pmode=$tmode;; esac If you were to use e.g., MODE as the marker instead of <mode>, you could drop the single quotes, too. Thanks!
