On 2020-01-02, Laslo Hunhold <d...@frign.de> wrote: > I would print something on stderr. POSIX is ignored often enough and a > ton of scripts are using the cancerous GNU extensions and other > extensions. If we just "ignore" them, there is no learning effect or > push for change for script writers, so maybe we could add a warning > while we ignore them, so when you run a script that makes use of these > mostly useless flags (which we could also tell them), then this might a > push in a good direction. What do you think?
I'm well aware that POSIX is ignored often. I send patches to projects every time I run into a script using non-POSIX options that break compatibility with sbase. But in this case, POSIX is not really relevant since as I mentioned, these are not POSIX tools, and there aren't many different implementations (BSDs have their own md5(1), sha256(1), ...), so coreutils is essentially the standard. Who are we to decide that -b and -t are not be valid flags for sha*sum and should not be used? I ran into a script that used `sha256sum -b`, but how could I justify removing that option to upstream? The flags may be useless on operating systems that sbase supports, but the "rb" is a valid mode for C99 fopen ("r" opens as a text file), so there must be operating systems where it makes a difference. Removing -b may break the script on those operating systems.