ip
💥 uhhhggghhhawk
💥 uhhhggghhhsed
💥 uhhhggghhhgrep
💥 uhhhggghhhjournalctl
💥 uhhhggghhhbrew install docker
? Nope! 💥 uhhhggghhh
Having to maintain TWO build chains (when the product only actually runs on Linux) and deal with all the extra tooling and complexity 💥💥💥💥💥💥 uhhhggghhh
my workaround for awk, sed, and grep has been to create symlinks in my $Path that point to gawk, gsed, and ggrep, which you can install through brew. Nix has also been useful to keep my sanity intact, tho it’s a learning curve too, and when Macos updates happen you will need to fix/reinstall it. Can I ask what terminal you are using?
The problem with this is that scripts assume the BSD versions of those tools. The scripts break if I do that. (I tried doing that.) 😢
The parts of the code base that can be built on both platforms is littered with:
if os == linux awk <something> else if os == mac awk <something else>
Why not set an
$awk
var once for the right platform and then use that instead of if-on-every-call? This seems incredibly error prone and verbose for no reason…Edit: To be clear I’m aware that it isn’t about the binary, but surely most of the calls have some baseline commonality that could be abstracted? Or: Can the Linux awk be built on bsd or vice versa? Ok fine, if it was the simple, it would’ve probably been the solution…