This is actually superior because, I’m guessing, aa probably won’t work with things like shell aliases, so simply using the shell to run your alert after the command finishes is a better option.
One option would be to just always run the command through the user’s shell. Yours isn’t the only otherwise-cool utility that doesn’t work on aliases. I recently found https://github.com/chneukirchen/nq and was super stoked until I realized it barfed on aliases. I may submit a PR to support that if I find time.
How do you like Rust? I haven’t found enough time or motivation to wrap my tiny brain around it yet :-(
I don’t think alert-after actually does this yet, but one advantage of using a wrapper is that it can check the exit status and automatically give you a success/failure notification. You could do that in shell too, but it’d be much more tedious to write out by hand every time.
A 101-character suffix on a command, versus the OP’s 2-character prefix on a command? Yes, I’d call that “much more tedious”.
(it would be pretty easy to turn your command into a shell-script wrapper with a two-character name, but that’s just an argument about implementation, not about whether a wrapper is a good design)
why not use
$ some –command; osascript -e ‘display notification “Done!!” with title “Done!”’
on your mac? (there are equivalents for gnome, kde, X etc.)
This is actually superior because, I’m guessing,
aaprobably won’t work with things like shell aliases, so simply using the shell to run your alert after the command finishes is a better option.Good point! I forgot about shell aliases. I opened an issue to track adding support for that:
https://github.com/frewsxcv/alert-after/issues/8
One option would be to just always run the command through the user’s shell. Yours isn’t the only otherwise-cool utility that doesn’t work on aliases. I recently found https://github.com/chneukirchen/nq and was super stoked until I realized it barfed on aliases. I may submit a PR to support that if I find time.
How do you like Rust? I haven’t found enough time or motivation to wrap my tiny brain around it yet :-(
I don’t think alert-after actually does this yet, but one advantage of using a wrapper is that it can check the exit status and automatically give you a success/failure notification. You could do that in shell too, but it’d be much more tedious to write out by hand every time.
Not awful.
If you simply want to notify with the exit code you can use
$?which is bound the exit code of the previous process, e.g.A 101-character suffix on a command, versus the OP’s 2-character prefix on a command? Yes, I’d call that “much more tedious”.
(it would be pretty easy to turn your command into a shell-script wrapper with a two-character name, but that’s just an argument about implementation, not about whether a wrapper is a good design)
sure, but it could be a 5 line shell script
shell has functions and aliases that you can add to your profile.
I use zenity on Linux for this exact purpose