Sometimes we want to use the results of the previous bash command in the next. For example, let’s say we’re clearing out some old text files. We start by listing them:
$ ls *txt info.txt mynotes.txt puppynames.txt techmeeting.txt
Of course this is a contrived example: with only four files listed, the “shortcut” below takes longer than just deleting the files you don’t want. But you knew that, right?
We realise we want to keep the the last file, techmeeting.txt, so we build a command that will just list the files we want to delete:
$ ls *txt|grep -v techmeeting.txt info.txt mynotes.txt puppynames.txt
In reality, we may well make this line more complex so that it lists only the files we want to delete.
What I used to do next was recall the last command, and then edit it like this:
$ rm $(*txt|grep -v techmeeting.txt)
The $(*txt|grep -v techmeeting.txt) construct is replace by the output of the command(s) between the parentheses, and so becomes:
$ rm info.txt mynotes.txt puppynames.txt
That works, but there’s a quicker way. At the shell, !! is replaced by the previous command, so we can just type:
$ rm $(!!)
Could this Linux Tip be improved?
Let us know in the comments below.
In the bash shell (and maybe others) from the command line we can use . to retrieve the last command (as well as !!)
Aargh – there was supposed to be {escape} in front of the dot, but I used angle brackets to display it and it got eaten 🙁
Thanks Robert – I didn’t know about ESC-. – although I actually think !! is easier to type, and at least it doesn’t get eaten!