Get first line of a shell command's output

问题:

While trying to read the version number of vim, I get a lot additional lines which I need to ignore. I tried to read the manual of head and tried the following command:

vim --version | head -n 1

I want to know if this is the correct approach?

回答1:

Yes, that is one way to get the first line of output from a command.

If the command outputs anything to standard error that you would like to capture in the same manner, you need to redirect the standard error of the command to the standard output stream:

utility 2>&1 | head -n 1

There are many other ways to capture the first line too, including sed 1q (quit after first line), sed -n 1p (only print first line, but read everything), awk 'FNR == 1' (only print first line, but again, read everything) etc.

回答2:

I would use:

awk 'FNR <= 1' file_*.txt

As @Kusalananda points out there are many ways to capture the first line in command line but using the head -n 1 may not be the best option when using wildcards since it will print additional info. Changing 'FNR == i' to 'FNR <= i' allows to obtain the first i lines.

For example, if you have n files named file_1.txt, ... file_n.txt:

awk 'FNR <= 1' file_*.txt

hello
...
bye

But with head wildcards print the name of the file:

head -1 file_*.txt

==> file_1.csv <==
hello
...
==> file_n.csv <==
bye