What's a good example of piping commands together?

I'm going to walk you through a somewhat complex example, based on a real life scenario.

Problem

Let's say the command conky stopped responding on my desktop, and I want to kill it manually. I know a little bit of Unix, so I know that what I need to do is execute the command kill <PID>. In order to retrieve the PID, I can use ps or top or whatever tool my Unix distribution has given me. But how can I do this in one command?

Answer

$ ps aux | grep conky | grep -v grep | awk '{print $2}' | xargs kill

DISCLAIMER: This command only works in certain cases. Don't copy/paste it in your terminal and start using it, it could kill processes unsuspectingly. Rather learn how to build it.

How it works

1- ps aux

This command will output the list of running processes and some info about them. The interesting info is that it'll output the PID of each process in its 2nd column. Here's an extract from the output of the command on my box:

$ ps aux
 rahmu     1925  0.0  0.1 129328  6112 ?        S    11:55   0:06 tint2
 rahmu     1931  0.0  0.3 154992 12108 ?        S    11:55   0:00 volumeicon
 rahmu     1933  0.1  0.2 134716  9460 ?        S    11:55   0:24 parcellite
 rahmu     1940  0.0  0.0  30416  3008 ?        S    11:55   0:10 xcompmgr -cC -t-5 -l-5 -r4.2 -o.55 -D6
 rahmu     1941  0.0  0.2 160336  8928 ?        Ss   11:55   0:00 xfce4-power-manager
 rahmu     1943  0.0  0.0  32792  1964 ?        S    11:55   0:00 /usr/lib/xfconf/xfconfd
 rahmu     1945  0.0  0.0  17584  1292 ?        S    11:55   0:00 /usr/lib/gamin/gam_server
 rahmu     1946  0.0  0.5 203016 19552 ?        S    11:55   0:00 python /usr/bin/system-config-printer-applet
 rahmu     1947  0.0  0.3 171840 12872 ?        S    11:55   0:00 nm-applet --sm-disable
 rahmu     1948  0.2  0.0 276000  3564 ?        Sl   11:55   0:38 conky -q

2- grep conky

I'm only interested in one process, so I use grep to find the entry corresponding to my program conky.

$ ps aux | grep conky
 rahmu     1948  0.2  0.0 276000  3564 ?        Sl   11:55   0:39 conky -q
 rahmu     3233  0.0  0.0   7592   840 pts/1    S+   16:55   0:00 grep conky

3- grep -v grep

As you can see in step 2, the command ps outputs the grep conky process in its list (it's a running process after all). In order to filter it, I can run grep -v grep. The option -v tells grep to match all the lines excluding the ones containing the pattern.

$ ps aux | grep conky | grep -v grep
 rahmu     1948  0.2  0.0 276000  3564 ?        Sl   11:55   0:39 conky -q

NB: I would love to know a way to do steps 2 and 3 in a single grep call.

4- awk '{print $2}'

Now that I have isolated my target process. I want to retrieve its PID. In other words I want to retrieve the 2nd word of the output. Lucky for me, most (all?) modern unices will provide some version of awk, a scripting language that does wonders with tabular data. Our task becomes as easy as print $2.

$ ps aux | grep conky | grep -v grep | awk '{print $2}'
 1948

5- xargs kill

I have the PID. All I need is to pass it to kill. To do this, I will use xargs.

xargs kill will read from the input (in our case from the pipe), form a command consisting of kill <items> (<items> are whatever it read from the input), and then execute the command created. In our case it will execute kill 1948. Mission accomplished.

Final words

Note that depending on what version of unix you're using, certain programs may behave a little differently (for example, ps might output the PID in column $3). If something seems wrong or different, read your vendor's documentation (or better, the man pages). Also be careful as long pipes can be dangerous. Don't make any assumptions especially when using commands like kill or rm. For example, if there was another user named 'conky' (or 'Aconkyous') my command may kill all his running processes too!

What I'm saying is be careful, especially for long pipes. It's always better to build it interactively as we did here, than make assumptions and feel sorry later.

My favourite is this one:

youtube-dl $1 -q -o - | ffmpeg -i - $2

downloads a video from the given youtube url passed by $1 and outputs it as the file given by $2. Note how the file is quietly -q output to STDOUT -o -, piped to ffmpeg and used as input there by -i -.

Especially for linux newbies this might be a pratical example why the command line can be useful and make things easier than using GUI tools. I am not sure how long it would take to download a video from youtube and convert its sound to an mp3. The above line can do that in some seconds.

The general use (read: the way I use it most of the times) is when, for some reason, I have to run some data through several tools to carry on different processing tasks.

So I'd say the use of pipes is as glue to assemble several building blocks (the different UNIX tools) together. Like Ulrich said, sort and uniq is a common stanza.

Depending on the audience, if you want to highlight this usage of pipes, you could, for example, start with: "hey, this syllabus has links to several interesting PDFs with papers and lecture notes, but some of them are repeated. can I somehow automate this?"

Then you could show how lynx --dump --listonly gets the list of links, how grep could filter for links ending in .pdf, how colrm or sed could get rid of the numbers lynx writes left to each URL, how sort and uniq could get rid of duplicates, and finally how wget -i - can be used to retrieve the files (using --wait to be gentle on the server, of course).

I'm afraid this is a complex example. On the other hand, it may help showing the power of pipes when you just pipe it and have the shell run it all at once.