04-loop.md

layout	title	subtitle	minutes
page	The Unix Shell	Loops	15

Learning Objectives {.objectives}

• Write a loop that applies one or more commands separately to each file in a set of files. (Create)
• Trace the values taken on by a loop variable during execution of the loop. (Analyze)
• Explain the difference between a variable's name and its value. (Understand)
• Explain why spaces and some punctuation characters shouldn't be used in file names. (Understand)
• Demonstrate how to see what commands have recently been executed. (Apply)
• Re-run recently executed commands without retyping them. (Apply)

Loops are key to productivity improvements through automation as they allow us to execute commands repetitively. Similar to wildcards and tab completion, using loops also reduces the amount of typing (and typing mistakes). Suppose we have several hundred genome data files named basilisk.dat, unicorn.dat, and so on. In this example, we'll use the creatures directory which only has two example files, but the principles can be applied to many many more files at once. We would like to modify these files, but also save a version of the original files, naming the copies original-basilisk.dat and original-unicorn.dat. We can't use:

$ cp *.dat original-*.dat

because that would expand to:

$ cp basilisk.dat unicorn.dat original-*.dat

This wouldn't back up our files, instead we get an error:

cp: target `original-*.dat' is not a directory

This problem arises when cp receives more than two inputs. When this happens, it expects the last input to be a directory where it can copy all the files it was passed. Since there is no directory named original-*.dat in the creatures directory we get an error.

Instead, we can use a loop to do some operation once for each thing in a list. Here's a simple example that displays the first three lines of each file in turn:

$ for filename in basilisk.dat unicorn.dat
> do
>    head -n 3 $filename
> done

COMMON NAME: basilisk
CLASSIFICATION: basiliscus vulgaris
UPDATED: 1745-05-02
COMMON NAME: unicorn
CLASSIFICATION: equus monoceros
UPDATED: 1738-11-24

When the shell sees the keyword for, it knows it is supposed to repeat a command (or group of commands) once for each thing in a list. In this case, the list is the two filenames. Each time through the loop, the name of the thing currently being operated on is assigned to the variable called filename. Inside the loop, we get the variable's value by putting $ in front of it: $filename is basilisk.dat the first time through the loop, unicorn.dat the second, and so on.

By using the dollar sign we are telling the shell interpreter to treat filename as a variable name and substitute its value on its place, but not as some text or external command. When using variables it is also possible to put the names into curly braces to clearly delimit the variable name: $filename is equivalent to ${filename}, but is different from ${file}name. You may find this notation in other people's programs.

Finally, the command that's actually being run is our old friend head, so this loop prints out the first three lines of each data file in turn.

Follow the Prompt {.callout}

The shell prompt changes from $ to > and back again as we were typing in our loop. The second prompt, >, is different to remind us that we haven't finished typing a complete command yet. A semicolon, ;, can be used to separate two commands written on a single line.

Same symbols, different meanings {.callout}

Here we see > being used a shell prompt, whereas > is also used to redirect output. Similarly, $ is used as a shell prompt, but, as we saw earler, it is also used to ask the shell to get the value of a variable.

If the shell prints > or $ then it expects you to type something, and the symbol is a prompt.

If you type > or $ yourself, it is an instruction from you that the shell to redirect output or get the value of a variable.

We have called the variable in this loop filename in order to make its purpose clearer to human readers. The shell itself doesn't care what the variable is called; if we wrote this loop as:

for x in basilisk.dat unicorn.dat
do
    head -n 3 $x
done

or:

for temperature in basilisk.dat unicorn.dat
do
    head -n 3 $temperature
done

it would work exactly the same way. Don't do this. Programs are only useful if people can understand them, so meaningless names (like x) or misleading names (like temperature) increase the odds that the program won't do what its readers think it does.

Here's a slightly more complicated loop:

for filename in *.dat
do
    echo $filename
    head -n 100 $filename | tail -n 20
done

The shell starts by expanding *.dat to create the list of files it will process. The loop body then executes two commands for each of those files. The first, echo, just prints its command-line parameters to standard output. For example:

$ echo hello there

prints:

hello there

In this case, since the shell expands $filename to be the name of a file, echo $filename just prints the name of the file. Note that we can't write this as:

for filename in *.dat
do
    $filename
    head -n 100 $filename | tail -n 20
done

because then the first time through the loop, when $filename expanded to basilisk.dat, the shell would try to run basilisk.dat as a program. Finally, the head and tail combination selects lines 81-100 from whatever file is being processed.

Spaces in Names {.callout}

Filename expansion in loops is another reason you should not use spaces in filenames. Suppose our data files are named:

basilisk.dat
red dragon.dat
unicorn.dat

If we try to process them using:

for filename in *.dat
do
    head -n 100 $filename | tail -n 20
done

then the shell will expand *.dat to create:

basilisk.dat red dragon.dat unicorn.dat

With older versions of Bash, or most other shells, filename will then be assigned the following values in turn:

basilisk.dat
red
dragon.dat
unicorn.dat

That's a problem: head can't read files called red and dragon.dat because they don't exist, and won't be asked to read the file red dragon.dat.

We can make our script a little bit more robust by quoting our use of the variable:

for filename in *.dat
do
    head -n 100 "$filename" | tail -n 20
done

but it's simpler just to avoid using spaces (or other special characters) in filenames.

Going back to our original file copying problem, we can solve it using this loop:

for filename in *.dat
do
    cp $filename original-$filename
done

This loop runs the cp command once for each filename. The first time, when $filename expands to basilisk.dat, the shell executes:

cp basilisk.dat original-basilisk.dat

The second time, the command is:

cp unicorn.dat original-unicorn.dat

Nelle's Pipeline: Processing Files

Nelle is now ready to process her data files. Since she's still learning how to use the shell, she decides to build up the required commands in stages. Her first step is to make sure that she can select the right files --- remember, these are ones whose names end in 'A' or 'B', rather than 'Z'. Starting from her home directory, Nelle types:

$ cd north-pacific-gyre/2012-07-03
$ for datafile in *[AB].txt
> do
>     echo $datafile
> done

NENE01729A.txt
NENE01729B.txt
NENE01736A.txt
...
NENE02043A.txt
NENE02043B.txt

Her next step is to decide what to call the files that the goostats analysis program will create. Prefixing each input file's name with "stats" seems simple, so she modifies her loop to do that:

$ for datafile in *[AB].txt
> do
>     echo $datafile stats-$datafile
> done

NENE01729A.txt stats-NENE01729A.txt
NENE01729B.txt stats-NENE01729B.txt
NENE01736A.txt stats-NENE01736A.txt
...
NENE02043A.txt stats-NENE02043A.txt
NENE02043B.txt stats-NENE02043B.txt

She hasn't actually run goostats yet, but now she's sure she can select the right files and generate the right output filenames.

Typing in commands over and over again is becoming tedious, though, and Nelle is worried about making mistakes, so instead of re-entering her loop, she presses the up arrow. In response, the shell redisplays the whole loop on one line (using semi-colons to separate the pieces):

$ for datafile in *[AB].txt; do echo $datafile stats-$datafile; done

Using the left arrow key, Nelle backs up and changes the command echo to bash goostats:

$ for datafile in *[AB].txt; do bash goostats $datafile stats-$datafile; done

When she presses Enter, the shell runs the modified command. However, nothing appears to happen --- there is no output. After a moment, Nelle realizes that since her script doesn't print anything to the screen any longer, she has no idea whether it is running, much less how quickly. She kills the running command by typing Ctrl-C, uses up-arrow to repeat the command, and edits it to read:

$ for datafile in *[AB].txt; do echo $datafile; bash goostats $datafile stats-$datafile; done

Beginning and End {.callout}

We can move to the beginning of a line in the shell by typing Ctrl-A and to the end using Ctrl-E.

When she runs her program now, it produces one line of output every five seconds or so:

NENE01729A.txt
NENE01729B.txt
NENE01736A.txt
...

1518 times 5 seconds, divided by 60, tells her that her script will take about two hours to run. As a final check, she opens another terminal window, goes into north-pacific-gyre/2012-07-03, and uses cat stats-NENE01729B.txt to examine one of the output files. It looks good, so she decides to get some coffee and catch up on her reading.

Those Who Know History Can Choose to Repeat It {.callout}

Another way to repeat previous work is to use the history command to get a list of the last few hundred commands that have been executed, and then to use !123 (where "123" is replaced by the command number) to repeat one of those commands. For example, if Nelle types this:

$ history | tail -n 5
  456  ls -l NENE0*.txt
  457  rm stats-NENE01729B.txt.txt
  458  bash goostats NENE01729B.txt stats-NENE01729B.txt
  459  ls -l NENE0*.txt
  460  history

then she can re-run goostats on NENE01729B.txt simply by typing !458.

Other history commands {.callout}

There are a number of other shortcut commands for getting at the history. Two of the more useful are !!, which retrieves the immediately preceding command (you may or may not find this more convenient than plain up-arrow), and !$, which retrieves the last word of the last command. That's useful more often than you might expect: after bash goostats NENE01729B.txt stats-NENE01729B.txt, you can type less !$ to look at the file stats-NENE01729B.txt, which is quicker than doing up-arrow and editing the command-line.

Variables in Loops {.challenge}

Suppose that ls initially displays:

fructose.dat    glucose.dat   sucrose.dat

What is the output of:

for datafile in *.dat
do
    ls *.dat
done

Now, what is the output of:

for datafile in *.dat
do
  ls $datafile
done

Why do these two loops give you different outputs?

Saving to a File in a Loop - Part One {.challenge}

In the same directory, what is the effect of this loop?

for sugar in *.dat
do
    echo $sugar
    cat $sugar > xylose.dat
done

1. Prints fructose.dat, glucose.dat, and sucrose.dat, and the text from sucrose.dat will be saved to a file called xylose.dat.
2. Prints fructose.dat, glucose.dat, and sucrose.dat, and the text from all three files would be concatenated and saved to a file called xylose.dat.
3. Prints fructose.dat, glucose.dat, sucrose.dat, and xylose.dat, and the text from sucrose.dat will be saved to a file called xylose.dat.
4. None of the above.

Saving to a File in a Loop - Part Two {.challenge}

In another directory, where ls returns:

fructose.dat    glucose.dat   sucrose.dat   maltose.txt

What would be the output of the following loop?

for datafile in *.dat
do
    cat $datafile >> sugar.dat
done

1. All of the text from fructose.dat, glucose.dat and sucrose.dat would be concatenated and saved to a file called sugar.dat.
2. The text from sucrose.dat will be saved to a file called sugar.dat.
3. All of the text from fructose.dat, glucose.dat, sucrose.dat and maltose.txt would be concatenated and saved to a file called sugar.dat.
4. All of the text from fructose.dat, glucose.dat and sucrose.dat would be printed to the screen and saved to a file called sugar.dat

Limiting Sets of Files {.challenge}

In the same directory, where ls returns (without the sugar.dat file):

fructose.dat    glucose.dat   sucrose.dat   maltose.txt

What would be the output of the following loop?

for filename in s*
do
    ls $filename 
done

1. No files are listed.
2. All files are listed.
3. Only fructose.dat, glucose.dat and maltose.txt are listed.
4. Only sucrose.dat is listed.

How would the output differ from using this command instead?

for filename in *s*
do
    ls $filename 
done

1. The same files would be listed.
2. All the files are listed this time.
3. No files are listed this time.
4. The file sucrose.dat will be listed twice, with the other files listed once each.

Doing a Dry Run {.challenge}

A loop is a way to do many things at once --- or to make many mistakes at once if it does the wrong thing. One way to check what a loop would do is to echo the commands it would run instead of actually running them.

Suppose we want to preview the commands the following loop will execute without actually running those commands:

for file in *.dat
do
  analyze $file > analyzed-$file
done

What is the difference between the two loops below, and which one would we want to run?

# Version 1
for file in *.dat
do
  echo analyze $file > analyzed-$file
done

# Version 2
for file in *.dat
do
  echo "analyze $file > analyzed-$file"
done

Nested Loops {.challenge}

Suppose we want to set up up a directory structure to organize some experiments measuring the growth rate under different sugar types and different temperatures. What would be the result of the following code:

for sugar in fructose glucose sucrose
do
    for temperature in 25 30 37 40
    do
        mkdir $sugar-$temperature
    done
done