Learning Objectives

Learners will be able to edit documents using the following editors:

- sed
- awk
- printf
- nano
- vi(m)
Stream Editor (sed)
The name sed is short for stream editor. The sed command is most often
used for finding and replacing, or searching and deleting. The sed utility in
its simplest form works like grep, you can use regular expressions to find a
string in a file. It’s much more powerful though as you can provide a
substitute string for the string you found, or delete every line that matches
a string.

First take a look at the man pages for the sed command.

man sed

Using sed to find a string in a file

Try the following command:

sed -n '/Pooh/p' myfile.txt

1. The -n means it will only produce output when explicitly told to via the
p command (the default is to print each line)
2. The '/Pooh/p' tells it to print out lines that have the word “Pooh” in
them.

You can also use sed to find and replace something in a file:

1. The s means substitute.

2. The % is the delimiter (we can use any characters here).
3. The search pattern follows the first %.
4. The replace pattern follows the second %.
5. The g stands for global replace - replace all occurrences in the file.

For example, you can use sed to delete underscores in a file. Take a look at
myfile.txt There are underscores around some words that probably have to
do with the formatted version of this material. To get rid of them we create
a pattern that looks for an underscore and replaces it with nothing.

Try the following command:

 sed 's%_%%g' myfile.txt

Try these variations:

Change the name of one of the characters in the text to something else.

Solution

The solution is similar to the previous one, you just need something in
the “replace” pattern slot

sed 's%Christopher%Chris%g' myfile.txt

Make a change and send the output to a file instead of the console.

Solution

Use the redirection operator &gt to accomplish this.

sed 's%_%%g' myfile.txt > newfile.txt

You can then view the contents of newfile.txt:

markdown cat newfile.txt

Make a change and save it in the original file. Hint: there is a command
line option for this.

Solution

The -i command line option means edit “in place”.

sed -i 's%_%%g' myfile.txt

Using sed to delete lines

There are numerous blank lines at the end of myfile.txt. Use the following
command to delete them:
 sed '45,54d' myfile.txt

1. In '45,54' the comma indicates a range.

2. The d lets sed know it will be a deletion

Try these variations:

Delete all blank lines. Hint: you need a pattern that will match a blank
line.

Solution

The ^ indicates the beginning of the line and the $ indicates the end of
the line and since there is nothing between the two, it means a blank
line. This is deleting all the blank lines, not just the ones at the end so
not the perfect solution for this case.

sed '/^$/d' myfile.txt

From a specific line to the end of the file.

Solution

The last line is the string “So I tried”, starting with this search pattern it
will delete everything until the end of the file, unfortunately including
this line. In this case, it isn’t exactly what we want.

sed '/So I tried/,$d' myfile.txt

Delete everything after a specific line and save it “in place”. Hint: you
want to not match the specific line.

Solution

The -i command line option means edit “in place”. Finally, the right
way to do this! The ’1,/So I tried/ specifies a range from the first line
up to and including “So I tried”. The !d means things is this range will
not be deleted but everything else will.
 sed -i '1,/So I tried/!d' myfile.txt

For more information, reference the documentation for the sed command.
awk
The name awk is a combination of the first letters of the last names of the
people who created the language - Alfred Aho, Peter J. Weinberger, and
Brian Kernighan.

The awk command provides a way to search for a pattern, and perform
actions on the found text. awk reads lines from the input file one at a time.
The line is scanned for each pattern in the program and if there is a match
the associated action is executed.

Either the pattern or the action can be omitted, but not both.
* If the pattern is omitted, then the action is performed for every line.
* If the action is omitted, then all lines that match the pattern will be
printed out.

Take a look at the man pages for the awk command:

man awk

As you will likely notice, awk is very powerful and you can write complete
programs to perform complex tasks using it.

A few built-in AWK variables (not a complete list)

Variable Description
$0 The entire line - not including the newline at the end
$1..$n The fields in a line (delimited by the field separator)
FNR Current line number - just spans the current file
FS Field separator - default is space
NF Number of fields
NR Current line number - spans multiple files
RS Record separator - the default is newline

AWK special patterns

Rules Description
BEGIN Startup actions
END Cleanup actions
Using awk to count the number of occurrences
of the word “Pooh” in the text

The awk command is followed by the pattern (such as /Pooh/)

and action (such as {x++}END{print x}) followed by the file name
(such as myfile.txt)

awk '/Pooh/{x++}END{print x}' myfile.txt

Try these variations:

Print the number of words in a file.

Solution

Add up the number of “fields” on each line. The default field separator
is space so this counts words and prints the total when it gets to the end
of the line.

awk '{ total = total + NF }; END {print total}' myfile.txt

Print the number of lines in a file.

Solution

The AWK variable NR contains the number of lines in a file so you just
have to print it at the end.

awk 'END{print NR}' myfile.txt

What happens if you don’t include the END?

Using awk to pull out items in field separated

data
The awk command line option -F sets the field separator, the default is
space.

Parse the comma-separated text and output two fields from each line. Don’t
print the header line.

awk -F"," '{if (NR!=1){ print $1 " wrong answers " $5 " out of
total " $8 }}' data.csv

Parsing comma separated text using the -F option and specifying

which column using the $ (such as $5)

Try these variations:

Print out just the second column of the file.

Solution

You don’t need the if statement because you’re printing the entire
column including the title line.

awk -F"," '{print $2}' data.csv

Print out the number of answered assessments for “Jane Smith”.

Solution

You use the same syntax you would use for any regular expression.

awk -F"," '/Jane Smith/{print $6}' data.csv

For more information, reference the documentation for the awk command
printf
The printf command formats and prints data. It works similarly to printf
in C, C++ and Java. In case you are unfamiliar or need a reminder, here is a
summary of the common parameters used with printf:
| Parameter | Description |
|———– |————————————-|
| %d | integer number printed in decimal |
| %f | floating point number |
| %c | character |
| %s | string |

Unlike echo, printf does not send an automatic newline at the end of the
output. The printf command allows you to optionally assign the result to a
variable rather than outputting to the screen (which is useful in Bash
scripts).

Take a look at the man pages for the printf command:

man printf

Let’s take a look at the first syntax pattern (printf format [arguments])
which prints ARGUMENT(s) according to FORMAT.

Try the following example to print a formatted string:

printf "%s got %s wrong answer(s)\n" "Jane" "1"

The printf command can be followed by a format (such as “%s

got %s wrong answer(s)”) followed by arguments (such as “Jane”
“1”). This example would output Jane got 1 wrong answer(s)

Try these variations:

Print the result of a math equation.

Solution

To execute a math expression in Bash you use a $ followed by the

expression enclosed in double parentheses. Use the %d to format a
decimal number.

printf "%d\n" $((8+4))

Print only the first 4 digits beyond the decimal point of a floating point
number.

Solution

You can use the precision modifier . followed by a number to set the
number of digits beyond the decimal point to display. Use the %f to
format a floating point number.

printf "%.4f\n" 3.1415926535

nano
Typing nano at the command line will bring up a no-frills text editor. You
can specify a file name on the command line and that file will open.

Ctrl-x exits from the nano editor. Nano does not autosave, you must type
Ctrl-o. The other commands that can be used with nano are listed at the
bottom of the edit screen.

Take a look at the man pages for the nano command

man nano

You can specify -B or --backup on the command line if you want to create a
backup version of the current file before changes are saved.

Using nano to create and edit files

Create a new file called hello.txt and enter the text Hello World! by
following these steps:
1. nano hello.txt
2. Type Hello World!
3. Type Ctrl-o and then the Return/Enter key to confirm the file name.
4. Type Ctrl-x to exit.
5. Type cat hello.txt to print a listing of the file to the console.
vim
Vim is a free and open-source text editor controlled from the keyboard.
Some versions of Vim do not support navigation with a mouse and it does
not provide menus.

Take a look at the man pages for the vim command:

man vim

Typing vim at the command line will bring up the vim text editor. You can
specify a file name on the command line and that file will open.

Overview of Vim navigation

Vim has different modes you can switch between. To add text to a file, you
will need to be in insert mode. To end a mode, you press esc in the top-left
of your keyboard.

important

The : in Vim
The : is used in Vim to enter command mode - always remember to
hold down the shift key to type it - otherwise you will type a ;

Here is a summary of mode switching

Key Description
i Enter insert mode
: Enter command mode
R Enter replace mode
v Enter visual mode (highlighting)
V Enter visual line mode (highlighting lines)
esc Return to normal mode from insert or replace mode
esc+esc Return to normal mode from command or visual mode
Aside from insert mode, Vim users also regularly use command mode. For
example, Vim does not auto-save, you must type :w and click Return or
Enter. The : enters command mode and the w specifies save. Return or
Enter tell vim you have completed the command.

Here is a summary of commands - remember to press : to

enter command mode and Enter or Return to end the
command:

Key Description
w Save the current file
Save the current file
wq and close it, exits vim
if no other open files
Save a copy of the
current file as newname
w newname
but continue editing
the current file
Save a copy of the
sav newname current file as newname
and edit newname
Close a file without
q!
saving
e filename Open another file
Write changes made to
x the current file and
close it

Create and edit a file using vim

Create a new file called testvim.txt and enter the text I'm learning about
the vim editor by following these steps:
1. Type vim testvim.txt
1. Type i to enter “insert” mode.
1. Enter the following text I'm learning about the vim editor
1. Press esc followed by :wq followed by the Return or Enter key to save
and quit
1. Type cat testvim.txt to list out the file you created

For more information, reference the documentation for vim. There is also a
quick reference.