Tuto3: Sed command.md

September 24, 2023 · View on GitHub

Pracatical for sed command

For this tutorial we will manupilate file by sed commands.
Sed is one of the most used rapid and powerfull unix commands. It stands for "stram eaditor". while it is like grep command, can fetch into files and search for spécified patterns "matching", it can also perform substitution, insertion and deletion functions. Its syntax :

sed OPTIONS.. [scripts] [Inputfile]

+++ Introduction : Command's examples :

sed 's/sad/happy/' myInputfile.txt

The “ s ” specifies the substitution operation, “ / ” are delimiters, the “sad” is the search pattern and the “happy” is the replacement string.

sed '/Thomas/d' Path_to_success.txt

The “ d ” specifies the deletion operation Deletes, and it deletes lines which contains the matched pattern from the input file.

sed -n '1,3p' Path_to_success.txt

This command print out the firt 3 lines, the print operation is specified by “ d ” and the “ -n ” is for suppressing automatic printing of pattern space.

Note that we can combine commands by “ -e ” using option.

sed -e 's/sad/happy/' -e'/Thomas/d' Path_to_success.txt

+++ Printing out practices

In this tutorial we will be working on a csv file named Laptops.txt
Please download the required file into Dir2 whether using wget command or manually from github.

First let check what our file looks like and how many lines there in?

Answers

cat Laptops.txt

wc -l Laptops.txt

Try this command and figure out what it does mean :

sed '1,5p' Laptops.txt

then print the last lines : use tail command:

tail Laptops.txt

- Notice there is no difference between the two outputs, means the first command printed out all the file. You could also use this specified command to print all the file :

sed -n 'p' Laptops.txt

Add "-n" to sed command then rexecute it;

sed -n '1,5p' Laptops.txt

Let print lines 1 to 4 , 7 to 9 and 24 :

sed -n '1,4p;7,9p;24p' Laptops.txt

Verify your output by comparing to this command which prints number lines.

cat -n Laptops.txt

We want to printout some lines with a condition ; first let us choose arbitrary any line, for example line number 13, for this line I'll print out how many character there in :

sed -n '13p' Laptops.txt | wc -m

Now based on this condition we will print lines which equal to, shorter and longer than line n°13:

line shorter than line n°13 :

sed -n '/^.\{0,49\}$/p' Laptops.txt

line equal to line n°13 :

sed -n '/^.\{50\}$/p' Laptops.txt

line longer line n°13 :

sed -n '/^.\{51,\}$/p' Laptops.txt

/^.{51,}/: Regular expression pattern we're matching :
^: Anchors the pattern to the beginning of the line.
.: Matches any character.
{51,}: Matches 51 or more occurrences of the preceding character.
$: Anchors the pattern to the end of the line.

let print out odd number lines

sed -n 'p;n’

let print out even number lines

sed -n ‘n;p’

* Not all patterns change as a result of a substitution because the `sed` command only replaces the first matching occurrence. To replace all the matches to the pattern regardless of the number of times it appears, use the global flag "g":

sed 's/ASUS/asus/g' Laptops.txt | cat -n

* Please check line 6 agian

Notice that changing "ASUS" to "asus" was done by simply typing the word in lowercase. There are other ways to achieve the same result, such as using a lowercase conversion function or method.

sed 's/ASUS/\L&/g' Laptops.txt

Try to change the word "year" to uppercase :