Cut from string linux

8 Cut Command Examples [Cut Sections of Line in File]

As Linux users, we interact with the text files on a regular basis. One of the common operations we perform on these files is text filtering. Linux provides many command-line utilities for text filtering, such as – grep, fgrep, sed, awk, and the list goes on.

However, in this guide, we will discuss one more text-filtering utility called cut, which is used to remove a particular section from the input line. The cut command performs filtering based on the byte position, character, field, and delimiter.

In this beginner’s guide, we will learn about the cut command with examples in the Linux command line. After following this guide Linux command line users will be able to use the cut command efficiently in their day-to-day life.

cut Command Syntax

The syntax of the cut command is just like any other Linux command:

In the above syntax, the angular bracket (<>) represents the mandatory arguments whereas the square bracket ([]) represents the optional parameters.

Now that we are familiar with the syntax of the cut command. Next, let’s create a sample file to use as an example:

Linux Sample Text File

1. Print First Byte of File

The cut command allows us to extract the text based on byte position using the -b option.

Let’s use the below command to extract the first byte from each line of the file:

Print First Byte from File

In this example, we can see that the cut command shows only the first character because all characters are one byte long.

2. Print Multiple Bytes of File

In the previous example, we saw how to select a single byte from the file. However, the cut command also allows us to select multiple bytes using the comma.

Читайте также:  Операционные системы linux убунту

Let us use the below command to select the first four bytes from the file:

Print Multiple Bytes from File

In this example, we have selected the consecutive bytes but that is not mandatory. We can use any valid byte position with the cut command.

3. Print a Range of Bytes of File

In the previous example, we used commas to select consecutive bytes. However, that approach is not suitable if we want to select a large number of bytes consecutively. In such cases, we can use the hyphen (-) to specify the byte range.

To understand this, let’s use the 1-12 as a byte range to select the first twelve bytes from each line:

In a similar way, we can select a byte range from the middle as well. For example, the below command selects bytes from column numbers 17 to 30:

Print Range of Bytes from File

4. Print Specify Start Byte Position of File

Sometimes, we want to extract all the text from a certain byte position. In such cases, we can omit the end byte position.

For example, we can use the following command to print all bytes starting from position 17:

Print Text from Certain Byte Position

In the above command, 17 represents the starting byte position whereas the hyphen (-) represents the end of the line.

5. Print Specify End Byte Position of File

In a similar way, we can also specify the end byte position only. For example, the below command prints all the bytes from the start of the line till the 12th column:

Priint Specify End Byte Position of File

In the above command, the hyphen (-) represents the start of the line whereas 12 represents the end byte position.

6. Cut First Byte by Character Position

In the last few sections, we saw how to perform text extraction based on byte position. Now, let’s see how to perform the text extraction by character position.

To achieve this, we can use the -c option to cut the first byte from the following multi-byte string:

$ echo école | cut -b 1  

In the above output, we can see that, the cut command shows a question mark instead of the character é . This happens because we are trying to print the first byte from the multi-byte character.

Читайте также:  Установка коды на линукс

Now, let’s use the -c option to cut the same multi-byte character and observe the result:

$ echo école | cut -c 1 é 

In the above output, we can see that now the cut command shows the expected output.

It is important to note that, not all versions of the cut commands support the multi-byte characters. For example, Ubuntu and its derivatives don’t support multi-byte characters.

To understand this, let’s run the same command on the Linux Mint which is derived from Ubuntu:

Cut Byte by Character Position

Here we can see that, the cut command treats both multi-byte characters and single-byte characters the same. Hence it doesn’t generate the expected output.

7. Cut a String by Delimiter in Linux

By default, the cut command uses a TAB character as a delimiter. However, we can override this default behavior using the -d option.

Oftentimes, the -d option is used in combination with the -f option which is used to select a particular field.

To understand this, let’s use the space character as a delimiter and print the first two fields using the -f option:

Cut String By Delimiter in File

In the above example, we have used the comma with the -f option to select multiple fields.

8. Print All Bytes or Characters Except Selected Ones

Sometimes, we want to print all the characters except a few ones. In such cases, we can use the —complement option. As the name suggests, this option prints all the columns except the specified ones.

$ cut -c 1 --complement file.txt

Print Not Selected Fields in File

In the above output, we can see that the —complement option prints all the characters except the first character.

It is important to note that, in this example, we have used commas to select multiple fields. However, we can also use the other supported ranges as well. We can refer to the first few examples of this tutorial to understand more about the ranges.

Conclusion

In this beginner’s guide, we discussed the practical examples of the cut command with text filtering based on the byte position, character position, and delimiter.

Do you know of any other best example of the cut command in Linux? Let us know your views in the comments below.

Источник

Читайте также:  Linux mint color themes

Shell script : How to cut part of a string

I pasted the contents of your example into a file named so.txt .

$ cat so.txt | awk '< print $7 >' | cut -f2 -d" mt24">
)" data-controller="se-share-sheet" data-se-share-sheet-title="Share a link to this answer" data-se-share-sheet-subtitle="" data-se-share-sheet-post-type="answer" data-se-share-sheet-social="facebook twitter devto" data-se-share-sheet-location="2" data-se-share-sheet-license-url="https%3a%2f%2fcreativecommons.org%2flicenses%2fby-sa%2f2.5%2f" data-se-share-sheet-license-name="CC BY-SA 2.5" data-s-popover-placement="bottom-start">Share
)" title="">Improve this answer
answered Sep 18, 2010 at 15:46
4
    Thaks this is exactly what I wanted !
    – deimus
    Sep 18, 2010 at 16:00
    1
    useless use of cat and cut. awk can do all of it.
    – ghostdog74
    Sep 19, 2010 at 14:57
    4
    @user229426: You mean unnecessary use of cat and cut, not useless, don't you?
    – Manoj Govindan
    Sep 20, 2010 at 5:00
    This is a great explanation of using awk in combination with cut. I tried to get this exact thing for another use case and struggled a lot with awk man pages and awk tutorials. Thanks a lot for taking the time to explain your code!
    – Kevin G.
    Dec 22, 2015 at 12:28
Add a comment|
3

Use a regular expression to catch the id number and replace the whole line with the number. Something like this should do it (match everything up to "id mt24">

)" data-controller="se-share-sheet" data-se-share-sheet-title="Share a link to this answer" data-se-share-sheet-subtitle="" data-se-share-sheet-post-type="answer" data-se-share-sheet-social="facebook twitter devto" data-se-share-sheet-location="2" data-se-share-sheet-license-url="https%3a%2f%2fcreativecommons.org%2flicenses%2fby-sa%2f2.5%2f" data-se-share-sheet-license-name="CC BY-SA 2.5" data-s-popover-placement="bottom-start">Share
)" title="">Improve this answer
)">edited Sep 18, 2010 at 16:12
answered Sep 18, 2010 at 15:51
2
    this method gives 9 and 1 but it should be 9 and 10
    – deimus
    Sep 18, 2010 at 15:57
    Right, I've made a mistake when pasting the sed line. Updated it to work correctly.
    – bluebrother
    Sep 18, 2010 at 16:12
Add a comment|
2

A perl-solution:

perl -nE 'say $1 if /id=(\d+)/' filename

Источник

Оцените статью
Adblock
detector