Understanding Sed Addressing

Oct 13, 2025 by wbrum.org 91 views

Introduction

Sed is a powerful command-line text processing tool that allows you to perform various operations on text files, including search and replace, filtering, and manipulation. However, when working with sed, you may encounter situations where you want to exclude specific lines from replacement. In this article, we will explore how to achieve this using sed.

Understanding Sed Addressing

Before we dive into the solution, it's essential to understand how sed addressing works. Sed uses a concept called "addressing" to specify which lines to operate on. You can use various addressing modes, including:

Line numbers (e.g., 1, 4-7, 38, 39)
Patterns (e.g., /pattern/)
Ranges (e.g., 1,10, 20,30)

Solution

To exclude specific lines from replacement, you can use sed's addressing modes in combination with the d command, which deletes the specified lines. Here's an example command that removes punctuation marks (,, :, {, }, [, ]) from all lines except those specified:

sed -e '1d; 4,7d; 38d; 39d; s/[^[:alnum:][:space:]]//g' input.txt

Let's break down this command:

1d: Delete the first line (line 1).
4,7d: Delete lines 4 through 7.
38d: Delete line 38.
39d: Delete line 39.
s/[^[:alnum:][:space:]]//g: Replace all characters that are not alphanumeric or whitespace with an empty string (i.e., remove punctuation marks).

Explanation

The s command in sed is used for substitution. The pattern [^[:alnum:][:space:]] matches any character that is not alphanumeric (letters or numbers) or whitespace. The g flag at the end of the command makes the substitution global, meaning it will replace all occurrences of the pattern in each line.

Example Use Case

Suppose you have a file input.txt containing the following text:

# This is a sample text
with punctuation marks: , : { } [ ]
and some numbers: 123
more text with punctuation: ! @ # $
and some special characters: ^ & * ( )

Running the sed command above on this file will produce the following output:

# This is a sample text
with punctuation marks
and some numbers: 123
more text with punctuation
and some special characters

As you can see, the punctuation marks (,, :, {, }, [, ]) have been removed from all lines except lines 1, 4-7, 38, and 39.

Conclusion

Introduction

In our previous article, we explored how to exclude specific lines from replacement in sed using addressing modes and the d command. In this article, we will answer some frequently asked questions (FAQs) related to this topic.

Q: How do I exclude multiple lines from replacement?

A: To exclude multiple lines from replacement, you can use sed's addressing modes in combination with the d command. For example, to exclude lines 1, 4-7, and 38-39 from replacement, you can use the following command:

sed -e '1d; 4,7d; 38,39d; s/[^[:alnum:][:space:]]//g' input.txt

Q: Can I exclude lines based on a pattern?

A: Yes, you can exclude lines based on a pattern using sed's addressing modes. For example, to exclude lines that contain the word "example" from replacement, you can use the following command:

sed -e '/example/d; s/[^[:alnum:][:space:]]//g' input.txt

Q: How do I exclude lines that contain a specific character?

A: To exclude lines that contain a specific character, you can use sed's addressing modes with a pattern that matches the character. For example, to exclude lines that contain the character "!" from replacement, you can use the following command:

sed -e '/!/d; s/[^[:alnum:][:space:]]//g' input.txt

Q: Can I exclude lines based on a range of lines?

A: Yes, you can exclude lines based on a range of lines using sed's addressing modes. For example, to exclude lines 10-20 from replacement, you can use the following command:

sed -e '10,20d; s/[^[:alnum:][:space:]]//g' input.txt

Q: How do I exclude lines that are empty?

A: To exclude lines that are empty, you can use sed's addressing modes with a pattern that matches an empty line. For example, to exclude empty lines from replacement, you can use the following command:

sed -e '/^$/d; s/[^[:alnum:][:space:]]//g' input.txt

Q: Can I exclude lines based on a specific file?

A: Yes, you can exclude lines based on a specific file using sed's addressing modes. For example, to exclude lines from a file called "excluded_lines.txt" from replacement, you can use the following command:

sed -e '/^>/d; s/[^[:alnum:][:space:]]//g' input.txt

Note: The > symbol is used to indicate the start of a line in the excluded_lines.txt file.

Conclusion

In this article, we answered some frequently asked questions related to excluding specific lines from replacement in sed. We covered topics such as excluding multiple lines, excluding lines based on a pattern, excluding lines that contain a specific character, excluding lines based on a range of lines, excluding empty lines, and excluding lines based on a specific file. By mastering sed's addressing modes and commands, you can perform complex text processing tasks with ease.