Bash Script to Find and Replace Text in a File

Use sed -i 's/old_text/new_text/g' filename in a Bash script to find and replace all occurrences of old_text with new_text directly in the file.

Examples

InputFile content: Hello world Command: sed -i 's/world/universe/g' file.txt

OutputFile content after: Hello universe

InputFile content: apple, banana, apple Command: sed -i 's/apple/orange/g' fruits.txt

OutputFile content after: orange, banana, orange

InputFile content: no match here Command: sed -i 's/xyz/abc/g' file.txt

OutputFile content after: no match here (unchanged)

How to Think About It

To replace text in a file using Bash, think about searching for the exact text you want to change and then replacing it everywhere it appears. The sed command is perfect because it can find and replace text directly inside the file without needing extra steps.

Algorithm

1

Get the filename, the text to find, and the text to replace from the user or script arguments.

2

Use a command that searches the file for the target text.

3

Replace all occurrences of the target text with the new text.

4

Save the changes directly to the original file.

Code

bash

#!/bin/bash

# Variables for old and new text and filename
old_text="apple"
new_text="orange"
file="fruits.txt"

# Replace all occurrences of old_text with new_text in the file
sed -i "s/${old_text}/${new_text}/g" "$file"

# Print the updated file content
cat "$file"

Output

orange, banana, orange

Dry Run

Let's trace replacing 'apple' with 'orange' in 'fruits.txt' containing 'apple, banana, apple'.

1

Initial file content

apple, banana, apple

2

Run sed command

sed -i 's/apple/orange/g' fruits.txt

3

File content after replacement

orange, banana, orange

Step	Action	File Content
1	Start	apple, banana, apple
2	Replace 'apple' with 'orange'	orange, banana, orange

Why This Works

Step 1: Using sed for replacement

The sed command edits text in a file by applying a substitution pattern.

Step 2: The -i option

The -i flag tells sed to edit the file in place, saving changes directly.

Step 3: The substitution pattern

The pattern s/old/new/g means replace all (g) occurrences of old with new.

Alternative Approaches

Using perl command

bash

perl -pi -e 's/old_text/new_text/g' filename

Perl can also do in-place replacement and supports more complex patterns but may not be installed by default.

Using temporary file with grep and mv

bash

grep -rl 'old_text' . | xargs sed 's/old_text/new_text/g' > temp && mv temp filename

This method uses a temporary file and is less efficient but works if <code>sed -i</code> is not available.

Complexity: O(n) time, O(1) space

Time Complexity

The script reads through the file once, so time grows linearly with file size.

Space Complexity

The replacement is done in place, so no extra space proportional to file size is needed.

Which Approach is Fastest?

Using sed -i is fastest and simplest for plain text replacements compared to alternatives like Perl or temporary files.

Approach	Time	Space	Best For
sed -i	O(n)	O(1)	Simple in-place text replacement
perl -pi	O(n)	O(1)	Complex patterns, regex support
temp file with sed	O(n)	O(n)	Systems without sed -i support