Deleting a Specific Line From a Text File in Command Line in Linux
On Linux, how to delete a specific line from a text file in command line? For example, to delete the 4th line from a file
aaa
bbb
ccc
ddd
eee
ffffff
You can use the “stream editor for filtering and transforming text” sed.
With GNU sed:
sed -i '4d' ./file
Here, -i means edit the file inplace. d is the command to “delete the pattern space; immediately start next cycle”. 4 means the 4th line.
The file content will be:
aaa
bbb
ccc
eee
ffffff
There are more combinations for deleting lines. Some examples are:
Remove the last line:
sed '$d' filename.txt
Remove all empty lines:
sed '/^$/d' ./file
or
sed '/./!d' ./file
Remove lines from 7 to 9:
sed '7,9d' ./file
Remove the line matching by a regular expression REGULAR:
sed '/REGULAR/d' ./file
For a simple example, remove the lines containing “oops”:
sed '/oops/d' ./file
This was very helpful.
Question.
Once we have the offending line removed, how do we (using the command line) overwrite the file with the new, corrected version?
For example…
sed ‘//d’ file.txt
removes the html expression but now I want to save this version to the same file.
Thank you for any suggestions.
Greetings. I found the answer. Insert -i after sed command.
sed -i ‘//d’ file.txt
overwrites the file with the corrected information.
But
What if the expression to remove contains a forward slash character. Haven’t figured out how to make this work.
sed -i ‘//d’ file.txt
There must be a way to escape the / character in the expression to remove. I’ve tried surrounding it with ‘ and \. Nether seem to be the solution.
Hai we can delete a particular line in text file using sed command. But again how to recover that deleted text in that filename after deleting if by a mistake?