CODEWITHSUNDEEP
unixsed
John Smith
John Smith

Reputation: 113

How to delete a specific line from a file in unix?

I have a very large file from which I need to delete a specific line (line number 941573 )

I'm somewhat new to this environment, but I've been googling the problem to no avail.

I've tried using the sed command as such, but it doesn't seem to be working 

sed -e '941572,941574d' filenameX > newfilenameY

I've also tried

sed -e '941573d' filenameX > newfilenameY

Yet the 'newfilenameY' file and the original file 'filenameX' both still contain the line that I'm trying to delete. It's a fastq file, though I don't see how that would make any difference. Like I said I'm new to unix so maybe I've gotten the sed command wrong

Upvotes: 9

Views: 23311

Answers (3)

cdarke
cdarke

Reputation: 44434

I generated a test file with 1000000 lines and tried your sed -e '941573d' filenameX > newfilenameY and it worked fine on Linux.

Maybe we have some other misunderstanding. Line numbers count from one, not zero. If you counted from zero then you would find line 941572 was missing.

Did you try a diff filenameX newfilenameY? That would highlight any unexpected changes.

I don't know much about FASTQ format, but are you sure we are talking about text file line numbers, and not sequence numbers?

There is a general line length limit of 4096 bytes, do any of your lines exceed that? (That's unlikely, but I thought it worth the question).

Upvotes: 0

user647772
user647772

Reputation:

d deletes a line/lines. So your second approach works.

$ sed '941573d' input > output

Long Example:

% for i in $(seq 1000000)
do
echo i >> input
done
% wc -l input
1000000 input
% sed '941573d' input > output
% wc -l output
999999 output
% diff -u input output                                      :(
--- input       2012-10-22 13:22:41.404395295 +0200
+++ output      2012-10-22 13:22:43.400395358 +0200
@@ -941570,7 +941570,6 @@
 941570
 941571
 941572
-941573
 941574
 941575
 941576

Short Example:

% cat input
foo
bar
baz
qux
% sed '3d' input > output
% cat output             
foo
bar
qux

Upvotes: 10

user1746291
user1746291

Reputation: 333

Here is how to remove one or more lines from a file.

Syntax: 

sed '{[/]<n>|<string>|<regex>[/]}d' <fileName>      
sed '{[/]<adr1>[,<adr2>][/]d' <fileName>
/.../=delimiters
n = line number
string = string found in in line
regex = regular expression corresponding to the searched pattern
addr = address of a line (number or pattern )
d = delete

Upvotes: 1

Related Questions