How to delete this pattern with sed or awk?

Question

#====================================================================== #NAME #====================================================================== gen ( id 111 descr Test 1 txcd content ) #====================================================================== #NAME #====================================================================== gen ( id 112 descr Test 2 txcd content ) #====================================================================== #NAME #====================================================================== gen ( id 123 descr Test 3 txcd content )

I have this pattern, and have tried to find a way to delete a certain id using sed or awk. I pass the ID using a function parameter, $1. I've looked it up on stack but couldn't find a solution that worked for me.

sed -i "/gen $/{:a;N;/$/!ba};/$1/d" file

At a certain point when editing this sed command it did delete a line, but just the line where the ID itself was.

This is the one I've been working with, I've made small changes but nothing worked thus far. I'm supposed to delete the hashtags on top as well, but to begin with, I was mainly trying to delete the gen pattern.

You were nearly there. When you quoted ( placed a \ before a character) the ( and ) you changed their meaning into a group instead of a literal ( or ). Perhaps a function f () { sed -i '/^gen (/{:a;N;/^)/M!ba;/'"$1"'/d' "${2:-file}"; } would meet your requirements (if you use bash)? — potong
– potong, Commented Feb 7, 2020 at 14:13
It is bash. I was getting an error and the parentheses were being indicated as the reason, so I thought I might have to escape them. — albuck97
– albuck97, Commented Feb 7, 2020 at 14:52
It is safer to surround sed commands by single quotes (as in my first comment) and then the side effects of bash interpolation will not occur. — potong
– potong, Commented Feb 8, 2020 at 17:46

kvantour · Accepted Answer · 2020-02-07 15:53:33Z

1

If a lone ) always terminates one record, this is easy enough in Awk.

awk -v RS=')' -v ORS=')' '!/id +112/' file

edited Feb 7, 2020 at 15:53

kvantour

26.9k4 gold badges57 silver badges81 bronze badges

answered Feb 7, 2020 at 12:47

tripleee

192k37 gold badges318 silver badges367 bronze badges

Sign up to request clarification or add additional context in comments.

7 Comments

albuck97 Over a year ago

Thank you so much, I did notice that I could do something with the closing parentheses, but I'm really inexperienced with sed, and even more inexperienced with awk. But I'm definitely noticing that it's more than worthwhile to really learn it. Thanks again.

tripleee Over a year ago

Tested on a Mac, which platform are you on?

albuck97 Over a year ago

I'm using red hat linux

albuck97 Over a year ago

Actually just noticed it leaves a parentehese

albuck97 Over a year ago

*at the end of the file with the edited solution, no matter where I delete, middle beginning or end, extra parenthesis at the end.

|

Collectives™ on Stack Overflow

How to delete this pattern with sed or awk?

1 Answer 1

7 Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

7 Comments

Related