this post was submitted on 15 Feb 2024
16 points (94.4% liked)

commandline

1747 readers
1 users here now

founded 1 year ago
MODERATORS
 

I have a list containing a set of tags, and would like to exclude one tag, unless another tag exists in the line.

Say I have the following list, and want to exclude B, unless A is present.

[A,B]
[A,C]
[B,C]
[A]
[B]

I can reverse grep for B:

> grep --invert-match "B"
[A,C]
[A]

How can I find the previous list, but also the item containing [A,B]?

top 6 comments
sorted by: hot top controversial new old
[–] borf@lemmynsfw.com 7 points 9 months ago (1 children)

I like to use awk instead of grep wherever possible, especially for weird logic like this.

awk '!/B/ || /A.*B/' is one way to skin that cat. If you don't care what order A and B are in on the lines containing both, then awk '!/B/ || (/A/ && /B/)' will work.

[–] TheLugal@lemmy.world 1 points 9 months ago

Thank you! This was exactly was I was looking for. :)

[–] Andy@programming.dev 7 points 9 months ago* (last edited 9 months ago) (2 children)
grep -E '(^[^B]*$|A)'

EDIT: Whoops, I meant to make this a top-level comment.

EDIT 2: On one client it looked like a nested comment and on this other client it looks top level and now I'm a confused old man.

[–] bizdelnick@lemmy.ml 1 points 9 months ago (1 children)

You don't need parentheses here.

[–] Andy@programming.dev 1 points 9 months ago

~~How dare you?~~

Thanks!

[–] TheLugal@lemmy.world 1 points 9 months ago

Haha, on Lemmy.World it looks like a top level. Thank you, either way :)