43

Is there a simple utility or script to columnate the output from one of my scripts? I have data in some form:

A aldkhasdfljhaf
B klajsdfhalsdfh
C salkjsdjkladdag
D lseuiorlhisnflkc
E sdjklfhnslkdfhn
F kjhnakjshddnaskjdh

but if this becomes two long, write the data in the following form (where still vertically ordered):

A aldkhasdfljhaf    D lseuiorlhisnflkc
B klajsdfhalsdfh    E sdjklfhnslkdfhn
C salkjsdjkladdag   F kjhnakjshddnaskjdh

From reading the manpage, I don't think that this is something column would be appropriate for but I'm not sure. It's easy enough to split in the form:

A B 
C D 
E F

by only printing \n every second line (what my current script does). Any ideas? Thanks!

Improve this question
1
  • an example: tree -L 2 -F | column | less -N which list dir & files up to 2 levels, with multiple columns. Commented Apr 26, 2024 at 19:25

4 Answers 4

Reset to default
36

column seems to be what you want:

$ cat file
A aldkhasdfljhaf
B klajsdfhalsdfh
C salkjsdjkladdag
D lseuiorlhisnflkc
E sdjklfhnslkdfhn

$ column file
A aldkhasdfljhaf    D lseuiorlhisnflkc
B klajsdfhalsdfh    E sdjklfhnslkdfhn
C salkjsdjkladdag   F kjhnakjshddnaskjdh
Improve this answer
3
  • I did try. But wrongly (with -t)... But for yours to work reliably you probably need --columns`. Commented Jul 4, 2013 at 14:43
  • I did try but for some reason I didn't realise that it depended on the width of your terminal window (seems obvious now)! Commented Jul 4, 2013 at 15:22
  • also I was testing this on two machines at once. Apparently the old version (circa 1993 on Scientific Linux) behaves differently to the newer 2004 version... Commented Jul 4, 2013 at 15:35
20

To columnate the output, pr converts text files for printing:

-COLUMN, --columns=COLUMN: output COLUMN columns
      and print columns down, unless -a is used.

-t, --omit-header : omit page headers and trailers

e.g.

ls /etc | pr -2 -t
abrt                                gtk-2.0
acpi                                hal
adjtime                             hba.conf
agent.properties                    host.conf
aliases                             hosts
aliases.db                          hosts.allow
...
Improve this answer
1
  • not working properly for large (long) file. Commented May 7, 2019 at 18:03
3

You can split the file in two (with the same number of lines or the first file having one line more) and then do this:

paste file1 file2

If the length of the lines is too different then is may be necessary to use printf to adapth the lengths by padding with spaces.

Improve this answer
1
  • 2
    Or ... | paste - - when "splitting long output"... Commented Apr 7, 2015 at 1:17
3

You can use just plain columns:

$ cat test.txt | columns
A aldkhasdfljhaf             B klajsdfhalsdfh             C salkjsdjkladdag
D lseuiorlhisnflkc           E sdjklfhnslkdfhn            F kjhnakjshddnaskjdh

NOTE: the columns command is part of the autogen package on my Fedora 14 system.

$ rpm -qf /usr/bin/columns
autogen-5.9.4-7.fc12.x86_64

References

Improve this answer
2
  • columns: command not found. autogen? Commented Jul 5, 2013 at 23:16
  • @Gilles - yes it's part of autogen. Just checked on my Fedora 14 system. I'll mention it in the answer as well. Commented Jul 5, 2013 at 23:18

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.