topic Re: Multi-line output to single line in Operating System - Linux

Multi-line output to single line

Patrick Ware_1 — Mon, 25 Feb 2008 05:49:33 GMT

Hello,

How can I take the following multi-lined output:

outputa
outputb
outputc

and turn it into single line ouput, with a single space between each field like below:

outputa outputb outputc

Re: Multi-line output to single line

Khairy — Mon, 25 Feb 2008 07:51:30 GMT

hi patrick,

try this. This format the output to a 10000 columns width line.

# | fmt -10000
OR
# command > file.txt
# cat file.txt | fmt -10000
outputa outputb outputc outputd

Please assign some points if this answer yr questions...

Thanx!

Re: Multi-line output to single line

Peter Nikitka — Mon, 25 Feb 2008 08:39:42 GMT

Hi,

if there is no need for the special handling, that fmt does, use
tr '\012' ' '
- the last entry is single_quote-space-single-quote
- read form stdin, writes to stdout

mfG Peter

Re: Multi-line output to single line

Patrick Ware_1 — Mon, 25 Feb 2008 09:41:05 GMT

Ok, so how does this work on a file?

Re: Multi-line output to single line

Patrick Ware_1 — Mon, 25 Feb 2008 09:42:59 GMT

I also ended up getting the following solution:

All on one line:

awk -v x="" '{ s=s sprintf(x "%s" x " ", $0) } END { sub(",$", "", s); print(s) }' file.txt

Re: Multi-line output to single line

Patrick Ware_1 — Mon, 25 Feb 2008 09:57:24 GMT

I tried the tr command, and got the following:

tr '\012' ' ' < sdisks1.txt
datadg01 datadg02 datadg03 datadg04 datadg05 root@server:>

Re: Multi-line output to single line

Peter Nikitka — Mon, 25 Feb 2008 10:16:25 GMT

Hi,

when you need the data in another file:
tr '\012' ' ' < sdisks1.txt >sdisk1.out

If you need a trailing newline:
(tr '\012' ' ' < sdisks1.txt; print) >sdisk1.out

mfG Peter

Re: Multi-line output to single line

James R. Ferguson — Mon, 25 Feb 2008 12:13:15 GMT

Hi Patrick:

Be aware that the 'fmt' command counts not only the string lengths but also the column separators in limiting the line width it composes. That may not be a problem in your case.

The use of 'cat' to read the input file and pipe it to 'fmt' runs a superflous process. Simply do:

# fmt -10000 file.txt

Peter's use of 'tr' shows the same economy of resources,

Regards!

...JRF...

Re: Multi-line output to single line

Dennis Handly — Sat, 17 Sep 2011 20:13:36 GMT

Did you only have 3 lines? If so, you use vi and join them. You could also use sed(1).

If you have sets of 3 lines and you want to combine all 3, then you can use awk:
awk '
{
getline l2
getline l3
print $0, l2, l3 }' file

If you have N lines and you want to to put them all together, you can use tr (and Peter's echo).

Re: Multi-line output to single line

James R. Ferguson — Mon, 25 Feb 2008 21:30:31 GMT

Hi (again):

When Dennis wrote:

-> print $0, $l2, $l3 }

...he meant:

-> print $0, l2, l3 }

...without the sigil.

I also suspect that for Peter's solution will be the fastest (and most generalized) with the smallest footprint, although there are (usually) many ways to do something :-)

Regards!

...JRF...