making two column-data

ovince · 03-10-2007, 10:06 AM

hi

I have file like

f1
g1
f2
g2
...
f10
g10

and I would like to make 2-column data like

f1 g1
f2 g2
....
f10 g10

I remember that I read somewhere that this is easily done with cat command. Something like

cat -2 fileName

but I can not find exactly how. Could somebody refresh my memory?

thanks
oliver

tronayne · 03-10-2007, 10:18 AM

I think you're talking about paste?

For example,

Code:

paste f1 g1 > f1g1

ovince · 03-10-2007, 10:23 AM

ahhh ... I was so unprecise

I have 1 file with content

f1
g1
f2
g2
...
f10
g10

I would like to make another file from it with content

f1 g1
f2 g2
....
f10 g10

paste will work job

ovince · 03-10-2007, 10:23 AM

paste will work job? how?

cfaj · 03-10-2007, 10:52 AM

Quote:

Originally Posted by ovince

ahhh ... I was so unprecise

I have 1 file with content

f1
g1
f2
g2
...
f10
g10

I would like to make another file from it with content

f1 g1
f2 g2
....
f10 g10

paste will work job

Code:

paste -d ' ' - - FILENAME > NEWFILE

ovince · 03-10-2007, 11:09 AM

thanks for reply

what if I wanna make 20 column data? do I write like

paste -d ' ' - - - - - - - - - - - - - - - - - - - - - ... FILENAME > NEWFILE

tronayne · 03-10-2007, 11:14 AM

Oops, didn't read that properly. ONE file, not many -- sorry about that.

ovince · 03-10-2007, 12:01 PM

I ask this because I remember in cat the 'column number' can be given by number

pwc101 · 03-10-2007, 12:42 PM

This is an ugly way of doing it (and it assumes you're looking for things which have "f" and "g" in them which are unique):

Code:

grep f input_file > 1.tmp && grep g input_file > 2.tmp && paste 1.tmp 2.tmp $$ \rm 1.tmp 2.tmp

I'm pretty certain that it can be done a lot more efficiently using awk, but I don't know enough awk to get it to work. Just in case you want to look into it more, I've got awk to print 1 column of the input file based on some unique indentifier in the column:

Code:

awk '/f/ {print $1}' input_file > output_file

What I can't get it to do is to sort the f's first, then the g's, something like:

Code:

awk '{if $1=f* print $1 else if $1=g* print $2}' input_file > output_file

Obviously that's totally wrong, but I'm sure it's possible.

Hope the first bit helps!

cfaj · 03-10-2007, 02:13 PM

Quote:

Originally Posted by ovince

thanks for reply

what if I wanna make 20 column data? do I write like

paste -d ' ' - - - - - - - - - - - - - - - - - - - - - ... FILENAME > NEWFILE

Try it and see.

ghostdog74 · 03-10-2007, 08:14 PM

Quote:

Originally Posted by ovince

hi

I have file like

f1
g1
f2
g2
...
f10
g10

and I would like to make 2-column data like

f1 g1
f2 g2
....
f10 g10

I remember that I read somewhere that this is easily done with cat command. Something like

cat -2 fileName

but I can not find exactly how. Could somebody refresh my memory?

thanks
oliver

try this:

Code:

sed -e 'N' -e 's/\n/ /g' file

ovince · 03-11-2007, 01:18 AM

found it

cat fileName | -n 20

will do the job

kshkid · 03-11-2007, 01:41 AM

Quote:

Originally Posted by ovince

found it

cat fileName | -n 20

will do the job

Am afraid, is there any command like '-n'

I dont think so?

kshkid · 03-11-2007, 01:43 AM

in awk

Code:

awk '{ printf "%s ", $0; getline; printf "%s \n", $0 }' filename

kshkid · 03-11-2007, 01:49 AM

in sed

Code:

sed 'h; N; h; {s/\n/ /g}' filename