UNIX for Advanced & Expert Users

Hi
I have multiple files each with two columns and I need to combine all those file into a tab delimited file. (multiple entry with same name separated by a comma)

The content of the files are as follows:
---
file1.txt:

name var1
aaa xx
aaa gg
bbb yy
ddd zz
---
file2.txt

name var2
bbb mm
ccc nn
ddd pp
ddd qq
----

file3.txt

name var3
aaa uu
ddd vv
eee ww
fff tt
fff uu
---------

I need to combine all these file into a tab delimited file in such a way that the output looks like follows:

* \t shows tab delimited space.

I was wondering how to do this in unix. I can do the same in other language but because of efficiency issue I was looking to unix as a potential rescuer.

Thanks for ur help

look into 'man paste'

I have different number of lines in each file and the output I need is not as straight forward.

Code:

$ cat agg.awk
BEGIN { OFS = "\t" }
{
  if ($1 != last) {
    if (NR > 1) print last, a
    last = $1
    a = $2
  } else a = a ", " $2
}
END { print last, a }

$ join -t$'\t' -a1 -a2 -e- -j1 -o0,1.2,2.2 <(awk -f agg.awk f1) <(awk -f agg.awk f2) |
join -t$'\t' -a1 -a2 -e- -j1 -o0,1.2,1.3,2.2 - <(awk -f agg.awk f3)

Try this out

Code:

 
#!/bin/ksh
cat f1 f2 f3 > tmp
_list=$(cat tmp | awk '{print $1}' | sort | uniq)
for i in ${_list}
do
        _inf1=$(cat f1 | grep $i | awk '{print $2}' | awk 'BEGIN{ORS="\t"}{print $0}')
        _inf2=$(cat f2 | grep $i | awk '{print $2}' | awk 'BEGIN{ORS="\t"}{print $0}')
        _inf3=$(cat f3 | grep $i | awk '{print $2}' | awk 'BEGIN{ORS="\t"}{print $0}')
        print "${i}\t${_inf1}${_inf2}${_inf3}"
done

This code is incredible.
This works exactly what i wanted. Awesome.
Can you please explain the switches you used here especially join. Also if I have variable number of input files then how can I extend this.

* I never did awk programming but this works for me. I was using matlab to do this and is damn slow.
Thanks again.

---------- Post updated at 12:30 AM ---------- Previous update was at 12:28 AM ----------

This code is just concatenating one file after another. Did not work for me.
Thanks again

Options to join used here are:
-t$'\t' use tab as input and output field separator
-a1 -a2 include unmatched lines from the first and second file
-e- use a "-" for missing fields
-j1 use the first field as key in joining
-o0,1.2,2.2 output line consists of the join field (0), second field of the first file (1.2), second field of the second file (2.2)
I wish Unix has a built in command to do the task of agg.awk.
Let me know how the code performs (time wise).