command paste with variables

Shell Programming and Scripting
21

You must be using a really old shell since the newer Korn and POSIX shells have the coprocess feature built into them.

22

samos, what's the OS you're working on?

23

CentOS

24

could just anyone explain me how to store the variable a in file open by exec? because there is a function paste which do everything i need but only with files...all would be solved if i can put variables somehow in the files (which a canot create) and then use paste...

25

The paste command works with std. input as well.

26

Try this one out...

echo "$var1"
var1line1
var1line2
var1line3
var1line4

echo "$var2"
var2line1
var2line2
var2line3

echo "$var1" | awk '{print $1;print a}' a="$var2" | paste - - - - | awk '{print $1, $(NR+1)}'
var1line1 var2line1
var1line2 var2line2
var1line3 var2line3
var1line4
27

Or, as I already stated, with process substitution if your shell supports it:

$ printf '%s\n' "$v1"
var1line1
var1line2
var1line3
var1line4
$ printf '%s\n' "$v2"
var2line1
var2line2
var2line3
$ paste <(printf '%s' "$v1") <(printf '%s' "$v2")
var1line1       var2line1
var1line2       var2line2
var1line3       var2line3
var1line4       
$

Assuming default CentOS installation the code above will work with bash.

28

yout code is working but only for lines without spaces...
e.g. var1='asd asd
fasd asd'
var2='das asd
f a '
result is:
asd das
fasd f

it concatenate only first words before space...

29

to radulov:
your code is notworking with print but with echo it worked but only in command line.
when i put it into script error occures with: unexpected token `('

30

I don't use print in my example, but printf. If you want that command to work in a script, you should use the bash shell:

bash <your_script>

or point the shebang to the correct interpreter:

#!/bin/bash
31

TIMTOWTDI

echo "$var1"
v1-1
v1-2
v1-3
v1-4
v1-5
v1-6
v1-7

echo "$var2"
v2-1
v2-2
v2-3
v2-4
v2-5
v2-6
v2-7
v2-8
v2-9

echo "$var1\n$var2" | \
awk '{
   if (NR <= a)
      x[NR] = $0
   else
      print x[++i], $0
} END {
   for (i++; x; i++)
      print x
}' a=$(echo "$var1" | wc -l)
32

Yep:

awk 'BEGIN { while ( "unix.com" ) {	
  getline _ < ARGV[1] ||  (_ = "    ") && ++no
  getline   < ARGV[2] || ($0 = "    ") && ++more
  if (no && more) break
  print _, $0 }
}' <(printf '%s' "$v1") <(printf '%s' "$v2")

For example:

% print $v1
v1-1
v1-2
v1-3
v1-4
v1-5
v1-6
v1-7
% print $v2
v2-1
v2-2
v2-3
v2-4
v2-5
v2-6
v2-7
v2-8
v2-9
% awk 'BEGIN {
  while ( 42 ) {
  getline _ < ARGV[1] ||  (_ = "    ") && ++no
  getline   < ARGV[2] || ($0 = "    ") && ++more
  if (no && more) break
  print _, $0
    }
}' <(printf '%s' "$v1") <(printf '%s' "$v2")
v1-1 v2-1
v1-2 v2-2
v1-3 v2-3
v1-4 v2-4
v1-5 v2-5
v1-6 v2-6
v1-7 v2-7
     v2-8
     v2-9
% awk 'BEGIN {
  while ( -9 ) {
  getline _ < ARGV[1] ||  (_ = "    ") && ++no
  getline   < ARGV[2] || ($0 = "    ") && ++more
  if (no && more) break
  print _, $0
    }
}' <(printf '%s' "$v2") <(printf '%s' "$v1")
v2-1 v1-1
v2-2 v1-2
v2-3 v1-3
v2-4 v1-4
v2-5 v1-5
v2-6 v1-6
v2-7 v1-7
v2-8     
v2-9
33

Ok, try this new version, it must work much faster than the precedent version

#!/bin/ksh

IFS='
'

FINAL_RESULT=""

VARIABLE1="var1 first line
var1 second line
var1 third line
var1 forth line
var1 fifth line"

VARIABLE2="var2 first line
var2 second line
var2 third line
var2 forth line
var2 fifth line"

CURRENT_LINE_NUMBER=1
for ITERATOR in $VARIABLE1
do
    CURRENT_LINE_IN_VARIABLE2=$(print "$VARIABLE2"\
    |head -n $CURRENT_LINE_NUMBER|tail -n 1)    
    (( CURRENT_LINE_NUMBER = CURRENT_LINE_NUMBER + 1 ))
    FINAL_RESULT="$FINAL_RESULT\n$ITERATOR$CURRENT_LINE_IN_VARIABLE2"
done

(( FINAL_LENGTH = ${#FINAL_RESULT} - 2 ))
FINAL_RESULT=${FINAL_RESULT:2:FINAL_LENGTH}
print "$FINAL_RESULT"

:smiley: