marge line

vaskarbasak

test.text:-
--------
601215856422|COM|0|0|CD|1|3.99
601215868029|COM|0|0|CD|1|3.99
801472092322|COM|0|0|CD|2|10.98
802097013822|COM|0|0|CD|2|10.98
601215856422|MOM|0|0|CD|1|3.99
601215856422|RNO|0|0|CD|1|3.99
601215856422|SOM|0|0|CD|1|3.99

..........

I have to do
601215856422|COM|0|CD|1|3.99|MOM|0|CD|1|3.99|RNO|0 |CD|1|3.99
....

using shell script

pls help me.

thanks!
vaskar

kaarthikeyapreyan

Quote:

Originally Posted by vaskarbasak

test.text:-
--------
601215856422|COM|0|0|CD|1|3.99
601215868029|COM|0|0|CD|1|3.99
801472092322|COM|0|0|CD|2|10.98
802097013822|COM|0|0|CD|2|10.98
601215856422|MOM|0|0|CD|1|3.99
601215856422|RNO|0|0|CD|1|3.99
601215856422|SOM|0|0|CD|1|3.99

..........

I have to do
601215856422|COM|0|CD|1|3.99|MOM|0|CD|1|3.99|RNO|0 |CD|1|3.99
....

The code is bit complicating but i have provided you flexibility to do a lot of things,I also believe there are a lot more simpler ways to do it

Expand|Select|Wrap|Line Numbers

 a=601215856422

filelengthmain=`wc -l input.txt |cut -d ' ' -f1`

val1=1

val1=`expr $val1 + 0`

for ((j=1; j<=$filelengthmain; j++))

do

ln=`head -$val1 input.txt|tail -1`

val=`echo $ln | cut -d '|' -f 1`

if test $a = $val

then

fvals[$val]=`echo $ln | grep -e $a |grep -v SOM | cut -d '|' -f 2-3,5-7`

b=`echo $b"|"${fvals[$val]}`

fi

val1=`expr $val1 + 1`

done

echo $a$b

Output :

601215856422|COM|0|CD|1|3.99|MOM|0|CD|1|3.99|RNO|0 |CD|1|3.99|

gpraghuram

You can use this script

Expand|Select|Wrap|Line Numbers

 
FULL_LINE=""
 
while read LINE

do

  if [ "$FULL_LINE" == "" ]

  then

      FULL_LINE="$LINE"

  else

    FULL_LINE=$FULL_LINE"|"$LINE

  fi

done < "ip.txt" 
 
echo "FULL_LINE is : ${FULL_LINE}"

Raghuram

radoulov

If I understand correctly:

Expand|Select|Wrap|Line Numbers

 awk '{ 

for (f=2; f<=NF; f++)

  if (f != 3)

    x[$1] = x[$1] FS $f

} END {

for (k in x)

  print k x[k]

}' OFS='|' FS='|' file

oberon

Here is a one-liner for the same job:

Expand|Select|Wrap|Line Numbers

 cat test.text | echo -n $(awk {'print $0 "|"'}) | sed -e 's/\ *//g'
 

The -n is to remove newline, and the last sed is to remove a whitespace that would otherwise sneak in.

vaskarbasak

Thanks all!

601215856422|COM|0|0|CD|1|3.99
601215868029|COM|0|0|CD|1|3.99
801472092322|COM|0|0|CD|2|10.98
802097013822|COM|0|0|CD|2|10.98
601215856422|MOM|0|0|CD|1|3.99
601215856422|RNO|0|0|CD|1|3.99
601215856422|SOM|0|0|CD|1|3.99

but i have to do

601215856422|MOM|1|3.99|RNO|1|3.99|SOM|1|3.99|COM| 1|3.99
801472092322|COM|0|0|CD|2|10.98
601215868029|COM|0|0|CD|1|3.99
.....

the file size is very large nearly 91 MB.So it takes lot of time.But i have to do it with in 5 min.

Still i am trying to solve the problem.

please help me...

oberon

Quote:

Originally Posted by vaskarbasak

601215856422|MOM|1|3.99|RNO|1|3.99|SOM|1|3.99|COM| 1|3.99
801472092322|COM|0|0|CD|2|10.98
601215868029|COM|0|0|CD|1|3.99

Your last post didn't really clearify what you are looking for. If you could please add a few lines of explanation, then you would have a greater chance of getting the right answer.

vaskarbasak

i need concate all duplicate lines like

794051400123|COM|24|MOM|62|SOM|25|RNO|73
794051413727|COM|11||MOM|4|RNO|8
............

and save it another file..

gpraghuram

Quote:

Originally Posted by vaskarbasak

i need concate all duplicate lines like

794051400123|COM|24|MOM|62|SOM|25|RNO|73
794051413727|COM|11||MOM|4|RNO|8
............

and save it another file..

The first part of the string that is the 794051400123 is the token with which you find whether it is a duplicate token?

Raghuram

marge line

re: marge line

re: marge line

re: marge line

re: marge line

re: marge line

re: marge line

re: marge line

re: marge line

Similar Unix / Linux / BSD bytes

/bytes/development

/bytes/it

/bytes/about