Shell Programming and Scripting

I have to replace a field in one file with a field from other file.

I came across this awk command to replace a field with one string
nawk -F'|' -v OFS='|' '$2="replace"' temp2 > temp3

I need to have something like cut -f2 -d "|" temp1 (a field from other file) instead of 'replace'

Is this possible in the above command or something else needs to be used??

The command you have does not have a way to determine which line in the other file to get field2 from in order to insert into temp3. If you have a list of data you want to replace field2 with in temp2 then you can execute a for loop to complete your task.

Code:

list.txt contains

john
sally
mike

for name in $(< list.txt); do
    nawk -F'|' -v OFS='|' '$2="$name"' temp2 > temp3
done

An alternative would be the file you want field2 from has the same number of lines and is pipe delimited just like file temp2.

Code:

IFS="|"
cat file_with_field2 | while read field1 field2 rest ; do
    nawk -F'|' -v OFS='|' '$2="$field2"' temp2 > temp3
done

Else you need to provide more information about the content of file1 and file2 along with the desired output.

Thanks ldapswandog..

I tried both the scripts.. the required field is getting replaced with $name/$field2..

For script

for name in $(<temp4); do
nawk -F'|' -v OFS='|' '$2="$name"' temp2 > temp3
done

Files
temp4
gg
re
tt
vv
qq

temp2
11|22|33|44|zz
11|22|33|44|zz
11|22|33|44|zz
11|22|33|44|zz
11|22|33|44|zz


outputf file(temp3)
11|$name|33|44|zz
11|$name|33|44|zz
11|$name|33|44|zz
11|$name|33|44|zz
11|$name|33|44|zz

required output
11|gg|33|44|zz
11|re|33|44|zz
11|tt|33|44|zz
11|vv|33|44|zz
11|qq|33|44|zz


My requirements
===================

file1
11|22|33|44|zz
11|22|33|44|zz
11|22|33|44|zz
11|22|33|44|zz
11|22|33|44|zz


file2
aa|bb|cc1|dd|55
aa|bb|cc2|dd|55
aa|bb|cc3|dd|55
aa|bb|cc4|dd|55
aa|bb|cc5|dd|55


required output
11|22|cc1|44|zz
11|22|cc2|44|zz
11|22|cc3|44|zz
11|22|cc4|44|zz
11|22|cc5|44|zz


I need to replace the 3rd field of file1 with 3rd field of file2 for all records(1st record should be replaced with 1st record only of the other file)
Both files have same number of lines as well as fields.

Try this:

Code:

awk -F"|" 'NR==FNR{a[NR]=$3;next}
{$3=a[FNR]}
1' OFS="|" file2 file1

Regards

cool.. it works...

Thanks so much Franklin..

I have never been into awk programming.. it would be great if you would let me know how this works.. just a small explanation.. this makes use of arrays??

Code:

awk -F"|" 'NR==FNR{a[NR]=$3;next}
{$3=a[FNR]}
1' OFS="|" file2 file1

-F"|" -> set fieldseparator

NR==FNR -> if we read the first file (file2)

a[NR]=$3;next -> set array a[NR] with the value of the 3th field and read the next line


The next lines of the code affect the 2nd file (file1):

{$3=a[FNR]} -> replace the 3th field with the content of a[FNR]

1 -> print the current line of the buffer


Have a read of this tutorial regarding the built-in variables NR and FNR:

Awk - A Tutorial and Introduction - by Bruce Barnett

Regards

Thanks again Franklin...