Hi

any enlightenment will help .
I am trying to substitute a string with a replacement for the 2nd occurence of it for each line.

Here's my command:

$ awk -v x=$strA -v y=$strB -v z=$strC '{ if ( match($2,z) > 0 ) { sub( /x/,y,$2) } ; print }' abc.dat

It finds the match alright, but the substitution refuses to work and print faithfully prints every line in the
file unchanged.

I am very surprised !? !

output of the above command : abc.dat is printed out as-is !!

$ awk -v x=$strA -v y=$strB -v z=$strC '{ if ( match($2,z) > 0 ) { sub( /x/,y,$2) } ; print }' abc.dat
starthere:server=REP60_PROD userid=AAAA/aaaa@PROD field1=value1 field2=value2 field3=value3 field4=value4 andso on....

start2here:server=REP60_PROD userid=AAAA/aaaa@PROD field1=value1 field2=value2 field3=value3 field4=value4 andso on....


regards to the gurus,

a fellow human being

Remove the slashes from around the match in sub(). They treat the value inside of them as a regex string, so you're trying to match a literal 'x'. Without slashes it's treated as a variable containing the expression to be matched, which is what you want.


PS: It's best to put all but the most trivial code strings inside [code][/code] tags. It preserves formatting (of both the code and the page) and makes things easier to read.

If I am understanding you correctly, how about this:

sed 's/string/replacement/2' filename

replaces the 2nd instance of "string" with "replacement"

Thank you for reply. I was tempted to use sed. I am familiar with sed with 's/ / / ' . But I was keen on learning to awk this time. Appreciate the early response.

Yes absolutely. It worked ! Thank you for taking the time to reply to my post. I hope to keep posting here as a way of learning and keeping in touch. Thanks again.