Shell Programming and Scripting

Greetings all,

i have csv file with pipe separated columns

SSN|NAME|ADDRESS|FILLER
123|abc|myaddress|xxx
234|BBB|my
add
ress
broken up|yyy

In the example above, the second record is broken into multiple lines. I need to keep going until I find a "|" since this issue is with the non-last column and therefore there definitely will be a pipe at the end of that column text.

is there any way i can remove \n (newline) from the address column?

Thanks for the help.

Try:

Code:

awk -F"|" 'NF<4{ORS=" ";p=1}NF==1{p=1}NF==4&&p{printf "\n";ORS="\n";p=0}1' file

meh. bartus11 beat me to it and his is more elegant.

Thanks guys. I guess as long as I know the number of columns in the record, I can tweak this awk code to strip out linefeeds embedded in columns.

If my number of columns is in a shell variable num_cols, how do I reference that from this awk code?

I guess this should work:

Code:

awk -F"|" -vn=$num_cols 'NF<n{ORS=" ";p=1}NF==n{p=1}NF==4&&p{printf "\n";ORS="\n";p=0}1' file

I get

nawk: can't open file NF<n{ORS=" ";p=1}NF==1{p=1}NF==n&&p{printf "\n";ORS="\n";p=0}1

---------- Post updated at 03:23 PM ---------- Previous update was at 03:22 PM ----------

works on command line though