Replace a field with line number in file

Shell Programming and Scripting
1

I am working on a script to convert bank data to a csv file. I have the format done - columns etc. The final piece of the puzzle is to change the second field (after the R) of every line to reflect its' line number in the file. I am stumped. I can use awk on each line but need help looping through the file.

Sample file below:

 
"R","","t0000123","","10/29/2013","10/29/2013","","","","100.00"

Any assistance greatly appreciated!

2

try:

awk '$2="\"" NR "\""' FS=, OFS=, infile
1 Like
4

in perl:

 perl -pe 's/""/$./' input

This will replace first "" one with current line number as per your input sample.

1 Like
5

Sed version

$ cat <<eof | sed -e 's/""/1/'
"R","","t0000123","","10/29/2013","10/29/2013","","","","100.00"
eof

"R",1,"t0000123","","10/29/2013","10/29/2013","","","","100.00"

OR

$ cat <<eof | awk '!x{x=sub(/""/,1)}1'
"R","","t0000123","","10/29/2013","10/29/2013","","","","100.00"
eof

"R",1,"t0000123","","10/29/2013","10/29/2013","","","","100.00"
6

@ akshay:
your solution will give always 1 as a replacement.

in sed we can try with

=

but i didnt succeed to replace in the required field.

7

@greet_sed try this awk

$ cat <<eof | awk 'x="";!x{x=sub(/""/,NR)}1' 
 "R","","t0000123","","10/29/2013","10/29/2013","","","","100.00"
"R","","t0000123","","10/29/2013","10/29/2013","","","","100.00"
eof

 "R",1,"t0000123","","10/29/2013","10/29/2013","","","","100.00"
"R",2,"t0000123","","10/29/2013","10/29/2013","","","","100.00"
8

That cannot work because = does not modify the pattern space. = writes directly to stdout and there is no way to capture its output to strip newlines from within the sed instance itself.

Regards,
Alister

1 Like
9

Various field separators:

awk '$2=NR' FS=\",\" OFS=\",\" file

awk '$4=NR' FS=\" OFS=\" file

--
JFF:

sed = file | sed 'N; s/\(.*\)\n\([^,]*,"\)/\2\1/'
10

Great! This did the trick:

awk '$2="\"" NR "\""' FS=, OFS=, infile

Now I need to add a comment string to a couple of fields for every line in the file. Different comment string for each field.

In below file, fields 7 and 8

"R","1","t0000684","","10/31/2013","10/31/2013","","","","750.23"

Thank you again.

11

try something like:

cmt1="this is comment 1"
cmt2="this is comment 2"
 
awk -v cmt1="$cmt1" -v cmt2="$cmt2" '
{$2="\"" NR "\"";
 $7="\"" cmt1 "\"";
 $8="\"" cmt2 "\""; } 1
' FS=, OFS=, infile