UNIX for Dummies Questions & Answers

Hi,
I have a file with 40 columns out of which 15 are amount fields. There are approximately 6 mn records in this file.
The file has data in following format:

Code:

123A,Ank,00.468,US,IL,780,53489
253A,Tng,-00.456,US,CA,452,46781
363A,nkk,-00.023,US,NJ,539,09625

I need to take all amount fields and do some logic where the last digit of each amount field is changed to a character based on the digit value.

So amount fields from above line should be converted like this
00.468 --> 0046T
-00.456 --> 0045K
-00.23 --> 002L
780--> 78}
452 --> 45B
539 --> 53X

My output file should be like this

123A,Ank,0046T,US,IL,78},53489
253A,Tng,0045K,US,CA,45B,46781
363A,nkk,002L,US,NJ,53X,09625
..

I was able to get the logic for converting one number at a time i.e I have coded a script which will take in a number and based on rules defined will give an output. e.g: if input to script id 780 it will give output 53X

What I need to know is how do I give this file as an input to my script, so that only the amount fields are changed and my output file format is maintained

Regards
Wah

You could do something like

Code:

cut -d ',' -f 3 data-file

to get the amount fields, and use a for loop around them.

I find this sort of icky text processing much easier in Perl, though; e.g.

Code:

# letter substitutions (incomplete)
@to_letter = ('}', '1', 'B', 'L', '4', '5', 'K', '7', 'T', 'X');
while(<>)
{
   split ',';
   for($_[2])
   {
     s/^\-//;    # remove -
     s/\.//;     # remove .
     s/(\d)$/$to_letter[$1]/;   # substitute letter for the last digit
   }
   print join( ',', @_);
}

which you would run with the data file name as the command-line argument.