CSV - How to add columns based on an existing column? - bash

Question

What is the best way to do this and how?
I gather things called sed, AWK and bash may be relevant.
I have used AWK once for one command, the others never.
I have searched and other apparently similar questions do not have an answer I need.
I have columns I have called fields in a CSV file:
_________________________
field1 | field2 | field3|
-------------------------
1990AB | 123456 | 123456|
-------------------------
I want to add fields based on these three original fields to appear as follows:
_______________________________________________________
field1 | field2 | field3 | field1a | field2a | field3a |
-------------------------------------------------------
1990AB | 123456 | 123456| 1990 | 12345 | 12345 |
-------------------------------------------------------
where:
field1a 1990 column 1 first 4 always digits then alpha
field2a 12345 column 2 is always 6 digits
field3a 12345 column 3 is always 6 digits
These are one-time-per-file actions, prior to database import.
macosx has about 6 million records. 2nd attempt at this question as my first was apparently not good. In this area I am a 100% novice.

Answer 1

awk to the rescue!
this should be easy to read even if you have no prior experience with awk
$ awk -F, -v OFS=, 'NR==1 {for(i=1;i<=3;i++) $(++NF)=$i"a"}
NR>1 {$(++NF)=substr($1,1,4);
$(++NF)=substr($2,1,5);
$(++NF)=substr($3,1,5)}1' file
NR is line number, special treatment for header, NF is number of fields, here incrementing for each additional column and $i is field value at position i. The last 1 is shorthand for printing the line. Initial options are for setting input field delimiter (F) and output field delimiter (OFS) to comma.