UNIX for Dummies Questions & Answers

Hi,

I am looking for an awk script which should help me to meet the following requirement:

File1 has records in following format

Code:

INF: FAILEd RECORD AB1234
INF: FAILEd RECORD PQ1145
INF: FAILEd RECORD AB3215
INF: FAILEd RECORD AB6114
.........................
.........................

Each record(basically the ID 'AB1234', 'PQ1145' and so on) of File1, should be validated against the following File2, which has format like:

Code:

AB0001-AB1111
AB1112-AB2222
.............
.............
PQ1111-PQ2222
.............
XY1234-XY5678
.............

Note that File2 has different ranges for code AB, PQ, XY and so on, so validation won't be straight forward and hence code should be compared first and then the ID should check if it lies in any of the ranges above, associated to that particular code. If it does then the record should be filtered out from File1.

I'll really appreciate if anyone please help me out. Thanks

Wrong language, but Perl would handle this well...

So match the text part and see if the numeric part falls within any of the resulting ranges
I'm not sure how you want to perform your validation, so take a look at the following and feel free to ask if you hit a snag

Code:

~/tmp]$ perl -e '
open (RANGES,"File2"); 
while(<RANGES>){
   $ranges{$1}=[$2,$3] if (/^([A-Z]{2})(\d{4})-\1(\d{4})/);
}
while (<>){
   chomp;
   print "$1 $2 - ", $ranges{$1}[0],"=>",$ranges{$1}[1],"\n" if /\s([A-Z]{2})(\d{4})$/;
}' File1
AB 1234 - 1112=>2222
PQ 1145 - 1111=>2222
AB 3215 - 1112=>2222
AB 6114 - 1112=>2222

Thanks @Skrynesaver, but this requirement needs to fit in an existing shell script, thats why I need to write in a shell script only. Though I am able to do it through a simple shell scripting, but I want to optimise it by using awk which 'does more in less code'.
Appreciate your thoughts on this.