Shell Programming and Scripting

I've got two files, both plain text. Each file is a datafeed of products, pipe delimited. The current file is in directory 1 and yesterday's file is in directory 2 (literally, those are the directory names). What I'm trying to do is compare the files and pull out products whose price has changed since yesterday and products that are new since yesterday. What I've come up with so far is a basic comparison to see if *anything* is different:

comm -23 1/datafeed.txt 2/datafeed.txt > 1/changes.txt

and that's a good start. It tells me, for example, when there's new products since yesterday, but it also tells me if they fixed a spelling error in one of the fields, which I don't really care about. Here's an abbreviated version of what these files look like:

Today's datafeed file:

ProductID|Name|Price|Color
84837423|Woot|2.50|Red
37238528|Blah|3.50|Blue
23452323|Rush|4.99|Black

Yesterday's datafeed file:

ProductID|Name|Price|Color
84837423|Woot|1.50|Red
37238528|Blah|3.50|Bluu

So, ideally, I'd like to have a newItems.txt file with the "Rush" line from today's datafeed and a changedPrices.txt file with the "Woot" line from today's datafeed. The "Blah" line, where they corrected a spelling error, doesn't interest me.

Thanks for any pointers you can give!

~Daniel

Absolutely perfect, thank you! I changed the file names, plugged it into my script and got exactly what I was looking for - and even better, I think I understand how it works. I love learning new things.

Thanks again!

Code:

#!/usr/bin/perl
open FH,"<old.txt";
while(<FH>){
	my @tmp=split("[|]",$_);
	$hash{$tmp[1]}->{PRC}=$tmp[2];
	$hash{$tmp[1]}->{DATA}=$_;
}
close FH;
open N_FH,">newitem.txt";
open C_FH,">changed.txt";
open FH,"<new.txt";
while(<FH>){
	my @tmp=split("[|]",$_);
	if (not exists ($hash{$tmp[1]})){
		print N_FH $_ ;
		next;
	}
	print C_FH $_ if $hash{$tmp[1]}->{PRC} ne $tmp[2];
}
close FH;