Hi,
I have files with records of 40,00,000& 39,00,000 and i want to find out the

content

1.which is existing in file1 and not in file2.
2.Which is exisitng in file2 and not in file1.

The format of the file will be like

404ABCDEFGHIJK|CDEFGHIJK|1234567890|1


If its a smaller one i used to do egrep -f .

Need your help to sort out.

So... are these files keyed? Are they sorted? I presume that they're text files with variable length records... what are the line endings?

What counts as a match? Should it be case sensitive? Does it have to be the entire line, or is one of the fields acting as a primary key? If so, are there any escaping rules we should know about before trying to split the fields, e.g. don't split on | if it's preceeded by a \.