Shell Programming and Scripting

Hi, I have two files roughly 1200 fields in length for each row, sorted on the 2nd field. I need to compare based on that 2nd column between file1 and file2 and print lines that exist in both files into separate files (I can't guarantee that every line in file1 is in file2).

Example:

Code:

File1: 
1,12345,1,1,1...
1,12344,1,1,0...
1,12343,1,1,0...
1,12342,1,1,0...
...

Code:

File2:
1,12346,1,1,1...
1,12343,1,1,1...
1,12342,1,1,1...
...

Desired Output:

Code:

_File1:
1,12343,1,1,0...
1,12342,1,1,0...

Code:

_File2:
1,12343,1,1,1...
1,12342,1,1,1...

I need to run actual value comparisons on the other fields later, so I need separate files containing their matching lines. Only the 2nd field should be compared on, differences in other fields are perfectly fine. That 2nd field is unique, won't appear more than once per file. Alternatively, if it's simpler to remove lines from both files that don't match each other (to get the same desired output), that works too. Thanks in advance.

Edit: If it helps, that 2nd field is always 15 digits in length (never letters) and no other 15-digit field exists in either file besides that 2nd field

Here is one way using 2 nawk statements to match the input files, just the order of the input files are different:

Code:

nawk -F"," 'FNR==NR{a[$2]++;next}a[$2]' file1 file2
1,12343,1,1,1...
1,12342,1,1,1...

Code:

nawk -F"," 'FNR==NR{a[$2]++;next}a[$2]' file2 file1
1,12343,1,1,0...
1,12342,1,1,0...