shell - compare two .csv files using awk\ unix -

file1:

sa, 5006, 12, , dj cn, bn, , bbb, 13 22, 67, gg, ff, 88 33, bb, aa, cc, 22 

file2:

sa, 5006, 12, 15 , dj cn, bn, , bbb, 13 empty line 33, cc, aa, dd, 22 

output:

sa, 5006, 12, 15 , dj, unmatch, 4 cn, bn, , bbb, 13, match empt, empt, empt, empt, empt, unmatch, 12345 33, cc, aa, dd, 22, unmatch, 24 

i need compare 2 .csv files line line, of field/lines can empty , output should in file3: 5 columns form file 2, match\unmatch, unmatch fields this:

c1, c2, c3, c4, c5, match/unmatch, concatenation of digits representing unmatch fields. 

i try new awk can help? :)

code use, think problem empty fields anf dont know how can print :

 ##set input , output field separators ':'. begin {     fs = ofs = ":" }   nr == fnr {     ## save line in array, lines saved like:     ## c1::c2::c3::c4::c5      ++a[$0]      ## process next line beginning.     next }  ## every line of second file. {       ## search line in array, if not exists means field     different       ## print line.     if ( !a[$0] ) {             $6 = "same"             print     }else {    $6 = " not same"             print } } 

you need use line number index of array save between files, can compare corresponding lines in 2 files.

begin { fs = ", "; } nr == fnr { a[fnr] = $0 } # in first file, save each line in array nr != fnr { if (a[fnr] == $0) { # compare line in 2nd file corresponding line in first file                 $6 = "match";             } else {                 $6 = "unmatch";                 split(a[fnr], b); # split fields first file                 $7 = ""                 (i = 1; <= 5; i++) { # compare each field                     if ($i != b[i]) { $7 = $7 i; } # add non-matching field numbers output                 }             }             print;         }