comparing the contents of two files

[agilo]

Hi,

I want to compare the contents of two text files using Awk. Both files are sorted. I want to see which values changed, deleted or newly added.
E.g.,

file1:

x1 x2 x3 x4 x5 x6
name1 4 5 6 8 10
name2 10 33 44 55 32
..................................
..................................

file2:

x1 x2 x3 x4 x5 x6 x7
name1 4 5 6 10 10 13
name2 10 33 44 45 12
.......................................
.......................................

The comparison should show that the value of x5 changed from 8 to 10, also x7 is a new column in the second file, the value of x6 for name2 in the second file has been deleted.

any ideas are welcomed..

Thanks,

Agilo

 

[johnherman]

You can use the Unix function sdiff (side by side difference).

Sometimes the grass is greener on the other side because there is more manure there - original.

 

[SteveR77]

Agilo,

I am certain that there is a way to do this with awk, but have you considered using the diff command to do the file comparison?

The way I would approach doing what you desire with awk would be to read the lines from each of the files then do a field comparison checking for the values not being equal.

While I am sure this method would work there is a lot of overhead for reading and comparing the files this way.


Hope this helps,

Steve

 

[CaKiwi]

Here's something to get you started. The spacing of the columns was different in the sample data you posted so I adjusted them to be 6 characters each. You may have to change the substr statements to reflect your actual data.

BEGIN {
if (!fn) fn="file3"
while ((getline < fn) > 0) a[++n] = $0
close(fn)
}
{
if (NR==1) {
n = split(a[1],b)
if (NF != n) print "different number of cols"
next
}
for (j=1;j<n;j++) {
a1 = substr($0,j*6,6)+0
a2 = substr(a[NR],j*6,6)+0
# print a1,a2
if (a1 != a2) print "your error message here"
}


CaKiwi

 

[Ygor]

This uses an array to hold all the values, then does a comparison.
[tt]
awk 'BEGIN {
FS=" {1,9}"
}
FNR==1{
afn[++fc]=FILENAME
split($0,acn)
for(x in acn) {
cn=acn[x]
ucn[cn]=cn
}
next
}
{
for(x=2;x<=NF;x++) {
cn=acn[x]
av[FILENAME,$1,cn]=$x
urn[$1]=$1
}
}
END {
for(z in urn)
for(y in ucn) {
f1=afn[1]
f2=afn[2]
if (av[f1,z,y] != av[f2,z,y])
printf "%s %s\t %s:%6s\t %s:%6s\n",
z, y, f1, av[f1,z,y], f2, av[f2,z,y]
}
}' File1 File2
[/tt]
...tested on the sample data...
[tt]
name1 x5 File1: 8 File2: 10
name1 x7 File1: File2: 13
name2 x5 File1: 55 File2: 45
name2 x6 File1: 32 File2:
name2 x7 File1: File2: 12[/tt]

 

[agilo]

Thanks very much guys,

the script from Ygor does what I need..

Agilo