comparing two unsorted files

Question

I have two tab separated files (please see the examples below):

File 1

Java    RAJ
PERL    ALEX
PYTHON  MAurice

(and so on)

File 2

ALEX    3.4
SAM     8.9
PEPPER  9.0

Now, if for instance say ALEX is also found in file 2 (it is not for sure that ALEX will be found) I should have a third file looking like this:

PERL ALEX 3.4

The code should check for all the values in column 2 of file 1 in file2.

Any suggestions for a bash script?

Answer 1

You received excellent answers using join and awk, so I thought I's post a pure bash-one:

#!/bin/bash

declare -A name2prog
declare -A name2num

while read a b; do name2prog[$b]=$a; done < file1
while read a b; do name2num[$a]=$b; done < file2

for i in "${!name2num[@]}"
do
    if  [[ ${name2prog[$i]} ]]; then
        echo ${name2prog[$i]} $i ${name2num[$i]}
    fi
done

outputs:

$ ./try.sh
PERL ALEX 3.4

Answer 2

You want to use join for that. First you need to sort according to join field though:

join -1 2 -2 1 <(sort +1 -2 file1) <(sort +0 -1 file2)

Answer 3

awk 'NR==FNR {val[$1]=$2; next} $2 in val {print $0, val[$2]}' file2 file1

Answer 4

Is a one-liner with PERL also ok? Works without sorting.. Assuming your files are called f1 and f2..

perl -e 'open(F1, shift); open(F2, shift); $ls = $/;undef $/;$f2 = <F2>;$/ = $ls; while(<F1>) { ($t1, $t2) = $_ =~ /^(\w+)\s+(\w+)$/; if($t1) { ($t3) = $f2 =~ /^$t2\s+(.+)$/m; print "$t1 $t2 $t3 \n" if ($t3); } }' f1 f2

With f1:

Java RAJ

PERL ALEX

PYTHON Maurice

And f2:

ALEX 3.4
SAM 8.9
PEPPER 9.0

Results in:

PERL ALEX 3.4