if your files (sets) are not huge, you can handle this easily using R, without resorting to Perl:

# read data in
a = read.table("~/tmp/a.txt",header=T)
b = read.table("~/tmp/b.txt",header=T)

# merge datasets
merge(a,b,by="id")
  id value1.x value2.x value3 value1.y value2.y
1 g1       v1       v2   good       v1       v2


# extract results
c = merge(a,b,by="id")
res=c[,c(1,6,5,4)]
res
  id value2.y value1.y value3
1 g1       v2       v1   good

# produce output
write.table(res, file="~/tmp/c.txt",quote=F)

R is more interactive - easier to re-run and take care about other things (comparison constraints, etc), plus it allows to run script in batch mode. But Perl is more streamlined and command-line oriented if you after that.

There are many ways to do this join task. To avoid small bugs from writing your own custom code, and the probable inefficiencies of a naive implementation, you could use the join utility on a Linux system. It should do what you want.

$ join -j 1 -t " " a.txt b.txt
id value1 value2 value3 value1 value2
g1 v1 v2 good v1 v2

You can use dplyr in R as well.

a = read.table("~/tmp/a.txt",header=T)  
b = read.table("~/tmp/b.txt",header=T)

library(dplyr)  
inner_join(a, b)

This can be done in two steps: 1) split each line of a.txt and build a hash, where the id is a key and value3 is the key's associated value, and 2) split each line of b.txt and print the line plus value3 if the current id exists as a key in the hash:

use strict;
use warnings;

my ( @cols, %hash );

while (<>) {
    $hash{ $cols[0] } = $cols[-1] if @cols = split;
    last if eof;
}

while (<>) {
    print +( join "\t", @cols, $hash{ $cols[0] } ), "\n"
      if @cols = split and exists $hash{ $cols[0] };
}

Usage: perl script.pl a.txt b.txt [>outFile.txt]

The last, optional parameter directs output to a file.

The last if eof; is placed within the first while since both files would be completely read without it. The + after the print in the second while helps perl (the interpreter) disambiguate what to do.

This one-liner will produce the same results:

perl -ane '$h{$F[0]}=$F[-1];last if eof;END{while(<>){@F=split; s/\K$/\t$h{$F[0]}/ and print if$h{$F[0]}}}' a.txt b.txt [>outFile.txt]

Hope this helps!