There must be some CPAN module available for this, but just becuase it seemed interesing, I implemented the following:

#!/usr/bin/perl

$howmany=3; # 3 lines before match, 3 lines after match
$match="match"; # What to search i.e. grep string
@buff=(); # Array that contains (($howmany*2)+1) elements max at any g
+iven time

$cLines=(($howmany*2)+1);

#Read first (($howmany*2)+1) lines into @buff
$count=0;
while(defined($line=<DATA>) && $count<$cLines)
{
   chomp($line);
    push(@buff,$line);
   $count++;
}

#start from @buff[0], go until index ($howmany)
for($idx=0;$idx<=$howmany and $idx <scalar @buff;$idx++)
{
    if($buff[$idx]=~/$match/)
   {
         print   "\n". "-" x 20;#Separator
          #Print $howmany lines before match
         for($jdx=0;$jdx<$idx and $jdx<scalar(@buff);$jdx++)
         {
                print "\n$buff[$jdx]";
         }

         print "\n $buff[$jdx] <---- match"; # print line that matches

         #Print $howmany lines after match
         for($jdx=$idx+1;$jdx<($idx+$howmany+1) and $jdx< scalar(@buff
+);$jdx++)
         {
                print "\n$buff[$jdx]";
         }
         print   "\n". "-" x 20;#Separator
   }
   else
   {
       print "\n$buff[$idx] <--- does not match";
   }
}

#Start 'shifting' one element out of the array until end of file
while(defined($line=<DATA>))
{
    chomp($line);
    shift @buff;
    push(@buff,$line);
   if($buff[$howmany]=~/$match/) #Always match element at index $howma
+ny
   {
       print "\n"."*" x 20;  #Separator
      #Print $howmany lines before match
        for($idx=0;$idx<$howmany;$idx++)
      {

          print "\n$buff[$idx]";
      }

         print "\n$buff[$idx] <----- match"; #Print line that matches

      #Print $howmany lines after match
        for($idx+=1;$idx<scalar(@buff);$idx++)
      {

          print "\n$buff[$idx]";
      }
       print "\n"."*" x 20;     #Separator
   }
   else
   {
       print "\n$buff[$howmany] <--- does not  match";
   }
}
print "\n". "=" x 20 . scalar(@buff). " elements";#Separator

#At end of file, there are <=(($howmany*2)+1) elements left in @buff
#BUt we have already seen $howmany elements in the while loop
#So start matching from $howmany+1
for($idx=$howmany+1;$idx<scalar @buff;$idx++)
{
   if($buff[$idx]=~/$match/)
   {
            print   "\n". "-" x 20;;
         #Print $howmany lines before match
         $left=($idx-$howmany);
         for($jdx=($left)>0?$left:0;$jdx<$idx and $jdx<scalar(@buff);$
+jdx++)
         {
            print "\n$buff[$jdx]";
         }

         print "\n $buff[$idx] <---- match"; # print line that matches

         #Print $howmany lines after match
         for($jdx=$idx+1;$jdx<($idx+$howmany+1) and $jdx< scalar(@buff
+);$jdx++)
         {
            print "\n$buff[$jdx]";
         }
         print   "\n". "-" x 20;;

   }
   else
   {
      print "\n$buff[$idx] <--- does not match";
         #shift @buff;
   }

}

__DATA__
first line contains match
second does not
third does not either
this third line contains one match
fourth one again matches
fifth
sixth
seventh
eighth
ninth
tenth
eleventh
enough! now match something
match this too
more match
ok
not any more
ENOUGH!!!!
no
no no no
nanananananana
hohhohohohohohp
lol
duck tecm
who are you and what nonsense is this
hahaha
match here!
[download]