Oh, Most Monkilicious Monks:

I have a packed string of integers and wish to index into the string to extract a single integer at a particular offset. I use the 'index-by-size' (e.g., 'x[N]') and group-repetition features of unpack to do this. This works, but  unpack behaves differently in scalar versus list context:

>perl -wMstrict -le
"my $ps  = pack 'N*', 11, 22, 33, 44, 55, 66;
 my $upk = '(x[N])3 N';
 my $n1   = unpack $upk, $ps;
 print qq{n1 $n1};
 my ($n2) = unpack $upk, $ps;
 print qq{n2 $n2};
"
n1 22
n2 44
[download]

The index-by-size works as expected in scalar context, but the repeat count on the indexing sub-template group is just ignored. (The behavior of the example code is the same in ActiveState 5.8.9 and Strawberry 5.10.1, both running under Windows 7.)

I don't understand the reason for the difference in behavior in different contexts. To quote the perlfunc unpack doc (first paragraph, emphasis added):

unpack does the reverse of pack: it takes a string and expands it out into a list of values. (In scalar context, it returns merely the first value produced.)

It seems to me there should be no difference.

Can anyone explain? Is this a feature? Mis-feature? Bug?