nysus has asked for the wisdom of the Perl Monks concerning the following question:

I checked the errata pages at O'Reilly but saw no mention of what I think might be an error on page 224 of "Programming Perl, 3rd Edition". More specifically, I'm talking about the following code: 
@common = inter ( \%foo, \%bar, \%joe );
sub inter {
    my %seen;
    for my $href (@_) {
        while (my $k  = each %$href ) {
            $seen{$k}++;
        }
    }
    return grep { $seen{$_} == @_ } keys %seen;
}
=======================================================

[download]
When I tried printing @common using the following data: 
__DATA__
my %foo = (
    'one' => 1,
    'two' => 2,
    'three' => 3,
);

my %bar = (
    'two' => 5,
    'four' => 6,
);

my %joe = (
    'three' => 7,
    'four' => 8,
    'five' => 89,
);

[download]
@common came back null for me. However, when I changed this line:

return grep { $seen{$_} == @ } keys %seen;
to this:
return grep { $seen{$_} } keys %seen;

@common returned: 

three five one two four

[download]
which judging by the description in the book, appears to be what the code is supposed to yield. I find it hard to believe I would catch an error like this. Can someone please back me up on this?

$PM = "Perl Monk's";
$MCF = "Most Clueless Friar Abbot";
$nysus = $PM . $MCF;

Edit by tye

Replies are listed 'Best First'.
Re: Code mistake in Camel book?
by wog (Curate) on Jun 26, 2001 at 08:42 UTC
    The code appears to be fine. It takes a bunch of hashes and returns all the keys that they have in common. In your example data there are no keys in common, so it correctly returns nothing. Your modification makes it return all the keys in any of the hashes, probably not what is wanted.

    update: I am assuming that inter implies intersection here. I don't have a copy of the Camel third edition personally.

    update 2: ... and that assumption is correct.

[reply]
[d/l]
      Ah, ok. I see now. The book reads: "Here's how you might write a function that does a kind of set intersection by returning a list of keys occurring in all the hashes passed to it..."

      The part that says "by returning a list of keys occurring in all the hashes passed to it..." led me to think "intersection" meant "union". It's been a while since I've had set theory. Still, "returning a list of keys common to each of the hashes passed to it" would be a lot less ambiguous.

      Thanks!

      $PM = "Perl Monk's";
      $MCF = "Most Clueless Friar Abbot";
      $nysus = $PM . $MCF;

[reply]
        But it is still worth sending this in as a suggestion. Often the hardest part of writing material which is meant to be read by people of different backgrounds is figuring out when language will be potentially unclear to your target audience. While anyone with a math background would pick up what "all" was supposed to mean from the word "intersection", the description of intersection itself is ambiguous and shouldn't be.

        However the usual way of talking in math would suggest one of the following two:

        • by returning a list of keys that occur in every hash passed to it.
        • by returning a list of keys that occur in all hashes passed to it.
        I believe that these should be clear to people without familiarity with the math, and also be clearer to people for whom English is a second language.
[reply]

Back to Seekers of Perl Wisdom