So, I finally bought a copy of HOP. In it, the author uses anonymous subs as iterators. The details of this aren't too important for this discussion. He then introduces a function called igrep, which applies a filter to an iterator, much like grep, but one that is evaluated lazily, in the tradition of an iterator.

So, here is an example implementation:

# Syntax sugar
sub Iterator (&) { $_[0] }

# Takes a sub followed by an iterator
sub igrep (&$) {
   my ($check, $iterator) = @_;
   return Iterator {
       # While the source iterator is not exhausted
       while (defined( local $_ = $iterator->() )) {
            return $_ if $check->();
       }
       return undef; # we are exhausted
   }
}
[download]

The author later introduces some examples of iterators that return lists on each invocation, instead of scalars. To account for this, he starts using a hypothetical igrep_l function that works with lists instead of a scalar. Not hard to implement, but I feel it gets ugly to have different functions for different contexts, especially iterators whose return value is dependant on context.

So, finally the question: how can this igrep function effectively propagate the calling context to the iterator? Here is something that I cobbled together. I was wondering if anyone had any simpler ideas. Note: the sub passed to igrep will have @_ as the param (in a scalar context) or params (in a list context) just retuned from the iterator. $_ is aliased to the first element returned by the iterator for the simple cases where an iterator is always called in a scalar context, or always returns a scalar.

#!perl -l

sub Iterator (&) { $_[0] }

sub igrep (&$) { 
    my ($check, $iterator) = @_;
    sub {
        local *_;
        my @data;
        my $wantarray = wantarray;
        while(@data = ($wantarray ? $iterator->() : scalar $iterator->
+())) {
            *_ = \ $data[0];
            return $wantarray ? @data : $data[0] if $check->(@data);
        }
        return undef;
    }
}
[download]

So, it seems like there is a lot of checks to wantarray here. An alternative would be to have a giant if (wantarray) dividing the sub into two implementations.

BTW, while reading this book made me think to ask it, I have actually had this problem come up in other domains. I can't remember where I have done before, but I do remember using a similar strategy.

Some other notes: I have intentionally not used strict or warning for the sake of brevity and clarity in these examples. Also, for the same reasons, I am not worried about the void context.

Well, I hope this question has interested more people than it has bored. :-)

---

---

Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!

Titles consisting of a single word are discouraged, and in most cases are disallowed outright.

Read Where should I post X? if you're not absolutely sure you're posting in the right place.

Please read these before you post! —

• How do I compose an effective node title?
• How do I post a question effectively?
• Markup in the Monastery

Posts may use any of the Perl Monks Approved HTML tags:

a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr

You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)

	For:		Use:
	&		&
	<		&lt;
	>		&gt;
	[		&#91;
	]		&#93;

Link using PerlMonks shortcuts! What shortcuts can I use for linking?

See Writeup Formatting Tips and other pages linked from there for more info.