in reply to Re^5: Pre vs Post Incrementing variables
in thread Pre vs Post Incrementing variables

The argument list to print is evaluated from left-to-right. ++${\++$i} is merely incrementing $i twice and returning an alias to $i.
$i = 0; print ++${\++$i}, 0+ ++$i, 0+ ++$i, ++${\++$i};;
              ^              ^        ^     ^
             (1)            (2)      (3)   (4)

[download]

(1) & (4) return aliases to $i. Since $i has been incremented six times, we get 6 for both of them.

(2) & (3) return copies to $i because of the zero additions 0+. We get 3 for (2) after three increments, and 4 for (3) with one more increment.

Replies are listed 'Best First'.
Re^7: Pre vs Post Incrementing variables
by BrowserUk (Patriarch) on Sep 13, 2010 at 05:20 UTC

    You appear to be saying that the ability to describe what the implementation does, justifies why it does it that way.

    The fact that this:

    $i=0; print $n = ++$i, $m = ++$i;;
1 2

    [download]
    produces a different result to this:
    $i=0; print ++$i, ++$i;;
2 2

    [download]

    Is just plain weird. Some, including me, would say 'broken'.

    Justifying the weirdness by saying "that's what it does", kinda reminds of Apple's "You're holding it wrong." explanation of Antennagate.

[reply]
[d/l]
[select]
      No, I made no implied justification thus far.

      The root of my reply was merely to reveal my experimentation with assigning the return of (++$i) to some value (since I also ran into the same error message you got).

      Then, I attempted to explain the snippet you posted so that others may find it helpful to see what's going on.

      My opinion? Yeah, the lvalue stuff is confusing and I don't quite see its usefulness. I would expect the behavior the OP expected.
[reply]
[d/l]
        I made no implied justification

        Then I apologise for misunderstanding your point.

        When I said: "Mind you, it might make explaining this one a tad awkward :)", I meant explaining the logic behind it, rather than the implementations behaviour.

        Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
        "Science is about questioning the status quo. Questioning authority".
        In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]

In Section Seekers of Perl Wisdom