in reply to Re: Threads and Shared Variables
in thread Threads and Shared Variables

Thanks, but I'm still unclear as to why this is happening. Since the $progref variable shares the same scrope with the sub, I don't understand why the sub can't see it. I've read the relevant section of perlsub, and as far as I can tell this should work.

Replies are listed 'Best First'.
Re^3: Threads and Shared Variables
by BrowserUk (Patriarch) on Jul 04, 2013 at 21:11 UTC
    and as far as I can tell this should work.

    It doesn't. And it won't whether you understand it or not.

    In Perl: DO NOT DEFINE SUBROUTINES WITHIN NESTED SCOPES!

    Weird bugs happen whether threads are involved or not. And when threads are involved, they can be even weirder. So don't do it.

    I could probably eventually explain why it doesn't work, but then I would have to think about it and probably do some experiments. But as there is no good reason to declare subs in async blocks; and many good reasons not to, I cannot be bothered to work out what does or does not work.

    With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
    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]
      If it's just a quirk then I can accept that, I was just trying to establish whether or not I missed something in the documentation (I can't find anything about not nesting subroutines). My reasoning behind doing this was not to pollute the package namespace. I am used to being able to do this in other languages (javascript).

      Thanks for the advice,
      Hermes
[reply]
        (I can't find anything about not nesting subroutines)

        See http://perldoc.perl.org/perlref.html#Function-Templates and read the three paragraphs where the first starts "Access to lexicals that change over time...".

        With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
        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]
        Apart from the experimental lexical subs introduced with 5.18, any named function created with sub (even those that are nested in an inner scope) become part of a package's namespace, so you can't avoid namesapce pollution in this way.
[reply]

In Section Seekers of Perl Wisdom