Beefy Boxes and Bandwidth Generously Provided by pair Networks
Clear questions and runnable code
get the best and fastest answer
 
PerlMonks  

tie(...) bug or select(...) bug?

by CardinalNumber (Prior)
on Oct 04, 2013 at 18:22 UTC ( #1056913=perlquestion: print w/ replies, xml ) Need Help??
CardinalNumber has asked for the wisdom of the Perl Monks concerning the following question:

While trying to clear up a few older bugs in a module I'm planning to take over, I found one I can't seem to wrap my head around. Take this:

use Tie::Scalar; tie (my $SLEEP_TIME, 'Tie::StdScalar', 0.25); select (undef, undef, undef, $SLEEP_TIME);

Tie::StdScalar is part of Tie::Scalar and does virtually nothing so parsing that mentally should result in a .25s sleep and then exit. But it doesn't. The code blocks as if $SLEEP_TIME was undefined. However, this...

use Tie::Scalar; tie (my $SLEEP_TIME, 'Tie::StdScalar', 0.25); print $SLEEP_TIME; select (undef, undef, undef, $SLEEP_TIME);

...works exactly as expected. The value of $SLEEP_TIME is printed and then select(...) sleeps a quarter second. According to Devel::Trace, my first example never calls the tied scalar's FETCH method but in my second example, it's correctly called twice: once for the print and again in the select statement.

I only have perl v5.16.3 installed so I'm not sure if this is a newly introduced bug or an old one. Before I bother fending off the dragons that inhabit p5p, anyone else have an idea or see this same behavior with newer/older versions of perl?

Update: Reported as RT#120102.

Comment on tie(...) bug or select(...) bug?
Select or Download Code
Re: tie(...) bug or select(...) bug (neither)?
by Anonymous Monk on Oct 04, 2013 at 18:32 UTC

    neither, read select, int $SLEEP_TIME is zero, which means block

      Yeah... not true and not the issue here. I could put tie(my $SLEEP_TIME, 'Tie::StdScalar', 5); and my first example would still block indefinitely. select(...) isn't correctly working with tie()'s magic to get the right value of $SLEEP_TIME and instead figures it's undef which is supposed to block. That's the bug I think I'm seeing.

Re: tie(...) bug or select(...) bug?
by dave_the_m (Parson) on Oct 04, 2013 at 19:42 UTC
    It's a bug in select(), and is still present in blead. Looks like select() isn't calling get magic on the timeout arg.

    Here's another example of using a magic var which fails unless its magic value has been previously retrieved (using the commented-out line):

    "5" =~ /(\d)/; #$x = $1; select (undef, undef, undef, $1);
    You may want to perlbug this.

    Dave.

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: perlquestion [id://1056913]
Approved by keszler
Front-paged by toolic
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others meditating upon the Monastery: (10)
As of 2014-09-16 03:48 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    My favorite cookbook is:










    Results (155 votes), past polls