To be, or not to be, an object ?

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

Replies are listed 'Best First'.
Re: To be, or not to be, an object ? by moritz (Cardinal) on Jul 24, 2008 at 11:34 UTC
I usually use Scalar::Util::blessed. Never failed me so far. Would your approach work with autobox? Update: Your suggested solution has a (nearly obvious) pitfall: `sub Foo::isa { die "bla" }; my $ref = bless {}, 'Foo'; eval { $ref->isa() }; if ($@) { print "Not an object\n"; } else { print "An object\n"; }` [download] Also in general this construct: `eval { ... } if ($@) { ... }` [download] Isn't all that reliable because a `DESTROY` sub might clear `$@` accidentally. Rather use `if (eval {...; 1}){ ... }` [download] Instead (Ok, this is nit-picking, but you asked for reliability...).	[reply] [d/l] [select]
Re^2: To be, or not to be, an object ? by Bloodnok (Vicar) on Jul 25, 2008 at 13:32 UTC
Hi moritz , JIC you were wondering about the justification for this diversion, I'm working in an environment whereby the ad-hoc downloading and use (pun intended:-) of CPAN modules is a non-trivial exercise:-( Back OT, if the conditional were to change to ... `if ($@ =~ /can't call method without a package or object ref/) {...`, then the problem would, methinx, be circumvented unless there was a package perverse enough to `die "can't call method..."` in an overloaded `isa()`. BTW, wouldn't your latter point put at risk all that we're told regarding the use (pun unintended this time:-) of `eval` to catch errors and handle them locally - the risk of `DESTROY` f%*&^ing things up in such circumstances must be almost omnipresent... mustn't it ?? Just a thought... At last, a user level that overstates my experience ... as I'm beginning to most ably demonstrate :-))	[reply] [d/l] [select]
Re^3: To be, or not to be, an object ? by moritz (Cardinal) on Jul 27, 2008 at 20:18 UTC
Back OT, if the conditional were to change to ... `if ($@ =~ /can't call method without a package or object ref/) {...`, then the problem would, methinx, be circumvented unless there was a package perverse enough to die `"can't call method..."` in an overloaded isa(). Yes. Except that I always hated this kind of hack, because it relies on the exact phrase of the error message. Which might be subject to change, to l10n or whatever. BTW, wouldn't your latter point put at risk all that we're told regarding the use (pun unintended this time:-) of eval to catch errors and handle them locally - the risk of DESTROY f%*&^ing things up in such circumstances must be almost omnipresent... mustn't it ?? Yes, it is indeed. There was a lengthy thread on p5p about that, and how to fix it. Thankfully only very few modules actually use a eval inside a `DESTROY` block without localizing `$@` first. If you don't need a return value from the eval block, you can also do this check: `my $success = eval { # your stuff here... 1; };` [download] and then check for the truthness of `$success`.	[reply] [d/l] [select]
Re^4: To be, or not to be, an object ? by Bloodnok (Vicar) on Jul 27, 2008 at 23:14 UTC
Re: To be, or not to be, an object ? by Corion (Patriarch) on Jul 24, 2008 at 11:33 UTC
See Scalar::Util::blessed	[reply]
Re: To be, or not to be, an object ? by Bloodnok (Vicar) on Jul 24, 2008 at 11:54 UTC
Thanx both for your quick responses. I know about Scalar::Util but hadn't even considered it - doh!! Having had a peek @ the code, AFAICT, it does something extremely similar - I'm now rather worried I might be gaining a little understanding ... looks somewhat worried Mmm, Autobox - what a thing that is ?! I'll try my approach with it and see ... might even let you know the results :-)) Update There are, 2 aspects to the differences between my attempt and Grahams' (author of Scalar::Util), Graham... creates a temporary method in the `UNIVERSAL` class uses a call to `ref` on the suspect, before `eval`ing the call to his new, temporary method Whereas I use the (now standard) `isa()` method and omitted the call to `ref` - since the call to `isa()` would, methinx, generate either a 'not an object' or 'not an reference' error - aka a short cut. At last, a user level that overstates my experience :-))	[reply] [d/l] [select]
Re^2: To be, or not to be, an object ? by betterworld (Curate) on Jul 24, 2008 at 14:30 UTC
There are, 2 aspects to the differences between my attempt and Grahams' (author of Scalar::Util), Graham... creates a temporary method in the UNIVERSAL class Scalar::Util uses UNIVERSAL::a_sub_not_likely_to_be_here, but it's not a temporary method, because (unless I'm mistaken) it stays in the UNIVERSAL namespace forever. However, the XS version of Scalar::Util is pretty straightforward, it's `blessed()` does not use any dirty tricks at all, it just determines whether something is blessed, using the appropriate API.	[reply]
Re^3: To be, or not to be, an object ? by Bloodnok (Vicar) on Jul 24, 2008 at 14:33 UTC
TFT, point taken. I would point out that temporary and forever are one and the same in this context i.e. the lifetime of the program `use`ing Scalar::Utils. At last, a user level that overstates my experience :-))	[reply] [d/l]
Re: To be, or not to be, an object ? by kyle (Abbot) on Jul 24, 2008 at 15:42 UTC
Scalar::Util::blessed is the way to do this. If you're exceptionally paranoid, you have to check if it's defined, not merely true. `use Scalar::Util 'blessed'; my $oh_bee_jay = bless {}, '0'; print "not blessed\n" if not blessed $oh_bee_jay; print "defined blessed\n" if defined blessed $oh_bee_jay; __END__ not blessed defined blessed` [download] I'd be surprised if this is an issue, though. Package `'0'` is not too popular.	[reply] [d/l] [select]