Scope::Guard - lexically-scoped resource management
my $guard = guard { ... };
# or
my $guard = scope_guard \&handler;
# or
my $guard = Scope::Guard->new(sub { ... });
$guard->dismiss(); # disable the handler
This module provides a convenient way to perform cleanup or other forms of resource management at the end of a scope. It is particularly useful when dealing with exceptions: the Scope::Guard
constructor takes a reference to a subroutine that is guaranteed to be called even if the thread of execution is aborted prematurely. This effectively allows lexically-scoped "promises" to be made that are automatically honoured by perl's garbage collector.
For more information, see: http://www.drdobbs.com/cpp/184403758
my $guard = Scope::Guard->new(sub { ... });
# or
my $guard = Scope::Guard->new(\&handler);
The new
method creates a new Scope::Guard
object which calls the supplied handler when its DESTROY
method is called, typically at the end of the scope.
$guard->dismiss();
# or
$guard->dismiss(1);
dismiss
detaches the handler from the Scope::Guard
object. This revokes the "promise" to call the handler when the object is destroyed.
The handler can be re-enabled by calling:
$guard->dismiss(0);
guard
takes a block and returns a new Scope::Guard
object. It can be used as a shorthand for:
Scope::Guard->new(...)
e.g.
my $guard = guard { ... };
Note: calling guard
anonymously, i.e. in void context, will raise an exception. This is because anonymous guards are destroyed immediately (rather than at the end of the scope), which is unlikely to be the desired behaviour.
scope_guard
is the same as guard
, but it takes a code ref rather than a block. e.g.
my $guard = scope_guard \&handler;
or:
my $guard = scope_guard sub { ... };
or:
my $guard = scope_guard $handler;
As with guard
, calling scope_guard
in void context will raise an exception.
0.21
chocolateboy <chocolate@cpan.org>
Copyright (c) 2005-2015, chocolateboy.
This module is free software. It may be used, redistributed and/or modified under the same terms as Perl itself.