package UNIVERSAL::require; $UNIVERSAL::require::VERSION = '0.11'; # We do this because UNIVERSAL.pm uses CORE::require(). We're going # to put our own require() into UNIVERSAL and that makes an ambiguity. # So we load it up beforehand to avoid that. BEGIN { require UNIVERSAL } package UNIVERSAL; use strict; use vars qw($Level); $Level = 0; =pod =head1 NAME UNIVERSAL::require - require() modules from a variable =head1 SYNOPSIS # This only needs to be said once in your program. require UNIVERSAL::require; # Same as "require Some::Module" my $module = 'Some::Module'; $module->require or die $@; # Same as "use Some::Module" BEGIN { $module->use or die $@ } =head1 DESCRIPTION If you've ever had to do this... eval "require $module"; to get around the bareword caveats on require(), this module is for you. It creates a universal require() class method that will work with every Perl module and its secure. So instead of doing some arcane eval() work, you can do this: $module->require; It doesn't save you much typing, but it'll make alot more sense to someone who's not a ninth level Perl acolyte. =head1 Methods =head3 require my $return_val = $module->require or die $@; my $return_val = $module->require($version) or die $@; This works exactly like Perl's require, except without the bareword restriction, and it doesn't die. Since require() is placed in the UNIVERSAL namespace, it will work on B module. You just have to use UNIVERSAL::require somewhere in your code. Should the module require fail, or not be a high enough $version, it will simply return false and B. The error will be in $@ as well as $UNIVERSAL::require::ERROR. $module->require or die $@; =cut sub require { my($module, $want_version) = @_; $UNIVERSAL::require::ERROR = ''; die("UNIVERSAL::require() can only be run as a class method") if ref $module; die("UNIVERSAL::require() takes no or one arguments") if @_ > 2; my($call_package, $call_file, $call_line) = caller($Level); # Load the module. my $file = $module . '.pm'; $file =~ s{::}{/}g; # For performance reasons, check if its already been loaded. This makes # things about 4 times faster. return 1 if $INC{$file}; my $return = eval qq{ #line $call_line "$call_file" CORE::require(\$file); }; # Check for module load failure. if( $@ ) { $UNIVERSAL::require::ERROR = $@; return $return; } # Module version check. if( @_ == 2 ) { eval qq{ #line $call_line "$call_file" \$module->VERSION($want_version); }; if( $@ ) { $UNIVERSAL::require::ERROR = $@; return 0; } } return $return; } =head3 use my $require_return = $module->use or die $@; my $require_return = $module->use(@imports) or die $@; Like C, this allows you to C a $module without having to eval to work around the bareword requirement. It returns the same as require. Should either the require or the import fail it will return false. The error will be in $@. If possible, call this inside a BEGIN block to emulate a normal C as closely as possible. BEGIN { $module->use } =cut sub use { my($module, @imports) = @_; local $Level = 1; my $return = $module->require or return 0; my($call_package, $call_file, $call_line) = caller; eval qq{ package $call_package; #line $call_line "$call_file" \$module->import(\@imports); }; if( $@ ) { $UNIVERSAL::require::ERROR = $@; return 0; } return $return; } =head1 SECURITY NOTES UNIVERSAL::require makes use of C. In previous versions of UNIVERSAL::require it was discovered that one could craft a class name which would result in code being executed. This hole has been closed. The only variables now exposed to C are the caller's package, filename and line which are not tainted. UNIVERSAL::require is taint clean. =head1 COPYRIGHT Copyright 2001, 2005 by Michael G Schwern Eschwern@pobox.comE. This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself. See F =head1 AUTHOR Michael G Schwern =head1 SEE ALSO L, L, L =cut 1;