#!/usr/local/bin/perl -w
use strict;
no strict 'refs';

use vars qw( $opt_m $opt_M );

package A;
use Data::Dumper;
use vars qw($AUTOLOAD);
my %cache;
use Carp 'cluck';

use optimizer extend => sub {
    my $name = $_[0]->name();
    cluck "dump ", Dumper \@_
    if 0
        or $name eq  "goto"
        # or $name eq  "print"
        or ref $_[0] =~ /CV/i
            ;
    # =~ /__ANON__/;
};

sub AUTOLOAD { # Use an extension of the default optimizer
    (my $meth = $AUTOLOAD) =~ s/.*:://;
    return if $meth eq 'DESTROY';
    my ($package, $filename, $line, $subroutine) = caller(0);

    my $cat = join '_', $meth, $package, $filename, $line;
    $cat =~ s|[\./:]||g;

    my $code;
    unless (exists $cache{$cat}) {
    $code = sub { shift @_; print " calling <$meth> ",$cat," (@_)\n" }
+;
    print "creating <$meth> $cat\n";
    $cache{$cat} = $code;
    *{'A::'.$cat} = $code if $main::opt_m; # vivify specific-usage nam
+e
    *{'A::'.$meth} = $code unless $main::opt_M; # vivify generic name
    } else {
    # runs when munging
    $code = $cache{$cat};
    print "fetching: $cat\n";
    }
    goto &$code;
}

package main;
use Getopt::Std;
getopts('mM') or die <<EOH;
    -m : enables munged sub-name vivify
    -M : suppresses non-munged sub-name vivify, causes fetch:
-M forces AUTOLOAD to run on 2nd pass, by not vivifying base name
EOH

#A::auto("string");
loop();
print "done\n";

sub loop {
    my $munge;
    foreach (1..2) {
    print "$_:\n";
    A->debug('aval');
    A->info('another');
    A->info('now that info() is built, the line numbers wrong.');
    A->info('more importantly, this call needs its own handler, not th
+e other one.');
    A->something('anything');
    A->other('anything');
    }
}

__END__

=head1 extending Log::Log4perl

package Log::Log4perl::AutoCategorize    # notional extension

Log::Log4perl provides a flexible logging system which allows
configuration of one or more output streams, each collecting messages
from a set of logging-categories / sources.

Id like to extend it by coupling the logging-category to caller()
info, following benefits accrue;

 1. END{} handler can report on Logger coverage.
 2. scope friendly filtering,
 3. well, 2 is something..

=head1 my code

Im using AUTOLOAD to implement A->$invoke(), for all values of
$invoke; Ill eventually delegate them to Log::Log4perl->debug(),
info(), warn(), or error(), dependent upon config.

AUTOLOAD uses caller() to make and test the logging-category, and woul
+d either write a log entry, or just return.  That yes/no decision can
+ result in a noop, or an appropriate print to stream.

method-munging guarantees that each $invoke() is lexically unique (fil
+e & line, approx), and carries context for purposes of logging and fi
+ltering.  But once name is munged, its not accessible by original cal
+ling point, and AUTOLOAD must regenerate the munged-name to look up t
+he cached coderef.

=head1 SYNOPSIS - Desired

# assuming a package like this
use Log::Log4perl::AutoCategorize ( alias_as => 'Logger' );

sub doit {
    ...
    # want simple invocations, like this
    Logger->debug(@_);

    # to really mean
    Logger->debug_<$callerpkg,$caller_sub,$caller_ln> (@_);
    # its really a macro
}
[download]

=head1 what I need

caller(), Log4perl::get_logger(), Log4perl::is_writable(), are quite expensive for repeated use in a system logger.

if I can use optimizer within AUTOLOAD to patch each invocation, I can build sub{}s or link to previously built ones, and eliminate subequent calls to AUTOLOAD. I also get to grab everything I need to provide a customized response for the caller.

aside: the contrib of the munged-name methods to the size of the symbol table is linear with the config size, not the count of invocations.

I hope this is an interesting enough application of B::Generate, optimizer modules to attract your attention / curiosity / spare clues. Thanks in advance.

---

---

Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!

Titles consisting of a single word are discouraged, and in most cases are disallowed outright.

Read Where should I post X? if you're not absolutely sure you're posting in the right place.

Please read these before you post! —

• How do I compose an effective node title?
• How do I post a question effectively?
• Markup in the Monastery

Posts may use any of the Perl Monks Approved HTML tags:

a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr

You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)

	For:		Use:
	&		&
	<		&lt;
	>		&gt;
	[		&#91;
	]		&#93;

Link using PerlMonks shortcuts! What shortcuts can I use for linking?

See Writeup Formatting Tips and other pages linked from there for more info.