From: Brad Bowman <bsb@strategicdata.com.au>
Date: Wed, 28 Sep 2011 01:49:31 +0000 (+1000)
Subject: inline_slot_access not inline_create_instance eg.
X-Git-Tag: 2.0301~16
X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=commitdiff_plain;h=2dbd5ebf42aff7da3355a9cde76f35bc87c4ef23;hp=251c65399c9432e8431e2752f063d1870c502bd8;p=gitmo%2FMoose.git

inline_slot_access not inline_create_instance eg.

The text refers to inline_slot_access, but the code
shown is for inline_create_instance.  This shows the
inline_slot_access code instead, as presumably intended.
---

diff --git a/lib/Moose/Cookbook/Meta/Recipe7.pod b/lib/Moose/Cookbook/Meta/Recipe7.pod
index b3746e6..b661d44 100644
--- a/lib/Moose/Cookbook/Meta/Recipe7.pod
+++ b/lib/Moose/Cookbook/Meta/Recipe7.pod
@@ -150,9 +150,9 @@ This level of indirection probably makes our instance class I<slower>
 than the default. However, when attribute access is inlined, this
 lookup will be cached:
 
-  sub inline_create_instance {
-      my ( $self, $class_variable ) = @_;
-      return 'do { my $sym = Symbol::gensym(); bless $sym, ' . $class_variable . ' }';
+  sub inline_slot_access {
+      my ( $self, $instance, $slot_name ) = @_;
+      return '*{' . $instance . '}->{' . $slot_name . '}';
   }
 
 The code snippet that the C<inline_slot_access> method returns will