Kurt Stephens

dm-more depends on merb_datamapper (!?!) version 0.9.3 which does not appear to exist yet:

++ gem install dm-more
ERROR:  Error installing dm-more:
	dm-more requires merb_datamapper (= 0.9.3, runtime)

Not sure what to do about it.

http://devblog.famundo.com/articles/2006/12/07/improving-postgres-listen...

Unfortunately this hack blocks all threads. It should use rb_thread_select(), not select().

I did this little hackage during a boring PGcon 2008 talk…

This patch allows the postgres gem 0.7.9.2008.01.28 to be compiled under Ruby 1.9 against PostgreSQL 8.3.0 client libraries.

It appears to work against a PostgreSQL 8.1 server, but I have not run any detailed testing, yet.

--- ext/extconf.rb.orig	2008-05-22 09:47:06.000000000 -0500
+++ ext/extconf.rb	2008-05-22 09:52:28.000000000 -0500
@@ -1,6 +1,9 @@

+# Ruby 1.9
+PLATFORM = RUBY_PLATFORM unless defined? PLATFORM
+

 # windows compatibility, need different library name
 if(PLATFORM =~ /mingw|mswin/) then
<br class="clear" />

Type tags in Ruby VALUE

Internally, values in Ruby are 32-bit (at least for 32-bit processors). Some of the least-significant bits are used to store type information. See the VALUE definition in include/ruby/ruby.h. Using type tag bits avoids allocating additional memory for commonly-used immutable values, like integers.

Ruby uses a single-bit tag of 0x01 as the Fixnum type tag. The remaining bits, are used to store the Fixnum’s signed value. This is an immediate value; it doesn’t require storage for the Fixnum value to be allocated, unlike the heap space that would be required for a String. Other types will use different tag bits.

Helpful in IRB:

class Regexp
  def to_proc
    @proc ||= lambda { | x | self.match(x) }
  end
end

As in:

irb(main):001:0> pp Object.methods.sort.select(&/meth/)
["instance_method",
 "instance_methods",
 "method",
 "method_defined?",
 "methods",
 "private_class_method",
 "private_instance_methods",
 "private_method_defined?",
 "private_methods",
 "protected_instance_methods",
 "protected_method_defined?",
 "protected_methods",
 "public_class_method",
 "public_instance_methods",
 "public_method_defined?",
 "public_methods",
 "singleton_methods"]
<br class="clear" />

Gem::SourceIndex does not honor GEM_PATH ordering.

See: http://rubyforge.org/tracker/index.php?func=detail&aid=14816&group_id=12...

Gem::SourceIndex#load_gems_in calls #add_spec for all gems found in #spec_dirs, in the order of Gem.path, however #add_spec
overwrites previous gem_specs in the @gems Hash:

So you’re coding some ruby, and you do the obligatory caching of a computation:

def foo(x)
  666
end

$cache = nil
def cryptic_cached_foo
  ($cache ||= [ foo("bar") ]).first
end

def nicey_cached_foo
  unless $cache
    $cache = [ foo("bar") ]
  end
  $cache.first
end

A discussion came up at work: is nicey_cached_foo “better” than cryptic_cached_foo? Obviously, nicey_cached_foo is more readable, but when I find myself doing the same pattern over and over, I prefer the terse cryptic_cached_foo. I expected cryptic_cached_foo to be faster. Oh boy, was I wrong….

Links about Caching in Rails.

• Object Caching in Rails
• CachedModel
• acts_as_cached
• ActiveRecord Explicit Caching

• “Expanded Caching Scope” – http://ryandaigle.com/articles/2007/2/21/what-s-new-in-edge-rails-expanded-caching-scope
  

http://eigenclass.org/hiki/Changes+in+Ruby+1.9

Excellent synopsis of changes to Ruby 1.9.

Other goodies:

http://eigenclass.org/hiki/non-synthetic-benchmarks-for-yarv

http://eigenclass.org/hiki/porting-rails-to-ruby-1.9

Here is a patch to rcov 0.8.0.2 to rethrow any exception generated by the scripts to be traced, after report generation (most importantly SystemExit). This allows the exit code from test runners (like spec) to be used under rake.