Yeah the new systems have much better performance on it. It's not needed in the framework though and it's a slowdown in 1.8.6, so it's not there.
If you want it use it, but it's not used inside merb. Cheers On Mon, Feb 2, 2009 at 10:47 AM, Jacques Crocker <[email protected]> wrote: > > Hah.. good catch! Thanks for the tip :-) > > Here's the new results: > > With to_proc shorthand: > real 0m1.449s > user 0m1.335s > sys 0m0.033s > > Without to_proc shorthand: > real 0m0.376s > user 0m0.347s > sys 0m0.016s > > Also ran it on 1.9.1. Seems to run exactly the same speed both ways. > > With to_proc shorthand: > real 0m0.204s > user 0m0.172s > sys 0m0.014s > > Without to_proc shorthand: > real 0m0.201s > user 0m0.180s > sys 0m0.013s > > I do sorta prefer the shorthand to_proc syntax. I'll probably start > using it more often since the performance difference on it isn't too > bad anymore, and looks to become increasingly irrelevant as Ruby 1.9 > and 2.0 come out. > > > On Feb 1, 3:35 pm, jonuts <[email protected]> wrote: > > Why are you redefining Symbol#to_proc? It's a core method in 1.8.7 > > > > On Feb 2, 1:25 am, Jacques Crocker <[email protected]> wrote: > > > > > Hey Jonuts. You sure they fixed it in 1.8.7? I'm still seeing 10x > > > slowdown on the to_proc shorthand with my 1.8.7 installation > > > > > With to_proc shorthand: > > > real 0m4.138s > > > user 0m3.893s > > > sys 0m0.066s > > > > > Without to_proc shorthand: > > > real 0m0.391s > > > user 0m0.350s > > > sys 0m0.015s > > > > > [...@macbookpro ~/Ruby/benchmarks]# ruby --version > > > ruby 1.8.7 (2008-08-11 patchlevel 72) [i686-darwin9] > > > > > Here's the code I'm using to benchmark:http://gist.github.com/56695 > > > > > On Feb 1, 3:17 pm, jonuts <[email protected]> wrote: > > > > > > Just for the record, on ruby 1.8.7 I get (slow machine): > > > > > > user system total real > > > > sym#to_p 3.170000 0.510000 3.680000 ( 3.710362) > > > > normal 2.450000 0.610000 3.060000 ( 3.070118) > > > > > > There is really no reason not to use sym2proc if you're on 1.8.7+ > > > > > > On Feb 2, 1:03 am, Daniel N <[email protected]> wrote: > > > > > > > Here's why it's not there: > > > > > > > class Foo > > > > > def foo > > > > > :foo > > > > > end > > > > > end > > > > > > > @foos = [Foo.new] * 1_000_000 > > > > > > > @foos.each(&:foo) vs @foos.each{|f| f.foo} > > > > > > > user system total > real > > > > > sym 2 proc 2.000000 0.020000 2.020000 ( 2.093694) > > > > > raw ruby 0.330000 0.000000 0.330000 ( 0.345703) > > > > > > > Cheers > > > > > Daniel > > > > > > > On Mon, Feb 2, 2009 at 9:57 AM, jonuts < > [email protected]> wrote: > > > > > > > > D> run ruby 1.8.7 > > > > > > > > On Feb 2, 12:45 am, Phlip <[email protected]> wrote: > > > > > > > Merbalists: > > > > > > > > > If I feel like using posts.map(&:name), leveraging the cute > > > > > > Symbol#to_proc trick > > > > > > > from both Rails and 'facets', what's the _most_supportable_ way > to get > > > > > > it? > > > > > > > > > A> stick all three lines at the bottom of config/init.rb? > > > > > > > B> require 'facets'? > > > > > > > C> require someone's merb-cute-hacks module from GitHub? > > > > > > > > > BTW, please debate whether Symbol#to_proc is better than sex, > or the last > > > > > > bell > > > > > > > before the Apocalypse, in space provided: [___]. Please write > legibly... > > > > > > > > > -- > > > > > > > Phlip > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "merb" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/merb?hl=en -~----------~----~----~----~------~----~------~--~---
