I noticed this evening we don't cache literal floats in either the
interpreter or the compiler, but I could not remember why. I've attached
a patch that fixes this, but does anyone remember if there's a reason
why we couldn't cache floats? Are they mutable or something?
- Charlie
>From c4a8d4b
Charles Oliver Nutter wrote:
I noticed this evening we don't cache literal floats in either the
interpreter or the compiler, but I could not remember why. I've
attached a patch that fixes this, but does anyone remember if there's
a reason why we couldn't cache floats? Are they mutable or someth
On Sat, Feb 21, 2009 at 2:42 AM, Charles Oliver Nutter
wrote:
> I noticed this evening we don't cache literal floats in either the
> interpreter or the compiler, but I could not remember why. I've attached a
> patch that fixes this, but does anyone remember if there's a reason why we
> couldn't ca
One can always:
class Float
def singleton_method_added *args;end
end
a = 1.0
class << a
def + foo;1;end
end
Marcin
-
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
Marcin Mielżyński wrote:
One can always:
class Float
def singleton_method_added *args;end
end
a = 1.0
class << a
def + foo;1;end
end
Ok, that's one possible reason...but it's a stretch. It seems like
caching them won't actually hurt anything we'd expect people to use.
- Charlie
Ola Bini wrote:
Charles Oliver Nutter wrote:
I noticed this evening we don't cache literal floats in either the
interpreter or the compiler, but I could not remember why. I've
attached a patch that fixes this, but does anyone remember if there's
a reason why we couldn't cache floats? Are they
[PATCH] FFI should allow passing nil as a callback parameter
Key: JRUBY-3436
URL: http://jira.codehaus.org/browse/JRUBY-3436
Project: JRuby
Issue Type: Bug
Components: Ex
