Normally I would choose performance over @nearly_all_else, but this
time I'm deferring the new feature until the POE::XS project is done.
Invoke will make local-session calls about 2-3 times faster in the
current code base. The gains may not be as significant when large
parts of the back-end
On (02/19 21:49), Scott wrote:
But it could result in
some very miffed new users screaming Why the hell cant you do this?.
funny. i thought that's what i was saying here.
why can't i post an event to another session from that session's object?
Basically, we simply have to check to ensure
Rocco Caputo wrote:
On Fri, Feb 20, 2004 at 08:37:39AM -0500, sungo wrote:
but really, you've changed the subject. i wasn't talking about aliases.
i was talking about event dispatch. we were talking about one specific
piece of api that can be fixed to be much more clear. i don't really
have
[..]
but really, you've changed the subject. i wasn't talking about aliases.
i was talking about event dispatch. we were talking about one specific
piece of api that can be fixed to be much more clear. i don't really
have interest in saving the world right now because i don't have time to
code
is the patchset and benchmarks for a new
POE::Kernel-invoke method,
Why not name it callDirect or callShort or something like that?
Invoke doesn't imply how it differs from call.
Rocco might buy call_direct, but he wont buy wikiWords in the poe kernel
namespace :)
I'm still partial to gosub
Rocco Caputo wrote:
On Wed, Feb 18, 2004 at 07:37:02PM -0800, Scott wrote:
The difference is very favorable, here are my benchmark results:
POE::Kernel-call(): 123 wallclock secs (111.82 usr + 0.86 sys = 112.68
CPU) @ 8874.69/s (n=100)
POE::Kernel-invoke(): 33 wallclock secs (28.74
On Thu, Feb 19, 2004 at 10:02:55AM -0800, Scott wrote:
Anyway, I very much realize the naming isn't entirely up to me but
personally I would prefer a short, single word name that denotes the
activity of the method as opposed to something that doesn't seem proper
in the context of the
On Thu, 2004-02-19 at 17:52, Rocco Caputo wrote:
On Thu, Feb 19, 2004 at 10:02:55AM -0800, Scott wrote:
Anyway, I very much realize the naming isn't entirely up to me but
personally I would prefer a short, single word name that denotes the
activity of the method as opposed to something
[..]
I like $kernel-invoke(), because by virtue of being a different name
than $kernel-call() is means something different. And you can look
up the meaning in perldoc POE::Kernel.
Otherwise, I'd suggest something like $kernel-local_call(). local
referring to the session.
But, $kernel-do() is
On Thu, Feb 19, 2004 at 04:54:56PM -0800, Scott wrote:
I think a week is more than long enough, so we're looking at next
thursday?
I'd rather the wait be shorter, but I figure a week's good enough to let
everybody think about it and raise any alternative they think of. If
there are no
[..]
i think you'd get better semantic meaning here if the method was a
Session method instead of a Kernel method.
one of the things that drives me nuts semantically with poe is all these
methods in Kernel space that actually only operate on the Session.
There are three answers to this. One
On (02/19 18:58), Scott wrote:
There are three answers to this. One is maintaining API sanity:
well, i'm not sure its sanity we're maintaining in keeping all these
session-only methods in the kernel. its a small matter but its one that
mildly irritates me.
The second is a weird form of
[..]
i'm not getting how having these methods in Kernel.pm instead of
Session.pm forces sanity on the user using the modules. care to clarify?
In that particular instance I was basically referring to the context
switching and garbage collection checks of the session. If we're not
In response to an agreement that it would in fact be beneficial to
follow the pattern of POE::Kernel-yield rather than implement invoking
in the session object, attached is the patchset and benchmarks for a new
POE::Kernel-invoke method, which is heavily optimized for use by
embedded
Tim Wood wrote:
At 07:37 PM 02/18/04, Scott wrote:
In response to an agreement that it would in fact be beneficial to follow the pattern of POE::Kernel-yield rather than implement invoking in the session object, attached is the patchset and benchmarks for a new POE::Kernel-invoke method
15 matches
Mail list logo
