You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@qpid.apache.org by "Xin Chen (JIRA)" <ji...@apache.org> on 2017/12/01 22:58:00 UTC
[jira] [Updated] (PROTON-1709) Python binding: ApplicationEvent
causing memory growth
[ https://issues.apache.org/jira/browse/PROTON-1709?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Xin Chen updated PROTON-1709:
-----------------------------
Description:
ApplicationEvent creates a new EventType object in the constructor. The EventType object adds itself to the TYPES dict always, even for the same typename. There is no way for the application to avoid this by providing an EventType object directly. Repro code below:
{code:java}
import threading
from proton.reactor import Container, ApplicationEvent, EventInjector
class Program:
def __init__(self, injector):
self.injector = injector
def on_reactor_init(self, event):
event.reactor.selectable(self.injector)
def on_hello(self, event):
print(event.subject)
def on_done(self, event):
event.subject.stop()
e = EventInjector()
r = Container(Program(e))
t = threading.Thread(target=r.run)
t.start()
for i in range(1, 10000000):
e.trigger(ApplicationEvent("hello", subject=str(i)))
e.trigger(ApplicationEvent("done", subject=r))
t.join()
{code}
was:
ApplicationEvent creates a new EventType object in the constructor. The EventType object adds itself to the TYPES dict always, even for the same typename. There is no way for the application to avoid this by providing an EventType object directly. Repro code below:
import threading
from proton.reactor import Container, ApplicationEvent, EventInjector
class Program:
def __init__(self, injector):
self.injector = injector
def on_reactor_init(self, event):
event.reactor.selectable(self.injector)
def on_hello(self, event):
print(event.subject)
def on_done(self, event):
event.subject.stop()
e = EventInjector()
r = Container(Program(e))
t = threading.Thread(target=r.run)
t.start()
for i in range(1, 10000000):
e.trigger(ApplicationEvent("hello", subject=str(i)))
e.trigger(ApplicationEvent("done", subject=r))
t.join()
> Python binding: ApplicationEvent causing memory growth
> ------------------------------------------------------
>
> Key: PROTON-1709
> URL: https://issues.apache.org/jira/browse/PROTON-1709
> Project: Qpid Proton
> Issue Type: Bug
> Components: python-binding
> Affects Versions: proton-c-0.18.1
> Reporter: Xin Chen
>
> ApplicationEvent creates a new EventType object in the constructor. The EventType object adds itself to the TYPES dict always, even for the same typename. There is no way for the application to avoid this by providing an EventType object directly. Repro code below:
> {code:java}
> import threading
> from proton.reactor import Container, ApplicationEvent, EventInjector
> class Program:
> def __init__(self, injector):
> self.injector = injector
> def on_reactor_init(self, event):
> event.reactor.selectable(self.injector)
> def on_hello(self, event):
> print(event.subject)
> def on_done(self, event):
> event.subject.stop()
> e = EventInjector()
> r = Container(Program(e))
> t = threading.Thread(target=r.run)
> t.start()
> for i in range(1, 10000000):
> e.trigger(ApplicationEvent("hello", subject=str(i)))
> e.trigger(ApplicationEvent("done", subject=r))
> t.join()
> {code}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@qpid.apache.org
For additional commands, e-mail: dev-help@qpid.apache.org