Tomasz Narloch pisze:
Hello,
I have a problem with the transfer of variables config and app_globals
to my own thread.
I need config and app_globals has been passed by reference.
Now, I like the one below, but it is not good because it throws a
segmentation fault at the end of threads.
I read something about register config and agg_globals to threads in
pylons but I do not know how to do that.
#Maybe I should remove reference to config and app_globals after
thread end?
I want the same config and app_globals in my thread and
when I change some params of app_globals (only one of three threads do
that) it has to be changed in other pylons threads.
Maybe exists some other way to do that?
my thread class:
----------------------------------------------------------------------------------------------
# -*- coding: utf-8 -*-
import threading
class Mythread(threading.Thread):
# class variables, not instances only
# Mythread.results[self.myId] = function()
results = []
vlock = threading.Lock()
id = 0 # I want to give an ID number to each thread, starting at 0
def __init__(self, function, **args):
# invoke constructor of parent class
threading.Thread.__init__(self)
# add instance variables
self.myId = Mythread.id
Mythread.id += 1
Mythread.results.append(None)
self.function = function
self.args = args
self.args['vlock'] = Mythread.vlock
def run(self):
# update results in an atomic manner
# nie ma wyścigu więc nie trzeba blokować
#Mythread.vlock.acquire()
Uncommented
Mythread.results[self.myId] = self.function(**self.args)
#Mythread.vlock.release()
Uncommented
@staticmethod
def flush():
Mythread.results = []
Mythread.id = 0
#end
----------------------------------------------------------------------------------------------
I run thread from:
----------------------------------------------------------------------------------------------
thread.Mythread.flush()
e0 = thread.Mythread(fun_1, names=lnames1, name=param)
e1 = thread.Mythread(fun_2, names=l.names2, name=param)
e2 = thread.Mythread(fun_3, names=lnames3, name=param)
for s in [e0, e1, e2]:
# I add additional param environ to function fun_1, fun_2, fun_3
s.args['environ'] = pylons.request.environ
s.start()
# wait for all threads to completed
for s in [e0, e1, e2]:
s.join()
nresult = thread.Mythread.results[-3]
hresult = thread.Mythread.results[-2]
eresult = thread.Mythread.results[-1]
----------------------------------------------------------------------------------------------
function as thread:
def fun_1(names, name, environ, vlock=None):
some_file.config = environ['pylons.pylons'].config
replaced by
pylons.config._push_object(environ['pylons.pylons'].config)
pylons.app_globals._push_object(environ['pylons.pylons'].config['pylons.app_globals'])
try:
root = some_file.expensiveFunctionThatNeedConfig(names)
....
finally:
pylons.app_globals._pop_object(environ['pylons.pylons'].config['pylons.app_globals'])
pylons.config._pop_object(environ['pylons.pylons'].config)
----------------------------------------------------------------------------------------------
some_file content:
from pylons import config as conf, app_globals
replace by default
from pylons import config as config, app_globals
...
#saveThreadConfig, for now I can replace empty conf to that from environ
# in theads area conf is almost empty so I rewrite it, but this
probably generate segmentation fault after threads end
config = conf
g = app_globals
...
removed above
def expensiveFunctionThatNeedConfig(names):
# some work with config and g (app_globals)
...
----------------------------------------------------------------------------------------------
Best Regards,
Tomasz Narloch
After this changes works.
Best Regards,
Tomasz Narloch
--
You received this message because you are subscribed to the Google Groups
"pylons-discuss" 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/pylons-discuss?hl=en.
