Mailman 3 September 2009 - Twisted

[Twisted-Python] ConFoo.ca call for speakers
by Yannick Gingras Sept. 23, 2009

Sept. 23, 2009

Greetings Twisters, I'm coordinator for the Python track at the ConFoo.ca conference and I have an announcement to make that will certainly be of interest to many of you. PHP-Québec, Montréal-Python, Ruby Montréal, W3Qc, and OWASP Montréal are organizing the first edition of the ConFoo.ca conference, which will be held in Montréal on March 10th through 12th at the Hilton Bonaventure Hotel. With over 500 expected attendees, ConFoo.ca is the largest Web development conference in North … [View More]

1 0

[Twisted-Python] VC++ client and a Twisted OpenSSL Server
by gary clark Sept. 23, 2009

Sept. 23, 2009

Hello, Firstly I'm new to twisted hence patience and a deep breath my be necessary. I wrote a simple VC++ client that communicates to a VC++ server using OpenSSL. It loads .pem file and communicates great. However I want the client to talk to a twisted server provided from TwistedMatrix: # Copyright (c) 2001-2004 Twisted Matrix Laboratories. # See LICENSE for details. from OpenSSL import SSL from twisted.internet.protocol import Factory from twisted.internet import ssl, reactor from … [View More]

2 1

Re: [Twisted-Python] Where is TwistedQuotes source code?
by David Ripton Sept. 23, 2009

Sept. 23, 2009

On 2009.09.22 22:14:15 -0400, John Crawford wrote: > I'm trying to find the source code for the TwistedQuotes application. > I've looked in every section of the website, and failed to find it, > alas. Is it still there, and if so, where? You can browse it using Trac here: http://twistedmatrix.com/trac/browser/trunk/doc/core/howto/listings/Twisted… Or just checkout the Twisted trunk from svn. -- David Ripton dripton(a)ripton.net

1 0

[Twisted-Python] Where is TwistedQuotes source code?
by John Crawford Sept. 23, 2009

Sept. 23, 2009

I'm trying to find the source code for the TwistedQuotes application. I've looked in every section of the website, and failed to find it, alas. Is it still there, and if so, where? I'm primarily interested in it, to use in understanding the documentation section, "Designing Twisted Applications", which uses TwistedQuotes as its example code. Thanks for any help. John C>

1 0

[Twisted-Python] Seeking advice on use of Perspective Broker
by Arthur Pemberton Sept. 22, 2009

Sept. 22, 2009

Good day, I have scanned through the docs a few times, and I am in need of some advice and explanation to determine if Twisted's Perspective Broker meetings my perceived needs What I am trying to do : create a client/server project where the server is a a typical access multiple data sources, enforce business logic, and pass data to client. I am looking to use RPC type technology to facilitate transparent use of the network to communicate between the client and server. What I am looked at so … [View More]

5 9

[Twisted-Python] Reviewer who understands basic numeric algorithms wanted
by Laurens Van Houtven Sept. 21, 2009

Sept. 21, 2009

Hello :-) twisted.positioning has just (read: last night) grown a new module, called util, which has two functions I needed for statistical analysis: the error function (erf) and the inverse normal cumulative distribution function (Phi). As most of you on this mailing list probably know, code headed for trunk needs to be reviewed. The reason I need to ask for this specifically is because the reviewer does not need to know anything about Twisted, but does need some experience with how … [View More]numerical algorithms work -- and that's about the opposite of most Twisted reviewers ;-) The default is to attempt to use scipy's implementation (which is hopefully faster), but otherwise fall back to a pure Python implementation (which from my tests is still way more than fast enough). The implementations rely on Chebyshev polynomial interpolation. The error function is a pretty simple one, the inverse normal cdf is slightly harder (three parts: the center is a simple rational, the 'tails' are rationals expressed in a linearizing parameter). There are two things that make this easier: 1. Other people have done similar things, and said coefficients can be looked up in other places (I'm not the first person to do the math -- I'm just the first person to implement both in Python with a shared Horner scheme evaluator, apparently) 2. It can be shown to be accurate by comparing the results at random points with scipy's reference implementation if we do it for 10**5 or 10**6 points or so -- there's already an automated unit test that does it for a much smaller number If you're not 100% sure you're up to it from just seeing the code, I can also talk you through it and show how I got there :-) It might be more useful to create a separate ticket for this piece of code for reviewing purposes, if anyone else feels that's the case I'll make the ticket. People who are interested: you can reply to this mail, mail me personally, or contact me on freenode (nick: lvh). (Oh, and before someone tells me I'm reinventing the wheel: yes, but not really -- the only commonly available python implementation is Scipy, that's *already* used as the default, and we do need something in case of no scipy (adding a dependency just because of one function on all of scipy is not exactly good either) and the only other simpler alternative would be a giant LUT with linear interpolation which would have been *much* worse near the edges of the interval of invncdf, which also happen to be the *interesting* bits of that function...) tia lvh [View Less]

3 5

[Twisted-Python] Weekly Bug Summary
by exarkun＠twistedmatrix.com Sept. 20, 2009

Sept. 20, 2009

1 0

[Twisted-Python] What is the difference in between reactor.running and reactor._stopped
by Gabriel Rossetti Sept. 20, 2009

Sept. 20, 2009

Hello, I tried using reactor.running for a test, to not try to stop the reactor if it's already stopped, but it doesn't work. If I use reactor._stopped instead, it works. I know I'm not supposed to access it sonce it is prefixed by the underscore, so my question is what is the difference in between reactor.running and reactor._stopped, and why does reactor.running not work as I expect (not reactor.running == reactor._stopped)? Thank you, Gabriel

3 7

Re: [Twisted-Python] Using reactor with multiple processes
by David Yoakley Sept. 18, 2009

Sept. 18, 2009

I promised a follow-up on what we found so here it is. As Phil indicated prior and is now clear to us, twisted is not designed to be fork safe. But as I considered patching it to set up all state when the reactor is started then I began to worry about whether I was really going to be happy with using twisted in a python multiprocess architecture since its ultimately not twisted aware and thus there are probably going to be blocking issues with using any of cool multiprocess abstractions. I … [View More]think we were just going in the wrong direction, attracted by a shiny new thing. To make a long story short, we think making twisted work with python multiprocess package is probably a dead end. The abstractions are nice and maybe the right direction but under the hood its just not going to work. Instead, we implemented a simple framework using twisted PerspectiveBroker (pb) where we spawn processes with python subprocess.Popen and use PB remote calls to manage the processes. Its now working well and is very simple. There are other aspects we didn't think about that are actually working out better for us with this PB based framework. Mostly, our child processes are much less coupled to the parent; they are able to be launched on their own from a command line for unit tests and such but each child is an independent server (implementing pb.PBServerFactory). They can be launched from our main process (implementing pb.PBClientFactory) in production and any number of other ways during development and test. Similarly, we could allocate the children even to separate machines if we wanted to since their connections are via tcp. The last thing is more subtle. We are now not tempted to use tighter synchronization (more coupling) such as shared memory or other mechanisms offered by python multiprocess in cases where its not strictly necessary. The processes all share configuration (such as child tcp port assignments) by having the parent send on the command line, an absolute path to a common config file when the process is launched and after that the children carry on pretty much independently, occasionally collaborating with each other and with the parent through remote calls as needed. When its time for the application to go away, the parent sends each child a stop command which blocks the parents shutdown completion (in the sense that its synchronous and the parent waits on the response, but NOT in the sense that any other logical twisted threads on the parent get blocked) until the child responds or times out. Cheers, David Yoakley [View Less]

2 1

[Twisted-Python] new contributors
by Glyph Lefkowitz Sept. 18, 2009

Sept. 18, 2009

We *still* don't have a totally official process for adding new contributors to SVN, but I've been noticing a steady stream of contributions from two new folks: lvh and jesstess. Would you folks like commit access? If so, does anyone have an objection on adding them to the Twisted Matrix Labs? (Feel free to contact me privately if you'd rather not be mean in public ;-)).

5 6