Mailman 3 November 2005 - Twisted

[Twisted-Python] twistd -noy unable to cancel using CTRL+C when using threads
by Clark C. Evans 16 Nov '05

16 Nov '05

Hello. The attached example works properly when called from twistd -nox test.py when one is not using threads. However, as soon as thread are introduced, you can no longer use CTRL+C to kill the process. I'm running FreeBSD 5.3-RELEASE and Python 2.4.1 Best, Clark

1 0

[Twisted-Python] Re: [Twisted-commits] r15100 - So I really do want this stuff
by Jean-Paul Calderone 16 Nov '05

16 Nov '05

On Fri, 11 Nov 2005 00:38:58 -0700, Jonathan Lange <jml(a)wolfwood.twistedmatrix.com> wrote: >Author: jml >Date: Fri Nov 11 00:38:56 2005 >New Revision: 15100 > >Added: > branches/jml/trial-setupclass-death/twisted/trial/dijkstra.py > - copied unchanged from r15099, sandbox/jml/testresources/dijkstra.py > branches/jml/trial-setupclass-death/twisted/trial/priodict.py > - copied unchanged from r15099, sandbox/jml/testresources/priodict.py > branches/jml/trial-setupclass-death/twisted/trial/resources.py > - copied unchanged from r15099, sandbox/jml/testresources/resources.py > branches/jml/trial-setupclass-death/twisted/trial/test/test_optimizing_test_suite.py > - copied unchanged from r15099, sandbox/jml/testresources/test_optimizing_test_suite.py > branches/jml/trial-setupclass-death/twisted/trial/test/test_resourced_test_case.py > - copied unchanged from r15099, sandbox/jml/testresources/test_resourced_test_case.py > branches/jml/trial-setupclass-death/twisted/trial/test/test_test_loader.py > - copied unchanged from r15099, sandbox/jml/testresources/test_test_loader.py > branches/jml/trial-setupclass-death/twisted/trial/test/test_test_resource.py > - copied unchanged from r15099, sandbox/jml/testresources/test_test_resource.py >Log: >So I really do want this stuff > This seems to beg the question: Why? So, consider it begged :) Jean-Paul

6 18

[Twisted-Python] UnicodeDecodeError with twisted.conch.ssh client
by Henning.Ramm＠mediapro-gmbh.de 16 Nov '05

16 Nov '05

Hi there! I get the following traceback while trying to connect to a server via conch.ssh: --- * --- starting service ssh-userauth Traceback (most recent call last): --- <exception caught here> --- File "D:\Python24\lib\site-packages\twisted\internet\selectreactor.py", line 139, in _doReadOrWrite why = getattr(selectable, method)() File "D:\Python24\lib\site-packages\twisted\internet\tcp.py", line 351, in doRead return self.protocol.dataReceived(data) File "D:\Python24\lib\site-packages\twisted\conch\ssh\transport.py", line 182, in dataReceived f(packet[1:]) File "D:\Python24\lib\site-packages\twisted\conch\ssh\transport.py", line 591, in ssh_SERVICE_ACCEPT self.setService(self.instance) File "D:\Python24\lib\site-packages\twisted\conch\ssh\transport.py", line 219, in setService self.service.serviceStarted() File "D:\Python24\lib\site-packages\twisted\conch\ssh\userauth.py", line 220, in serviceStarted self.askForAuth('none', '') File "D:\Python24\lib\site-packages\twisted\conch\ssh\userauth.py", line 224, in askForAuth self.transport.sendPacket(MSG_USERAUTH_REQUEST, NS(self.user) + \ File "D:\Python24\lib\site-packages\twisted\conch\ssh\transport.py", line 107, in sendPacket packet = struct.pack('!LB', totalSize+lenPad-4, lenPad)+ \ File "D:\Python24\lib\encodings\utf_8.py", line 16, in decode return codecs.utf_8_decode(input, errors, True) exceptions.UnicodeDecodeError: 'utf8' codec can't decode bytes in position 2-3: invalid data connection lost Stopping factory <twisted.internet.protocol._InstanceFactory instance at 0x02D2B210> --- * --- I use password auth, and both user and password use only 7-bit ASCII chars. It always happens on one machine, but works on another. Both run a german Windows XP with Twisted 2.0, but this one has Python 2.4 and the other 2.3 Is it perhaps a known bug (and fixed in Twisted 2.1)? Best regards, Henning Hraban Ramm Südkurier Medienhaus / MediaPro Support/Admin/Development Dept.

2 1

RE: [Twisted-Python] Traceback bug?
by Henning.Ramm＠mediapro-gmbh.de 16 Nov '05

16 Nov '05

>>This looks like a problem of Twisted's Failure: >>Should I file a bug, or is it my fault? > >You didn't say what version of Twisted you're using or how you >caused this to happen, but it looks to me like a bug that has >been fixed already. Try with 2.1 if you're using 2.0, or >trunk if you're using 2.1. If it doesn't go away, then file a bug. Sorry, I reckognized that I forgot all data after I posted the mail. Yes, I'm still on 2.0 on Win, because I wanted to stick to the binary installers. But, however, I can avoid the bug, and my whole project including my job is on being dropped... Best regards, Henning Hraban Ramm Südkurier Medienhaus / MediaPro Support/Admin/Development Dept.

1 0

[Twisted-Python] Weekly Bug Summary
by James Y Knight 16 Nov '05

16 Nov '05

KIND == 'create' Bug summary ______________________ Since 2005-11-09 07:00:01.570000: Bugs opened: 14 Bugs closed: 11 Total open bugs: 483 (+3) Urgent: 1 (+0) Bug: 226 (-5) Feature: 173 (+4) Wish: 83 (+4) New / Reopened Bugs ______________________ ==== Urgent ==== [CORE] "Sumo" is confusing and unnecessary (3 days ago) http://twistedmatrix.com/bugs/issue1331 opened by exarkun ==== Bug ==== [BACKPORT, CORE] FTP server does not work with ange-ftp (2 days ago) CLOSED http://twistedmatrix.com/bugs/issue1332 opened by exarkun IOCP AcceptExOp.ovDone not called when expected (4 days ago) http://twistedmatrix.com/bugs/issue1326 opened by justinj document reactor reentrancy policy (5 days ago) http://twistedmatrix.com/bugs/issue1323 opened by justinj [CORE, TESTS] twisted.test.test_udp.testRebind makes no sense (2 weeks ago) CLOSED http://twistedmatrix.com/bugs/issue1295 opened by exarkun ==== Feature ==== [TRIAL] Custom Trial reactor to track resources and handle cleanup (yesterday) http://twistedmatrix.com/bugs/issue1334 opened by jml [PATCH, sorta] Socks V5 functionality (3 days ago) http://twistedmatrix.com/bugs/issue1330 opened by jadestorm [CONCH] [patch] Performance of Conch/SSH (5 days ago) http://twistedmatrix.com/bugs/issue1325 opened by cooz [CORE, PB] Cacheable/RemoteCache does not work with new style classes (5 days ago) http://twistedmatrix.com/bugs/issue1324 opened by karlb ==== Wish ==== [TRIAL] unittest.main() (4 days ago) http://twistedmatrix.com/bugs/issue1329 opened by jml [CORE, TRIAL] twistd and trial --help lists all reactors, regardless of OS (4 days ago) http://twistedmatrix.com/bugs/issue1327 opened by PenguinOfDoom [FTP] Survey existing FTP servers and clients to figure out how LIST works (2 days ago) http://twistedmatrix.com/bugs/issue1333 opened by exarkun [DOCUMENTATION, TRIAL] People keep asking about _trial_temp (4 days ago) http://twistedmatrix.com/bugs/issue1328 opened by jml multicast support in iocpreactor (6 days ago) http://twistedmatrix.com/bugs/issue1322 opened by ghazel Bugs Closed ______________________ ==== Urgent ==== [CORE, TESTS, TRIAL] trial doesn't run test suite all the way on win32eventreactor (2 weeks ago) http://twistedmatrix.com/bugs/issue1306 opened by exarkun, closed by jml ==== Bug ==== socket.error: (53, 'Software caused connection abort') (1 month ago) http://twistedmatrix.com/bugs/issue1274 opened by ghazel, closed by teratorn [CORE, TRIAL] trial --until-failure is broken (1 month ago) http://twistedmatrix.com/bugs/issue1263 opened by exarkun, closed by jml [CORE, TRIAL] Unexpected successes are not reported in the summary section of trial (2 weeks ago) http://twistedmatrix.com/bugs/issue1293 opened by glyph, closed by jml [TRIAL] trial shouldn't print ascii color codes on Windows (1 week ago) http://twistedmatrix.com/bugs/issue1316 opened by justinj, closed by jml [BACKPORT, CORE] FTP server does not work with ange-ftp (2 days ago) http://twistedmatrix.com/bugs/issue1332 opened by exarkun, closed by <unknown> [CORE, WIN32] iocpreactor (at least) breaks event concept, leads to crashes (1 month ago) http://twistedmatrix.com/bugs/issue1240 opened by ghazel, closed by justinj IOCP dataReceived called after connectionLost (7 days ago) http://twistedmatrix.com/bugs/issue1321 opened by justinj, closed by justinj [CORE, WIN32] iocpreator can drop a listening socket by mistake (1 month ago) http://twistedmatrix.com/bugs/issue1244 opened by ghazel, closed by justinj [CORE, TESTS] twisted.test.test_udp.testRebind makes no sense (2 weeks ago) http://twistedmatrix.com/bugs/issue1295 opened by exarkun, closed by exarkun ==== Wish ==== [WIN32] Implement spawnProcess for IOCP reactor (6 months ago) http://twistedmatrix.com/bugs/issue1008 opened by justinj, closed by justinj

1 0

[Twisted-Python] Traceback bug?
by Henning.Ramm＠mediapro-gmbh.de 15 Nov '05

15 Nov '05

This looks like a problem of Twisted's Failure: tb = a.getTraceback() File "C:\PROGRA~1\Python\Lib\SITE-P~1\twisted\python\failure.py", line 312, in getTraceback self.printTraceback(file=io, elideFrameworkCode=elideFrameworkCode, detail=detail) exceptions.TypeError: printTraceback() got an unexpected keyword argument 'detail' Should I file a bug, or is it my fault? Best regards, Henning Hraban Ramm Südkurier Medienhaus / MediaPro Support/Admin/Development Dept.

2 1

[Twisted-Python] How to talk to IM's
by Max Slimmer 14 Nov '05

14 Nov '05

I am trying to get into twisted, I have gone through the examples and have purchased and read most of the book, but am still having problems understanding just how things hang together. Any pointers to other sample code, or documentation would be great. What I am trying to do is to use the WORDS modules and be able to connect to various IM servers and through them to known IM users. I have gotten the oscardemo to work connecting to ICQ, if I log in as user who is on another users buddy list that user sees the demo logon and then I can send messages to the demo which in turn echos the messages back. All good. Now I want to run another module in the reactor which can have access to the IM (oscar) module and use it to send messages to some given IM client. I am having difficulty understanding how to get from one module to the other. The oscardemo is initiated with protocol.ClientCreator and that returns a deferred, I can then pass the deferred into the factory of my control module and have access to it, but I don't know if that will help. It seems I need access to the BOSconnection object in order to call its methods, but maybe I need to re code the demo to use defereds and don't quit know where / how to do that either. I would like to be able to connect to twisted say with telnet to a command module, and have it initiate a chat session in the IM module, it seems that I only need a single instance of the IM module, and then a means to send messages to it and get received messages from it and pass them back to some instance of command module. Any help, pointers to docs or good sample code would be much appreciated. thanks, max -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.362 / Virus Database: 267.13.0/167 - Release Date: 11/11/2005

1 0

Re: [Twisted-Python] Re: [Twisted-commits] r15100 - So I really do want this stuff
by Brian Warner 13 Nov '05

13 Nov '05

>>This is a bug, not a feature. > > Why is it a bug? I've been bitten by this behavior.. basically it means that the TestCase's __init__ method is not invoked for each test method, so you have to remember to clear status flags (say, things that are set from a Deferred callback, like maybe a .oopsSomethingFailed flag) in setUp() rather than just initialize them in __init__ . The annoying thing is that it (the __init__ approach) works fine when you run the tests one at a time, which I frequently do as I set up a brand new test. It only breaks when you run multiple tests, and even then only when the leftover results of one test get interpreted by a later one. I think part of the "clean all lingering state after every test" philosophy should include cleaning attributes from the test methods themselves (or just not re-using those objects). That said, I still want a place to amortize expensive setup across multiple tests. The buildbot tests, for example, have about half a dozen tests for each supported version control system. There is a module-level check that sees which ones you have installed (arch, baz, cvs, svn, darcs), then sets some flags which cause certain tests to be run or skipped. I don't want to re-do these checks for each test method. It may work to re-do them for each TestCase class (say, the setUpClass for the SVN TestCase class checks to see if SVN is available), but for code factoring purposes there might be multiple SVN TestCase classes, and it would be nice to avoid duplicate checks. Incidentally, these checks are the only remaining use of deferredResult() in the buildbot test suite, at least when it is run against Twisted-2.0 or newer, because they were originally run at import time. In the current code, they are performed by the first setUpClass that gets run, and the results are stashed in a module-level global to avoid running them again. If I were confident that all supported (by buildbot) versions of Twisted allowed setUpClass to return a Deferred, I could get rid of that last deferredResult() call. cheers, -Brian

1 0

Re: [Twisted-Python] Unjelly - recursion limit reached
by David K. Hess 12 Nov '05

12 Nov '05

On Nov 7, 2005, at 8:42 AM, David K. Hess wrote: > Bottom line, if all of these things are true, then I'm not sure > what is going wrong. Is it possible that jelly is at times walking > my object graph (which is interconnected to a certain degree and > changes dynamically) in such a way that it has to perform a > significant amount of recursion? In other words, maybe this isn't a > bug but I just need to increase the runtime recursion limit? > > FYI, here's a snippet of the exception: I hate replying to my own post but I'm beginning to believe that recursion is the root of the problem and want to focus on it. From the exception trace, it looks like it takes on average 3.5 stack frames to unjelly a nested object. With a default recursion limit of 1000 stack frames, this means that Twisted is limited to transferring an object graph that has a depth first traversal through it of length 285 objects (roughly speaking). You could of course easily create a linked list of 285 objects to force the situation. This is a pretty scary limitation. Is anybody working on a non- recursion based version of jelly? Am I going to have to move to Stackless Python? Thanks. Dave

3 3

Re: [Twisted-Python] Unjelly - recursion limit reached
by David K. Hess 11 Nov '05

11 Nov '05

On Nov 11, 2005, at 8:46 AM, Jean-Paul Calderone wrote: > On Fri, 11 Nov 2005 08:26:03 -0600, "David K. Hess" > <dhess(a)verscend.com> wrote: >> >> On Nov 7, 2005, at 8:42 AM, David K. Hess wrote: >>> Bottom line, if all of these things are true, then I'm not sure >>> what is going wrong. Is it possible that jelly is at times >>> walking my object graph (which is interconnected to a certain >>> degree and changes dynamically) in such a way that it has to >>> perform a significant amount of recursion? In other words, maybe >>> this isn't a bug but I just need to increase the runtime >>> recursion limit? >>> >>> FYI, here's a snippet of the exception: >> >> I hate replying to my own post but I'm beginning to believe that >> recursion is the root of the problem and want to focus on it. >> From the exception trace, it looks like it takes on average 3.5 >> stack frames to unjelly a nested object. With a default recursion >> limit of 1000 stack frames, this means that Twisted is limited to >> transferring an object graph that has a depth first traversal >> through it of length 285 objects (roughly speaking). You could of >> course easily create a linked list of 285 objects to force the >> situation. > > Linked list? Hee hee. Crazy. > >> >> This is a pretty scary limitation. Is anybody working on a non- >> recursion based version of jelly? Am I going to have to move to >> Stackless Python? > > You can raise the recursion limit. See sys.setrecursionlimit(). Yes, but with the standard Python build on Windows, you can only go up to 2000. So now, I can do 570 objects. I'm working on trying to track down editbin.exe so I can raise the stack size on python.exe and see how much farther I can go. > Stackless has nothing to do with this. It won't help fix this > problem. Are we talking about the same thing? The key architectural feature of Stackless Python is "practically unlimited recursion". Forget about all the tasklet and channel stuff. > The real way to "fix" this is to re-implement the jellying process > iteratively. Or, don't use such deeply nested objects. My object graph is a natural and clean data structure. I don't want to lose that just because of limitations in my tools. Is this currently being addressed or in a road map? Am I unknowingly volunteering for it? :-o Dave > > Jean-Paul > > _______________________________________________ > Twisted-Python mailing list > Twisted-Python(a)twistedmatrix.com > http://twistedmatrix.com/cgi-bin/mailman/listinfo/twisted-python

1 0