Mailman 3 March 2010 - Twisted

[Twisted-Python] R: Twisted Python on Nokia E71x (Running the phone Python Interpreter)
by lormayna＠libero.it 06 Mar '10

06 Mar '10

>Date: Thu, 4 Mar 2010 15:57:43 -0600 >From: Prince Riley <wmarketing3(a)gmail.com> >Subject: [Twisted-Python] Twisted Python on Nokia E71x (Running the > phone Python Interpreter) >To: twisted-python(a)twistedmatrix.com >Message-ID: > <fa04eb751003041357o7ca889a9uc3985667c34731b0(a)mail.gmail.com> >Content-Type: text/plain; charset="iso-8859-1" > >Has anyone experimented with twisted-python on the Nokia E71x or E72 >smartphones? > >Prince Riley Hi, I had tried to use Python for S60 (not Twisted, but standard socket module) writing a NNTP client, but I found many problem specially about networking. There is no high-layer abstraction layer for networking, so you have to implement all the logic from low level. I tried to use select (that may be useful for doing asynchronous networking apps), but is not working correctly on s60 phones. There is no porting for Twisted. My experience is that PyS60 is not mature for a production environment.

3 3

[Twisted-Python] Escaping variable names
by Kamil Wasilewski 05 Mar '10

05 Mar '10

Hi, Ive got an issue where a variable name needs to have a minus sign (-) in it. from SOAPpy import WSDL wsdlFile = "http://webapi.allegro.pl/uploader.php?wsdl" server = WSDL.Proxy(wsdlFile) server.soapproxy.config.argsOrdering = {'doGetCountries': ['country-code', 'webapi-key'] } server.doGetCountries(country-code=COUNTRYID}, webapi-key=WEBAPIKEY) The above method allows me to specify the order and name of variables being sent, but i have a problem since the variables have a "-" sign in the middle, is there a way to escape it in the variable name or get around the problem? Thanks!

3 5

[Twisted-Python] SSHSessionProcessProtocol.inConnectionLost() behavior and Git over Conch
by Bo Shi 05 Mar '10

05 Mar '10

Hi All, I've been struggling with this issue off and on for the better part of a month now. Having implemented a simple git SSH server using some of the Conch examples in the official documentation, I encountered an issue that would cause a git client to throw an error only part of the time. The little program to reproduce the issue is found at http://gist.github.com/321403 (raw version for download) http://gist.github.com/raw/321403/82ab2111a2709c8fe50a77aabb08565749087408/… and can be executed to start the sample server. Run a command like # The /abspath/to/git/repo should be readable by the server user $ git clone ssh://user@localhost:2222/abspath/to/git/repo (password "user") To test the server. On my workstation, I am able to reproduce the error at least once every 5 tries. It's definitely not consistent. The behavior I see is attached at the end of this email. The client fails due to a premature inConnectionLost() call in the SSHSessionProcessProtocol that sends an EOF. As a workaround, when TraceProcessProtocol.inConnectionLost() is overriden to do nothing, the client error goes away. This is somewhat foreign territory for me so I'm not sure whether it's a bug in git or whether it's a bug in the twisted ProcessProtocol implementation. RFC 4254 isn't terribly helpful here: 5.3. Closing a Channel When a party will no longer send more data to a channel, it SHOULD send SSH_MSG_CHANNEL_EOF. So on the surface the current behavior of sending an EOF appears fine, however, I can't really find any definitive cases of this type of problem popping up via, say, default OpenSSH/git combinations. Any advice? Is the gitconnbug.py implementation flawed? Should I open a ticket? Any git experts know of a case where git-upload-pack might close it's stdout pipe? Thanks, Bo Successful case (server side logging, timestamp cut out): [SSHChannel session (0) on SSHService ssh-connection on SSHServerTransport,1,127.0.0.1] executing command "git-upload-pack '/Users/bshi/sandbox/poop'" [-] TPP.outReceived(...) 199 bytes [-] TPP.outReceived(...) 146 bytes [-] TPP.outReceived(...) 8 bytes [-] TPP.outReceived(...) 33 bytes [-] TPP.outReceived(...) 41 bytes [-] TPP.outReceived(...) 77 bytes [-] TPP.outReceived(...) 1228 bytes [-] TPP.outReceived(...) 8192 bytes [-] TPP.outReceived(...) 4567 bytes [-] TPP.inConnectionLost() [-] sending eof [-] exitCode: 0 Successful case (client-side shell session): $ GIT_TRACE=1 git clone ssh://user@localhost:2222/Users/bshi/sandbox/poop trace: built-in: git 'clone' 'ssh://user@localhost:2222/Users/bshi/sandbox/poop' Initialized empty Git repository in /tmp/poop/.git/ trace: run_command: 'ssh' '-p' '2222' 'user@localhost' 'git-upload-pack '\''/Users/bshi/sandbox/poop'\''' user@localhost's password: trace: run_command: 'index-pack' '--stdin' '-v' '--fix-thin' '--keep=fetch-pack 763 on Bo-Shis-MacBook-Pro.local' trace: exec: 'git' 'index-pack' '--stdin' '-v' '--fix-thin' '--keep=fetch-pack 763 on Bo-Shis-MacBook-Pro.local' trace: exec: 'git-index-pack' '--stdin' '-v' '--fix-thin' '--keep=fetch-pack 763 on Bo-Shis-MacBook-Pro.local' trace: run_command: 'git-index-pack' '--stdin' '-v' '--fix-thin' '--keep=fetch-pack 763 on Bo-Shis-MacBook-Pro.local' remote: Counting objects: 50, done. remote: Compressing objects: 100% (35/35), done. remote: Total 50 (delta 16), reused 43 (delta 14) Receiving objects: 100% (50/50), 12.36 KiB, done. Resolving deltas: 100% (16/16), done. Failure case (server side): [SSHChannel session (0) on SSHService ssh-connection on SSHServerTransport,2,127.0.0.1] executing command "git-upload-pack '/Users/bshi/sandbox/poop'" [-] TPP.outReceived(...) 345 bytes [-] TPP.outReceived(...) 8 bytes [-] TPP.outReceived(...) 33 bytes [-] TPP.outReceived(...) 8192 bytes [-] TPP.inConnectionLost() [-] sending eof [-] TPP.outReceived(...) 5903 bytes [-] exitCode: 0 [-] sending request exit-status [-] sending close 0 Failure case ("trace:" messages similar, cut out for readability): $ GIT_TRACE=1 git clone ssh://user@localhost:2222/Users/bshi/sandbox/poop user@localhost's password: ... ... remote: Counting objects: 50, done. fatal: The remote end hung up unexpectedly fatal: early EOF fatal: index-pack failed

4 4

[Twisted-Python] Twisted 9.0.0?
by Mark van Lent 05 Mar '10

05 Mar '10

Hi everyone! I am a bit puzzled about the way new version of Twisted is released. Or better: the way the old version seems to be taken out of service. Last weeks I've been developing agains version Twisted 9.0.0. Yesterday I wanted to deploy my code to my production server, but to my surprise I could not easy_install that version anymore: $ easy_install Twisted==9.0.0 Searching for Twisted==9.0.0 Reading http://pypi.python.org/simple/Twisted/ Reading http://www.twistedmatrix.com Reading http://twistedmatrix.com/products/download Reading http://twistedmatrix.com/projects/core/ Reading http://twistedmatrix.com/ No local packages or download links found for Twisted==9.0.0 Best match: None (Okay, actually I'm using buildout to setup my environment, but the above demonstrates my problem more clearly.) I'm sure the new version of Twisted is better and improved but I still wonder why version 9.0.0 has effectively been decommissioned. Or am I missing something and is it still possible to easy_install version 9.0.0? Thanks, Mark -- Mark van Lent m.van.lent(a)zestsoftware.nl

8 18

[Twisted-Python] Twisted Python on Nokia E71x (Running the phone Python Interpreter)
by Prince Riley 05 Mar '10

05 Mar '10

Has anyone experimented with twisted-python on the Nokia E71x or E72 smartphones? Prince Riley

2 1

[Twisted-Python] IMAP4 Client extrange behavior
by César García 04 Mar '10

04 Mar '10

Hello all, I have this imap4 client http://www.pastebin.com/m4e387f1a most of the code is barrowed from a IRC friend, bob_f , it works fine but as soon as more that 1 email arrrives to the inbox, it only prints the first new messages but not the others, I wonder what am I doing wrong Thanks a lot Here is the code: #!/usr/bin/env python #coding=utf-8 """ Client de IMAP4 que descarga contenido del INBOX de una cuenta en especifico para luego extraer el numero de telefono que debe venir en los correos enviados """ import StringIO import sys from Config import retCredentials from twisted.internet.task import LoopingCall from twisted.internet import protocol from twisted.internet import defer from twisted.mail import imap4 from twisted.python import util from twisted.python import log from twisted.internet import reactor debug = 1 class IMAP4Client(imap4.IMAP4Client): def serverGreeting(self,caps): """ Metodo llamado cuando el servidor contesta """ if debug: print "On serverGreeting" self.serverCapabilities = caps if self.greetDeferred is not None: d, self.greetDeferred = self.greetDeferred, None d.addCallback(self.cbLogin) d.callback(self) def cbLogin(self, proto): """ Callback to IMAP login """ if debug: print "On Login" login = self.login(self.factory.username, self.factory.password) login.addCallback(self.startPolling) def startPolling(self, proto): """ Callback to poll every x seconds """ call = LoopingCall(self.selectMailbox,mailbox="INBOX") call.start(20, now=True) def selectMailbox(self, mailbox): """ Select the mailbox to examin """ if debug: print "On selectMailbox" mailbox = self.factory.mailbox return self.select(mailbox).addCallback(self.cbSelectSuccess) def cbSelectSuccess(self, selected): """ Examine the INBOX mailbox for new mails """ if debug: print "On cbSelectSuccess" self.messageCount = selected['EXISTS'] print "Messages: ", self.messageCount unseen = selected['EXISTS'] - selected['RECENT'] if selected['RECENT'] == 0: print "No new messages" return return self.fetchMessage("%s:*" % (unseen) ).addCallback(self.cbProcMessages) def cbProcMessages(self,messages): print messages class IMAP4ClientFactory(protocol.ClientFactory): protocol = IMAP4Client def __init__(self, username, password, onConn): self.username = username self.password = password self.mailbox = 'INBOX' self.onConn = onConn def buildProtocol(self,addr): if debug: print "On buildProtocol" p = self.protocol() p.factory = self p.greetDeferred = self.onConn auth = imap4.CramMD5ClientAuthenticator(self.username) p.registerAuthenticator(auth) return p def clientConectionFailed(self, connector, reason): d, self.onConn = self.onConn, None d.errback(reason) def ebConnection(reason): log.startLogging(sys.stdout) log.err(reason) reactor.stop() PORT = 143 RESULT = "INBOX" def main(): credentials = retCredentials() hostname = credentials['server'] username = credentials['mailbox'] password = util.getPassword('IMAP4 Password: ') onConn = defer.Deferred( ).addErrback(ebConnection ) factory = IMAP4ClientFactory(username, password, onConn) reactor.connectTCP(hostname, PORT, factory) reactor.run() if __name__ == "__main__": main() -- http://celord.blogspot.com/

1 1

[Twisted-Python] debugging a memory leak
by Alec Matusis 03 Mar '10

03 Mar '10

I modified a tolerably leaking (about 40MB/day) Twisted server: when the new code was pushed, the memory leak became catastrophic (about 100MB/hr). We could tolerate 40MB/day, but the new code needs to be debugged. First, I reverted to the old version, that is leaking 40MB/day (The leak rate is actually proportional to the number of new connections per second, (which correlates with the CPU utilization of the process): if CPU as measured by $top jumps to >90%, the leak can accelerate to 50MB/hr) I took two steps to debug the leak: 1) Using guppy/heapy via manhole: >>> hp = hpy() >>> h = hp.heap() >>> h Partition of a set of 1157084 objects. Total size = 140911144 bytes. Index Count % Size % Cumulative % Kind (class / dict of class) 0 785798 68 48463680 34 48463680 34 str 1 7357 1 24660664 18 73124344 52 dict of service.TagSession 2 11735 1 12298280 9 85422624 61 dict of twisted.internet.base.DelayedCall 3 7377 1 7731096 5 93153720 66 dict of twisted.internet.tcp.Server 4 7375 1 7729000 5 100882720 72 dict of protocols.TagProtocol 5 30925 3 7174600 5 108057320 77 __builtin__.set 6 9193 1 6373336 5 114430656 81 dict (no owner) 7 15557 1 3396904 2 117827560 84 list 8 44833 4 3227976 2 121055536 86 types.BuiltinFunctionType 9 38142 3 3051360 2 124106896 88 types.MethodType <328 more rows. Type e.g. '_.more' to view.> Note that the total size of all objects is 140911144 bytes The 1st, 3d and 4th items in this list show the actual number of connected clients. I wait for about 48 hrs, and then execute the same command, and I see approximately the same Total size, of 130MB. So the total size that Heapy sees via the manhole is stable, fluctuating around 140MB. The problem is that the total RSS size of the process visible by the OS is much larger, it is 871680KB = 851MB: $ps -o pid,vsz,rss,sz,size -p 11236 PID VSZ RSS SZ SZ 11236 1303180 871680 325795 1174388 It is this total RSS size that keeps leaking at 40MB per day. As far as I understand, this means that this is not the problem with purely Python code, since Heapy shows that the total size of all Python objects is more or less constant. Is this a correct assumption? 2) So now I turn to valgrind. I am no expert in using valgrind, so what I did was based only on general logic/rough guesses. Since I cannot run this under valgrind on a production machine due to performance reasons, I recompile python on the staging machine: ./configure --enable-shared --without-pymalloc --prefix=/nail/encap/python-2.6.4-valgrind I then follow the instructions in http://svn.python.org/projects/python/trunk/Misc/README.valgrind Then I run twistd process like this: valgrind --tool=memcheck --suppressions=/nail/sys/src/Python-2.6.4/Misc/valgrind-python.supp --leak-check=full --log-file=/tmp/valgrind.log /usr/local/bin/twistd --no_save --reactor=epoll --pidfile=logs/tagserv.pid --logfile=logs/tagserv.log --python=tagserv.py The memory for the process shown by the OS is now 5x normal, and the performance is about 5x worse, since it's running inside valgrind's synthetic CPU. Because this is done on the staging box where I cannot accurately reproduce the real load, the memory leaks from simulated load seen by $ps -o rss are pretty small, about 1 to 10MB. Still, I am interested in finding out what they are. Now I encounter the problems with my understanding of how to use valgrind. This may not be the appropriate list for this, but perhaps someone could recognize the problem. When I start the server, about 240 lines is written to valgrind log file, --log-file=/tmp/valgrind.log When I shut it down, it adds another 100 lines. No matter what I do in between, it always results in the log file with exactly 343 lines. I can have server runs with a leak of 1MB, or with 10MB, but in the end, I get pretty much the same log file. Moreover, when look for all lost memory reports: $grep lost valgrind.log.1 17,352 bytes in 31 blocks are possibly lost in loss record 49 of 62 203,312 bytes in 478 blocks are possibly lost in loss record 57 of 62 definitely lost: 0 bytes in 0 blocks. possibly lost: 220,664 bytes in 509 blocks. 64 bytes in 2 blocks are definitely lost in loss record 12 of 63 17,352 bytes in 31 blocks are possibly lost in loss record 50 of 63 203,824 bytes in 479 blocks are possibly lost in loss record 58 of 63 definitely lost: 64 bytes in 2 blocks. possibly lost: 221,176 bytes in 510 blocks. 47 bytes in 1 blocks are definitely lost in loss record 8 of 63 128 bytes in 4 blocks are definitely lost in loss record 16 of 63 584 (104 direct, 480 indirect) bytes in 2 blocks are definitely lost in loss record 26 of 63 1,008 bytes in 6 blocks are definitely lost in loss record 31 of 63 22,296 bytes in 41 blocks are possibly lost in loss record 50 of 63 183,368 bytes in 381 blocks are possibly lost in loss record 59 of 63 definitely lost: 1,287 bytes in 13 blocks. indirectly lost: 480 bytes in 20 blocks. possibly lost: 205,664 bytes in 422 blocks. If I add up all those numbers, I get less than 1MB. How do I track down the 10MB leak? Are there any alternative strategies in finding this leak?

6 11

[Twisted-Python] Does the default session expiry work?
by markscottwright＠gmail.com 03 Mar '10

03 Mar '10

I'm using twisted 8.2 and (after reading glyph's latest "Twisted in 60 seconds" entry), I've realized that my sessions never expire. I never added any session expiration code, but I thought that the default code expires after 15 minutes. In the below example, I modified twisted\web\server to make Session.sessionTimeout be 30 seconds (I also put a print statement in Session's ctor, as the log shows), then did two GETs separated by more than 30 seconds. As the log shows, it looks like the exact same session uid is being returned. I would expect it to be deleted by Session.expire, since time - lastModifiedTime > 30. Am I missing something? import sys from twisted.web import server, static from twisted.web.resource import Resource from twisted.web.server import Session, NOT_DONE_YET from twisted.internet import reactor, ssl from twisted.python import log class TestSessionResource(Resource): isLeaf = True def render_GET(self, request): session = request.getSession() print session.sessionTimeout print session.uid print session return "<html><body><p>Session Test</p></body></html>" root = TestSessionResource() reactor.listenTCP(9000, server.Site(root)) log.startLogging(sys.stdout) reactor.run() 2010-03-02 12:16:49-0600 [-] Log opened. 2010-03-02 12:16:52-0600 [HTTPChannel,0,127.0.0.1] making a new session!!! 2010-03-02 12:16:52-0600 [HTTPChannel,0,127.0.0.1] 30 2010-03-02 12:16:52-0600 [HTTPChannel,0,127.0.0.1] 338e4aa450f671b2f0115513737e643b 2010-03-02 12:16:52-0600 [HTTPChannel,0,127.0.0.1] <twisted.web.server.Session instance at 0x01042AD0> 2010-03-02 12:16:52-0600 [HTTPChannel,0,127.0.0.1] 127.0.0.1 - - [02/Mar/2010:18:16:51 +0000] "GET / HTTP/1.1" 200 45 "-" "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/532.5 (KHTML, like Gecko) Chrome/4.0.249.89 Safari/532.5" 2010-03-02 12:16:52-0600 [HTTPChannel,0,127.0.0.1] 30 2010-03-02 12:16:52-0600 [HTTPChannel,0,127.0.0.1] 338e4aa450f671b2f0115513737e643b 2010-03-02 12:16:52-0600 [HTTPChannel,0,127.0.0.1] <twisted.web.server.Session instance at 0x01042AD0> 2010-03-02 12:16:52-0600 [HTTPChannel,0,127.0.0.1] 127.0.0.1 - - [02/Mar/2010:18:16:51 +0000] "GET /favicon.ico HTTP/1.1" 200 45 "-" "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/532.5 (KHTML, like Gecko) Chrome/4.0.249.89 Safari/532.5" 2010-03-02 12:17:45-0600 [HTTPChannel,0,127.0.0.1] 30 2010-03-02 12:17:45-0600 [HTTPChannel,0,127.0.0.1] 338e4aa450f671b2f0115513737e643b 2010-03-02 12:17:45-0600 [HTTPChannel,0,127.0.0.1] <twisted.web.server.Session instance at 0x01042AD0> 2010-03-02 12:17:45-0600 [HTTPChannel,0,127.0.0.1] 127.0.0.1 - - [02/Mar/2010:18:17:44 +0000] "GET / HTTP/1.1" 200 45 "-" "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/532.5 (KHTML, like Gecko) Chrome/4.0.249.89 Safari/532.5" 2010-03-02 12:17:45-0600 [HTTPChannel,0,127.0.0.1] 30 2010-03-02 12:17:45-0600 [HTTPChannel,0,127.0.0.1] 338e4aa450f671b2f0115513737e643b 2010-03-02 12:17:45-0600 [HTTPChannel,0,127.0.0.1] <twisted.web.server.Session instance at 0x01042AD0> 2010-03-02 12:17:45-0600 [HTTPChannel,0,127.0.0.1] 127.0.0.1 - - [02/Mar/2010:18:17:44 +0000] "GET /favicon.ico HTTP/1.1" 200 45 "-" "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/532.5 (KHTML, like Gecko) Chrome/4.0.249.89 Safari/532.5"

7 9

[Twisted-Python] Reverse HTTP proxy code review
by Fred C 02 Mar '10

02 Mar '10

Hello, I am totally new with Twisted and this the first code I have ever written using that package. One of my main problem was to figure out how to fit things together and I am not even sure I understand everything I am doing here. Anyway, for a project I have to write a proxy that figure out where is located a resource by requesting a database. then figure out if the resource is accessible and them open a proxy connection to download that resource. I was wandering if there is someone in that mailing list who can review that code and tell me how I can improve things. I know I am not accessing the database the right way but I can't figure out how to work with the Callbacks. Any help or advice will be very much appreciated. The code is available here: http://pastebin.com/BGcqzh4Y -fred- -- http://kiq.me/oi

2 3

[Twisted-Python] PyCon 2010 sprint report
by Jonathan Lange 02 Mar '10

02 Mar '10

Residents of the once-great city of Atlanta are only now beginning to piece their lives back together after the event that Glyph Lefkowitz described as "the best Twisted sprint ever". Quite apart from the unprecedented levels of wanton destruction, Twisted's PyCon 2010 sprint was also better attended than any previous sprints, which have laid waste to the outskirts of Boston, inner-city Dallas and regions of Australia. At its peak, there were eighteen people seated around the two tables where the sprint took place. People even joined in remotely, working with the Atlanta team through #twisted on freenode. Although buried deep within a secret bunker in Atlanta, the sprint still managed to attract a fair number of new contributors and Twisted users who wanted to help out. Attendees who made themselves known to your correspondent include: * Jonathan Lange * Jean-Paul Calderone * David Reid * Thomas Hervé * Tim Couper * Pavel Pergamenshchik * Maciej Fijalkowski * Glyph Lefkowitz * Itamar Turner-Trauring * Terry Jones jesstess, ralphm and spiv also participated remotely. Attendees all had immense fun and expressed a desire to have another sprint soon. One observed that it had been far too long since giant mecha robot warriors descended on Paris. Twisted prevails, jml More simply, we had a great time at the sprint and got a ridiculous amount of stuff done. We fixed many, many bugs, including all of the known regressions since the 9.0 release. Once we fixed the regressions, we started work on getting the first 10.0 pre-release out the door. The pre-release probably would not have been possible without the sprint since it allowed the new release manager ready access to Jean-Paul, who has quite a lot of knowledge stuck in his head. One result of this is a new release process document, which is being developed at http://twistedmatrix.com/trac/wiki/ReleaseProcess alongside the actual 10.0 release. We also reached an in-principle agreement to aim for a release every three months. This means that although hardly any of the good work from the sprint is in the 10.0 release, you won't have to wait long for it to appear in 10.1. Hopefully. I plan write up how this will work as soon as the 10.0 release is done. The sprint proved to be a perfect opportunity to talk about some of the more gutly bits of Twisted, including Deferred cancellation (see #990) and endpoints (see #1442). The former is done, and the latter has made good progress. Stay tuned. Thomas Hervé did a lot of work on upgrading our Trac instance, which was getting kind of old and currently crashes all of the time. We talked a bit about how we could make it easier for people to contribute to Twisted without compromising our high standards of quality. We put as many of the ideas that we could remember on to http://twistedmatrix.com/trac/wiki/ContributionIdeas. Maciej Fijalkowski was at the sprint working on getting PyPy benchmarks for Twisted and on making PyPy work faster with Twisted. Personally, it's really exciting to see PyPy live up to its promise of a faster Python, particularly with a big codebase like Twisted. As I said, we fixed lots of bugs. Here's the complete list: * #4179 - HTTP proxy hanging in Twisted 9.0 * #3056 - twisted.spread.pb.IUsernameMD5Password's docstring should say it accepts an MD5 digest of a password, not plaintext * #3289 - startLoggingWithObserver raises IndexError * #4178 - Function singleSearchStep was renamed to _singleSearchStep not everywhere * #4192 - Move the content from exarkun's Twisted Web in 60 Seconds series into the Twisted Web howto * #4284 - t.e.adbapi leaks threads * #4282 - _release.filePathDelta is broken * #3718 - twisted.internet.posixbase._Waker not defined on Jython * #4143 - twisted.web.wsgi incorrectly unquotes QUERY_STRING * #4255 - BaseLogFile should provide a method to reopen logs * #972 - Add linux inotify support to twisted core * #2367 - twisted.test.test_failure.FailureTestCase.testBrokenStr breaks #trial in debug mode * #1083 - gtk2reactor thread initialization confuses python-2.3 * #3462 - FTP server: upload should support async close * #990 - Deferred cancellation * #733 - twisted's SIGCHLD handler breaks popen. * #2376 - release-twisted should generate guides with links to API docs * #4272 - flavors.py still talks about getObjectAt * #4172 - UDP crashes python.exe when using IOCPReactor * #3925 - test_addresses fails with Glib2 and Gtk2 reactors * #4255 - BaseLogFile should provide a method to reopen logs * #4293 - Update links in testing documentation * #4311 - XMPP Stanza Error 'remote-server-timeout' yield no type and code * #3811 - Allow aborting HTTP client connections in the _newclient API We also worked on: * Getting rid of 3k warnings and generally doing 2to3 * Event support for iocpreactor * Making Deferreds more iterative (see #411) * Twisted Web features and bug fixes * Fixing obscure Windows/gtk2 compatibility issues Thank you very much to PyCon US, the Python Software Foundation and the many sponsors of the Twisted Software Foundation for their hard-work and support in making this all possible. Thanks also to the attendees for making it such a blast. If you want to encourage more of this sort of behaviour, then please donate to the Twisted Software Foundation: http://twistedmatrix.com/trac/wiki/TwistedSoftwareFoundation#HowtoDonate jml

3 3