| 00:00.18 | *** join/#storm wallflower (n=wallflow@ip-205-246-113-216.pool.grokthis.net) |
| 01:50.41 | *** join/#storm kov (n=kov@debian/developer/kov) |
| 02:39.28 | *** join/#storm cornbread (n=max@adsl-71-150-248-78.dsl.mdsnwi.sbcglobal.net) |
| 02:39.32 | *** join/#storm Sidnei (n=Sidnei@plone/dreamcatcher) |
| 02:39.52 | cornbread | Hey guys. I am trying to use storm in a parallel program (using the processing or pp modules), but I am having pickle issues with my models |
| 02:40.21 | cornbread | whenever I use a model in any parallel code, I get errors saying that I can't pickle thread.lock objects or weakref objects on these modules. |
| 02:40.41 | cornbread | Does anyone have any experience using storm with parallel (multi-process) python? |
| 03:28.41 | *** join/#storm cornbread (n=max@adsl-76-208-70-61.dsl.mdsnwi.sbcglobal.net) |
| 04:03.29 | *** join/#storm cornbread (n=max@adsl-76-208-70-61.dsl.mdsnwi.sbcglobal.net) |
| 06:32.50 | *** join/#storm jukart (i=lovely@lsfw01.lovelysystems.com) |
| 06:39.25 | *** join/#storm cornbread (n=max@adsl-76-208-70-61.dsl.mdsnwi.sbcglobal.net) |
| 07:36.58 | *** join/#storm cornbread (n=max@adsl-76-208-70-61.dsl.mdsnwi.sbcglobal.net) |
| 09:33.56 | jamesh | cornbread: you probably need to implement one of these methods on the class you're trying to pickle: http://docs.python.org/library/pickle.html#pickling-and-unpickling-normal-class-instances |
| 09:34.50 | jamesh | cornbread: that said, storing pickles in the database is not generally the best option -- you can't easily query on them |
| 09:54.54 | *** join/#storm mcella (n=mcella@ip-143-37.sn3.eutelia.it) |
| 10:05.39 | *** join/#storm lightyear (n=ben@o.bcn.fluendo.net) |
| 10:06.28 | *** join/#storm vds (n=vds@78.13.139.106) |
| 10:11.05 | vds | hi all |
| 10:11.40 | vds | I have some problem trying to understand how WHERE works in a join |
| 10:13.18 | vds | the tutorial has just a very simple example, I wonder how can a test something like the value of column in a join is null |
| 10:13.51 | vds | or if the value of a column is in a list |
| 10:45.53 | *** join/#storm niemeyer (n=niemeyer@201-25-142-51.ctame705.dsl.brasiltelecom.net.br) |
| 11:19.25 | *** join/#storm rockstar (n=rockstar@canonical/launchpad/rockstar) [NETSPLIT VICTIM] |
| 11:35.46 | *** join/#storm artista_frustrad (n=artista_@201-40-200-100.ctame704.dsl.brasiltelecom.net.br) |
| 12:06.36 | *** join/#storm kov (n=kov@debian/developer/kov) |
| 13:00.14 | *** join/#storm joerg (n=joerg@p548894BA.dip0.t-ipconnect.de) |
| 13:00.16 | joerg | hey |
| 13:00.36 | *** join/#storm fcorrea (n=fcorrea@189.72.251.144) |
| 13:01.03 | joerg | I just did apt-get install python-storm on my ubuntu 8.10 box.... |
| 13:01.15 | joerg | anybody did that before? |
| 13:01.37 | joerg | a "from storm.locals import *" doesn't really work.... |
| 13:02.13 | fcorrea | hello there. Is there any examples on how to sort a query on date for example? I am trying to find better ways on filtering, say, week |
| 13:24.19 | therve | joerg: what do you mean? |
| 13:24.56 | therve | fcorrea: you'd have to use order_by with some tricky functions |
| 13:25.08 | therve | fcorrea: which is probably specific to the db you're using |
| 13:31.09 | fcorrea | therve: thanks. I think I got that sorted out by filtering the resultset. Will try to find better ways later |
| 13:59.10 | *** join/#storm fcorrea (n=fcorrea@189.72.251.144) |
| 14:17.04 | *** join/#storm goschtl (n=goschtl@p5B0BBF56.dip.t-dialin.net) |
| 14:23.12 | cornbread | jamesh: thanks. I am not storing the pickled instances, they are being used for IPC since the parallel processing creates multiple processes instead of threads and must pickle the state of each process I assume (I am new to it) |
| 14:28.15 | joerg | therve, lol? |
| 14:28.39 | therve | joerg: no? |
| 14:54.38 | *** join/#storm cornbread (n=max@adsl-76-208-70-61.dsl.mdsnwi.sbcglobal.net) |
| 14:56.30 | joerg | therve, it didn't find the lib |
| 14:56.35 | joerg | but I solved it. |
| 14:56.42 | joerg | it was because I named my file storm.py |
| 15:01.11 | joerg | can I somehow change the charset for my mysql connection? |
| 15:01.36 | joerg | I am querying the db with storm, but the strings I get back are not displayed correctly. |
| 15:01.40 | joerg | they are in utf8 in the DB |
| 15:01.52 | joerg | and my terminal is set to utf8 as well. |
| 15:02.06 | joerg | with the mysql command line client it works perfectly. |
| 15:30.58 | *** join/#storm Sidnei_ (n=Sidnei@201-11-209-206.cslce701.dsl.brasiltelecom.net.br) |
| 15:34.04 | vds | I have a connection to the database in a method and a different one in the tests |
| 15:34.29 | vds | when I update the database from the method the tests don't see the changes |
| 15:35.08 | vds | is it something strange or that is correct and I'm missing something ? |
| 15:37.03 | sidnei | vds: do you have transactions involved? sounds like a side effect of transaction isolation level |
| 15:37.14 | vds | sidnei: nope |
| 15:39.10 | vds | the test opens its own connection |
| 15:39.18 | vds | than calls the method |
| 15:39.27 | vds | which actually opens a new one |
| 15:39.31 | vds | make the canges |
| 15:39.35 | vds | changes |
| 15:39.49 | vds | then the control is back to the test |
| 15:40.08 | vds | which checks if the changes are correct |
| 15:40.18 | vds | but can't find any changes... |
| 15:40.51 | vds | in the method I do store.add(obj); store.commit(); store.flush() |
| 15:41.10 | vds | sorry first flush than commit |
| 15:49.41 | *** join/#storm jamesh_ (n=james@canonical/launchpad/jamesh) |
| 15:50.21 | *** join/#storm jukart_ (i=lovely@lsfw01.lovelysystems.com) |
| 16:01.45 | sidnei | not much i can help with, i think |
| 16:07.49 | vds | sidnei: thanks :) |
| 16:14.52 | vds | niemeyer or jml can I ask you the same question? |
| 16:14.55 | *** join/#storm andrea-bs (n=andrea-b@ubuntu/member/beeseek.developer.andrea-bs) |
| 16:15.47 | niemeyer | vds: That's the expected way it should work, if I understand what you're saying |
| 16:16.08 | niemeyer | vds: You're working with postgres, right? |
| 16:16.13 | vds | niemeyer: yes |
| 16:16.29 | niemeyer | vds: So you probably have a serializable transaction |
| 16:17.34 | niemeyer | vds: Which means changes are only seen in new transactions after they are committed, and on going transactions won't see changes performed while the current transaction is running |
| 16:20.52 | vds | niemeyer: if I do store.close() in the method and create a new connection in the test just after calling the method, that would work? |
| 16:21.31 | niemeyer | vds: Yes, but you don't have to close the connection.. just get a fresh transaction by committing/rolling back |
| 16:22.26 | sidnei | just like zope, vds. *wink* |
| 16:23.06 | vds | niemeyer: testresources will like it? |
| 16:23.36 | vds | sidnei: :) |
| 16:23.36 | vds | trying... |
| 16:23.36 | niemeyer | vds: I don't know what that means :( |
| 16:23.49 | niemeyer | vds: What's testresources, and what does it have to like? :-) |
| 16:24.27 | vds | testresources is what I'm using to create my tescases |
| 16:25.37 | *** join/#storm lightyear (n=ben@o.bcn.fluendo.net) |
| 16:26.26 | vds | woow worked! |
| 16:27.08 | vds | niemeyer: thanks! as usual your help is invaluable! |
| 16:33.21 | niemeyer | vds: Sweet! You're welcome :-) |
| 16:46.56 | *** join/#storm fcorrea (n=fcorrea@189.72.251.144) |
| 17:08.35 | *** join/#storm Sidnei_ (n=Sidnei@201-11-209-206.cslce701.dsl.brasiltelecom.net.br) |
| 17:31.05 | *** join/#storm Sidnei_ (n=Sidnei@201-11-209-206.cslce701.dsl.brasiltelecom.net.br) |
| 17:58.02 | *** join/#storm lightyear (n=ben@o.bcn.fluendo.net) |
| 18:17.51 | *** join/#storm jamesh_ (n=james@canonical/launchpad/jamesh) |
| 18:47.58 | *** join/#storm andrea-bs (n=andrea-b@ubuntu/member/beeseek.developer.andrea-bs) |
| 18:49.27 | *** join/#storm jamesh__ (n=james@canonical/launchpad/jamesh) |
| 19:37.39 | *** join/#storm Sidnei (n=Sidnei@plone/dreamcatcher) |
| 19:47.47 | *** join/#storm jukart (n=jukart@d91-130-142-16.cust.tele2.at) |
| 21:02.26 | *** join/#storm Sidnei (n=Sidnei@plone/dreamcatcher) |
| 21:48.48 | jkakar | rockstar: Can I help? :) |
| 21:49.35 | rockstar | jkakar, well, not now. I need to be doing work, and this isn't work related. |
| 21:49.53 | rockstar | (I glanced at it while cooking lunch, and the problem is just stewing in my head...) |
| 21:50.24 | jkakar | rockstar: Ah, okay. Feel free to ping when you're back at it, will be happy to try and help out if I can. |
| 21:50.38 | rockstar | jkakar, I definitely will be hacking on it tonight. |
| 21:52.26 | jkakar | rockstar: I'll be teaching a drum lesson and then jamming with some guys... but maybe you're night will happen before mine. |
| 21:52.30 | jkakar | your, even. |
| 21:52.54 | rockstar | jkakar, I think you're only an hour behind me. Worse comes to worse, I'll post the mailing list. |
| 21:53.27 | jkakar | rockstar: Okay. I'm in -8, FYI. |
| 21:53.53 | rockstar | Yea, I'm -7. We're the last of the Canoni-folks to experience a day. |
| 21:54.40 | jkakar | Yes... -7 must be a similarly sucky IRC timezone as -8. :) |
| 21:55.30 | rockstar | Most of my cow orkers are +11 and +13 though. :) |
| 21:56.04 | jkakar | +13? Does that come about because of DST? |
| 22:15.13 | *** join/#storm mcella (n=mcella@ip-143-37.sn3.eutelia.it) |
| 22:53.57 | *** join/#storm artista_frustrad (n=artista_@201-14-24-77.ctame704.dsl.brasiltelecom.net.br) |