| 00:01.06 | *** join/#storm wallflower (n=wallflow@ip-205-246-113-216.pool.grokthis.net) |
| 04:33.12 | *** join/#storm oubiwann (n=oubiwann@97-119-4-38.omah.qwest.net) |
| 04:40.27 | *** join/#storm jfishman (i=jfishman@139.140.73.239) |
| 05:24.47 | *** join/#storm jukart (i=lovely@lsfw01.lovelysystems.com) |
| 05:50.14 | *** join/#storm rreis (n=rreis@201-13-68-86.dsl.telesp.net.br) |
| 05:50.25 | *** join/#storm rreis_ (n=rreis@201-13-68-86.dsl.telesp.net.br) |
| 06:14.40 | *** join/#storm dobee (i=lovely@lsfw01.lovelysystems.com) |
| 09:01.58 | *** join/#storm lightyear (n=ben@nat/fluendo/x-17a7435647585625) |
| 11:07.17 | *** join/#storm andrea-bs (n=andrea@ubuntu/member/beeseek.developer.andrea-bs) |
| 11:15.26 | *** join/#storm jukart (i=lovely@lsfw01.lovelysystems.com) |
| 12:20.12 | *** join/#storm jdo (n=jdobrien@238.205.119.70.cfl.res.rr.com) |
| 12:56.02 | *** join/#storm thumper_laptop (n=tim@canonical/launchpad/thumper) |
| 12:59.55 | *** join/#storm HazzaUK (n=HazzaUK@dsl-217-155-201-84.zen.co.uk) |
| 13:00.29 | HazzaUK | HeyHey |
| 13:01.00 | *** join/#storm fcorrea (n=fcorrea@c953f461.virtua.com.br) |
| 13:04.46 | HazzaUK | I'm wondering if anyone could inform be about how the expressions evaluations work before I have a look at the code. I'm using the flushed hook to inform 'filter' interfaces of new/updated/deleted objects. I'd like to initialize the filters with a storm expression to be evaluated with the flushed item as input. |
| 13:07.15 | HazzaUK | At the moment I'm looping over expression primitives and evaluating them one by one. What would you suggest. I'm doing things like expr.expr1.__get__( obj ) == expr.expr2.get() |
| 13:07.54 | *** join/#storm jkakar (n=jkakar@92.54.156.162) |
| 13:08.00 | HazzaUK | Which really doesn't feel right. |
| 13:08.28 | *** join/#storm jkakar (n=jkakar@92.54.156.162) |
| 13:09.35 | HazzaUK | A quick glance at expr.py is quite daunting, so if anyone could give some pointers that would be great. |
| 13:11.23 | *** join/#storm oubiwann (n=oubiwann@97-119-4-38.omah.qwest.net) |
| 13:26.37 | niemeyer | HazzaUK: Object-basde expressions are compiled into their SQL format by an instance of the Compile class, which is somewhat of a generic functions pattern |
| 13:27.07 | niemeyer | HazzaUK: Functions are hooked into this instance to let it know of how to compile a given expression type into its SQL equivalent |
| 13:27.30 | niemeyer | HazzaUK: and these functions may call the compile instance again to compile subexpressions contained therein |
| 13:27.50 | niemeyer | HazzaUK: That's pretty much it |
| 13:27.58 | HazzaUK | ok, cheers |
| 13:28.07 | niemeyer | HazzaUK: There are details around this, but I'm sure you'll be able to figure them out with that basic info |
| 13:34.50 | *** join/#storm kov (n=kov@debian/developer/kov) |
| 13:50.01 | *** join/#storm munga (n=abate@ununtrium.pps.jussieu.fr) |
| 13:50.46 | munga | what is the best way to very if a record is in the db ? store.find(table, where).count() > 0 ? |
| 13:52.05 | therve | that's probably the less costly |
| 13:52.23 | therve | "best" is a bit wide to tell you :) |
| 13:53.06 | munga | actually the point is that I might try to insert a record twice (while I should not). I could just wait for an Integrity error ... but I find this quite inelegant . So I might check if the record is already in the db, but this way I'll add an additional query to the count... |
| 13:54.00 | munga | also because the integryty error is quite generic ... |
| 13:54.27 | munga | and moreover it's db specific ... |
| 13:55.44 | niemeyer | munga: Something like config(limit=1).values(TheClass.primary_key) |
| 13:55.56 | niemeyer | munga: count() may be expensive depending on DB |
| 13:56.15 | niemeyer | We'll implement an exists() in the ResultSet at some point, to make that easier |
| 14:01.44 | munga | niemeyer: thanks ! |
| 14:03.26 | mup | storm/variable-referenceset-leak r274 committed by thomas@canonical.com |
| 14:03.26 | mup | No need to re-hook object-deleted |
| 14:31.01 | mup | storm/variable-referenceset-leak r275 committed by thomas@canonical.com |
| 14:31.01 | mup | DIrectly call hook on flush, because we may have already missed |
| 14:31.01 | mup | start-tracking-changes the second time. |
| 14:37.05 | *** join/#storm Sidnei (n=Sidnei@houston.enfoldsystems.com) |
| 14:37.27 | *** join/#storm fcorrea (n=fcorrea@c953f461.virtua.com.br) |
| 14:57.06 | *** join/#storm fcorrea (n=fcorrea@c953f461.virtua.com.br) |
| 15:20.20 | *** join/#storm jjrojo (n=jjrojo@195.Red-79-145-145.staticIP.rima-tde.net) |
| 15:30.40 | *** join/#storm dobee (n=dobee@85-124-200-100.static.xdsl-line.inode.at) |
| 15:58.26 | *** join/#storm andrea-bs (n=andrea@ubuntu/member/beeseek.developer.andrea-bs) |
| 18:46.53 | *** join/#storm dobee (n=dobee@85-124-200-100.static.xdsl-line.inode.at) |
| 19:05.55 | *** join/#storm Zenom (n=Zenom@unaffiliated/aj1973) |
| 19:11.41 | *** join/#storm jjrojo (n=jjrojo@195.Red-79-145-145.staticIP.rima-tde.net) |
| 19:15.03 | *** join/#storm jfishman (i=jfishman@dhcp-199-59.wireless.bowdoin.edu) |
| 19:28.38 | *** part/#storm Zenom (n=Zenom@unaffiliated/aj1973) |
| 19:37.00 | *** join/#storm dobee (n=dobee@85-124-200-100.static.xdsl-line.inode.at) |
| 19:57.38 | *** join/#storm dobee (n=dobee@85-124-200-100.static.xdsl-line.inode.at) |
| 21:53.53 | *** join/#storm jfishman (i=jfishman@139.140.175.154) |
| 23:00.44 | *** join/#storm jukart (n=jukart@213162066137.public.t-mobile.at) |
| 23:40.24 | *** join/#storm jukart_ (n=jukart@213162066130.public.t-mobile.at) |
| 23:52.59 | *** join/#storm jukart__ (n=jukart@213162066131.public.t-mobile.at) |