00:25.16 | *** join/#storm artista_frustrad (~artista_f@189.30.139.238) |
05:34.43 | *** join/#storm stub (~stub@ppp-58-8-12-220.revip2.asianet.co.th) |
09:40.52 | *** join/#storm Suv123 (~chatzilla@110.44.126.19) |
09:43.15 | Suv123 | hi guys i got error like this http://fpaste.org/raRf/ |
09:43.43 | Suv123 | can anyone help me to understand whats going under the hood? what may be possible reason for error? |
09:52.05 | marienz | Suv123: not without failing code |
10:08.29 | *** join/#storm zmijunkie_ (~sascha@i59F5F2BC.versanet.de) |
10:24.43 | *** join/#storm sidnei (~sidnei@189.30.217.33) |
10:38.11 | *** join/#storm elmom_ (~elmom@a88-114-241-230.elisa-laajakaista.fi) |
10:38.47 | *** join/#storm danilo_ (~danilo@static-213-198-236-193.adsl.eunet.rs) |
10:58.37 | *** join/#storm zmijunkie (~sascha@i59F5ED08.versanet.de) |
11:04.42 | *** join/#storm zmijunkie_ (~sascha@i59F5CD97.versanet.de) |
11:28.58 | *** join/#storm niemeyer (~niemeyer@187.53.254.172) |
11:44.11 | *** join/#storm jdobrien (~john@24.18.88.184.cfl.res.rr.com) |
11:49.09 | *** join/#storm sidnei (~sidnei@189.30.217.33) |
12:20.14 | *** part/#storm stub (~stub@ppp-58-8-12-220.revip2.asianet.co.th) |
12:22.04 | *** join/#storm stub (~stub@canonical/launchpad/stub) |
12:54.56 | *** join/#storm fcorrea (~fcorrea@187.3.31.9) |
13:12.04 | therve | stub, around? |
13:12.17 | stub | Yes |
13:12.28 | therve | stub, you need to pass unicode for things to work |
13:12.36 | therve | this is indeed something that changed with psycopg 2.2 |
13:13.17 | stub | Foo.col.is_in(['foo']) works but In(Foo.col, ['foo']) does not |
13:14.11 | stub | The former the string gets correctly cast to unicode, the second not so much. It looks like the casting can be moved from is_in to In, as the former invokes the latter |
13:14.41 | therve | true |
13:14.50 | therve | I'd rather have both of them fail, though |
13:15.41 | stub | I have an unknown number of call sites to change and a huge number of doctests |
13:16.05 | stub | Why do you want them both to fail? |
13:16.23 | stub | u'foo' == 'foo' in Python just fine |
13:17.12 | therve | well, if you have ascii sure |
13:17.33 | therve | also, not in python 2 |
13:17.35 | therve | also, not in python 3 |
13:18.48 | stub | The string constants are automatically unicode in Python 3 so that is irrelevant I think |
13:19.42 | therve | if you're only talking about string in tests, sure |
13:19.56 | therve | but I suspect they actually come from a user input |
13:20.08 | therve | so enforcing it to be unicode is a good thing |
13:20.17 | stub | If they come from user input, they will be strings (unicode). |
13:20.44 | therve | well, you don't know if the API silently accepts encoded string |
13:21.09 | *** join/#storm gary_poster (~anonymous@99-98-191-175.lightspeed.rlghnc.sbcglobal.net) |
13:21.51 | stub | oh... str is still returned from open().read() and friends |
13:22.10 | therve | I'm not really sure why it works for is_in and not In |
13:22.20 | therve | the compilation paths are fairly different |
13:22.40 | stub | is_in does casting then returns In() |
13:23.00 | therve | ah right |
13:23.08 | therve | the variable doesn't matter when In is used directly |
13:23.25 | stub | Right, which seems incorrect. |
13:25.34 | therve | I *guess* |
13:26.32 | stub | You would also be able to have both spellings fail if you want, as you make the casting fail to promote str to unicode ;) |
13:26.57 | stub | I'm trying to determine if we have an impractical number of call sites or not |
13:28.42 | stub | And the fix is pretty much always casting the str to unicode (we can't fix the source because it is impractical to find out where the source is), so Storm's purity doesn't really help me ;) |
13:29.26 | stub | Mostly the source are tests, as almost all the user input comes in as unicode |
14:39.17 | *** join/#storm sidnei (~sidnei@201-35-74-151.cslce701.dsl.brasiltelecom.net.br) |
16:17.02 | *** join/#storm keppla (~keppla@i577B0DBB.versanet.de) |
16:28.43 | *** join/#storm jwstasiak (~jwstasiak@cpe-74-64-104-182.nyc.res.rr.com) |
16:33.01 | jwstasiak | Hi guys. What's the best way to load a sql file (defining tables, etc) into a sqlite db? I'm currently reading the file (block = f.read()) and trying store.execute(block), but fails saying it 'can only execute one statement at a time', the standard sqlite api in python has an executeScript - does storm have something similar? |
16:52.19 | *** join/#storm niemeyer_ (~niemeyer@189-72-21-191.pltce701.dsl.brasiltelecom.net.br) |
16:58.58 | *** join/#storm sidnei (~sidnei@201-35-74-151.cslce701.dsl.brasiltelecom.net.br) |
17:01.30 | *** join/#storm sidnei (~sidnei@201-35-74-151.cslce701.dsl.brasiltelecom.net.br) |
18:54.48 | *** join/#storm elmom_ (~elmom@hoasnet-fe29dd00-137.dhcp.inet.fi) |
19:53.13 | *** join/#storm keppla1 (~keppla@i577B23F2.versanet.de) |
21:43.52 | *** join/#storm jcsackett (~jcsackett@cpe-098-122-187-054.nc.res.rr.com) |
21:44.30 | jcsackett | hi all, how does one get at the sql storm is sending to the db? i need better debugging of a query that's all goofed up. |
21:44.54 | therve | jcsackett, storm.tracer.debug(True) iirc |
21:45.15 | jcsackett | therve: does that just force everything to dump to stdout? |
21:45.27 | therve | jcsackett, yeah |
21:55.35 | jcsackett | therve: that works. thanks! |
21:58.04 | jcsackett | a new question. any idea why storm would be converting a condition like (Product.somefield == 20) to FALSE ? |
22:03.23 | *** join/#storm shaunm (~shaunm@74.83.18.167) |
22:17.33 | jcsackett | has anyone seen a storm query replace a condition with FALSE? |
22:51.12 | jcsackett | attemping to paint the picture more clearly: https://pastebin.canonical.com/39408/ |
22:51.20 | jcsackett | that's storm query and output sql. |
22:52.19 | jcsackett | and it was copied imperfectly; there are two ')' before .config |
23:25.53 | *** join/#storm fcorrea (~fcorrea@189-54-100-52-nd.cpe.vivax.com.br) |
23:35.24 | *** join/#storm jcsackett (~jcsackett@cpe-098-122-187-054.nc.res.rr.com) |
23:55.56 | *** part/#storm bac (~bac@canonical/launchpad/bac) |