| 00:15.55 | spats | hey all. I'm trying to do this: "mary.name = Select(Person.name, Person.id == 1)" to generate an insert or update, and storm is generating the sub-select without prens around it giving a syntax error from the database (sqlite or postgres.) |
| 00:16.15 | spats | I can work around this via monkey-patch "from storm import compile; compile.set_precedence(9, Select)" but I'm wondering if there isn't a better way that also isn't a global hack - and will doing this break anything horribly? |
| 00:28.42 | spats | http://dpaste.com/43852/ |
| 01:03.46 | *** join/#storm paranoidi (n=paranoid@av41.adsl.tnnet.fi) [NETSPLIT VICTIM] |
| 01:04.50 | *** join/#storm wgrant (n=wgrant@ubuntu/member/wgrant) [NETSPLIT VICTIM] |
| 01:04.50 | *** join/#storm munin (n=munin@209.9.227.50) [NETSPLIT VICTIM] |
| 01:06.06 | *** join/#storm oubiwann (n=oubiwann@97-119-81-240.omah.qwest.net) [NETSPLIT VICTIM] |
| 01:16.13 | *** join/#storm bac (n=bac@cpe-065-190-186-163.nc.res.rr.com) [NETSPLIT VICTIM] |
| 02:23.44 | wgrant | spats: I think I saw a bug about that - IIRC they worked around it by putting an extra Select() around it, but I could be wrong. |
| 02:25.43 | spats | wgrant: thanks - as a work-around I'm just executing the expression if it happens to be a select (the expression to be assigned comes from a function and could equally be 'current_timestamp' or some such. |
| 02:34.38 | jamesh | spats: please file a bug |
| 02:36.13 | spats | jamesh: it turns out that my precedence hack only works in some simple cases, so I have no fix to propose. none the less it would be nice to have Select work in a more general way, so I'll learn myself the bug-filing process ;) |
| 02:36.34 | jamesh | spats: bug reports don't need to come with fixes |
| 02:36.47 | jamesh | methods to reproduce are good though |
| 02:37.09 | jamesh | (and I don't mean http://xkcd.com/583/) |
| 02:37.11 | spats | okay, I'll file what I have. |
| 02:37.36 | spats | jamesh: hehe |
| 03:13.06 | *** join/#storm shaunm (n=shaunm@c-98-212-133-244.hsd1.il.comcast.net) |
| 04:02.57 | *** join/#storm bigdog (n=scmikes@72-197-8-8-arpa.cust.cinci.current.net) |
| 04:58.57 | *** join/#storm jukart (n=jukart@d91-128-122-124.cust.tele2.at) |
| 05:36.25 | *** join/#storm akm1 (n=akm1@122.166.39.10) |
| 05:57.53 | *** join/#storm jukart (i=lovely@81.189.156.94) |
| 06:22.53 | akm1 | Hi all, i am getting error while creating binary egg for storm. i am using mingw for compiling c extensions...but the compiler is expecting PC folder in my python directory...am i missing something...? |
| 06:25.37 | akm1 | am using windows and py2.5...any help much appreciated... |
| 06:35.24 | mup | storm/twisted-integration r222 committed by thomas@canonical.com |
| 06:35.24 | mup | Use TestHelper/MakePath to manage sqlite file. |
| 06:36.46 | therve | akm1: see https://bugs.launchpad.net/storm/+bug/268151 |
| 06:36.48 | mup | Bug #268151: Installation to Win32 is fiddly <Storm:Confirmed> <https://launchpad.net/bugs/268151> |
| 06:37.17 | therve | there are various solutions, one is to not build the extension at all, the other is to apply a patch on the C extension |
| 07:31.24 | akm1 | therve, i need to build the extension...any simple solution ? |
| 07:31.44 | therve | akm1: apply the patch which is in that bug page |
| 07:32.37 | akm1 | therve, is that patch not update to storm repository ? |
| 07:32.57 | therve | akm1: I don't understand that sentence |
| 07:33.09 | akm1 | therve, is that patch not updated to storm repository ? |
| 07:33.24 | therve | I still don't :) |
| 07:39.30 | jamesh | this two review policy makes it pretty difficult to get things merged in a timely manner :( |
| 07:40.40 | therve | jamesh: are you waiting for a review? |
| 07:41.12 | jamesh | therve: I do, but I was partly thinking about patches like the one on that bug |
| 07:41.44 | therve | jamesh: well it's not even in review... |
| 07:43.24 | jamesh | therve: right. But for trivially correct low impact patches like that it'd be nice to just merge them after one review. |
| 07:44.08 | jamesh | I think the two review policy is a great idea for API changes and risky patches though. |
| 07:46.45 | therve | jamesh: well in this case the problem is more that I can't test the patch |
| 07:47.04 | therve | and also that we can't ensure it will continue to work |
| 07:47.19 | jamesh | therve: you can test that the patch doesn't break the build on Linux and issues no warnings. |
| 07:47.41 | jamesh | i.e. we can easily check that the patch does no harm |
| 07:47.51 | therve | sure |
| 07:48.25 | jamesh | for this kind of portability issue, I think that is enough |
| 07:49.15 | jamesh | regularly running the tests on Windows would be nice too, but if we don't have the resources to do so then so be it. |
| 07:49.46 | jamesh | leaving the build broken after a problem has been reported though makes it difficult for others to do this testing for us |
| 07:53.06 | therve | jamesh: in this case the patch outputs a fair bunch of warnings |
| 07:53.23 | therve | and... it doesn't work |
| 07:53.28 | jamesh | therve: I guess we'd want a warningless version then ... |
| 08:00.15 | akm1 | therve, i applied the patch...but im getting this error.. build\temp.win32-2.5\Release\storm\cextensions.o(.text+0x353c):cextensions.c: undefined reference to `_imp__PyDict_Type'collect2: ld returned 1 exit status error: command 'gcc' failed with exit status 1 |
| 08:01.19 | therve | akm1: hum this doesn't seem related |
| 08:01.27 | therve | akm1: but anyway the patch doesn't seem to work, sorry |
| 08:01.45 | akm1 | therve, np |
| 08:34.50 | *** join/#storm goschtl (n=goschtl@p5B0BFB99.dip.t-dialin.net) |
| 11:05.10 | jamesh | akm1: Could you give the following branch a try? https://code.launchpad.net/~jamesh/storm/bug-268151 -- it includes a tweaked version of the patch from that bug. |
| 11:06.06 | akm1 | jamesh, sure..let me try |
| 11:21.45 | akm1 | jamesh, im getting this error...C:\MinGW\bin\gcc.exe -mno-cygwin -mdll -O -Wall -Ic:\Python25\include -Ic:\Python25\PC -c storm/cextensions.c -o build\temp.win32-2.5\Release\storm\cextensions.o |
| 11:21.46 | akm1 | storm/cextensions.c: In function `prepare_type': |
| 11:21.46 | akm1 | storm/cextensions.c:2161: error: `PyTPFLAGS_HAVE_GC' undeclared (first use in this function) |
| 11:21.46 | akm1 | storm/cextensions.c:2161: error: (Each undeclared identifier is reported only once |
| 11:21.46 | akm1 | storm/cextensions.c:2161: error: for each function it appears in.) |
| 11:21.47 | akm1 | error: command 'gcc' failed with exit status 1 |
| 11:28.30 | jamesh | akm1: Thanks. I'm just pushing r308 to fix that bug. |
| 11:28.47 | jamesh | If you do a "bzr pull", that bug should be gone |
| 11:29.05 | akm1 | jamesh, gr8 ! |
| 11:29.27 | akm1 | jamesh, will check... |
| 12:01.29 | akm1 | jamesh, am getting different error...build\temp.win32-2.5\Release\storm\cextensions.o(.text+0x429):cextensions.c: undefined reference to `_imp__PyExc_TypeError' |
| 12:01.30 | akm1 | build\temp.win32-2.5\Release\storm\cextensions.o(.text+0x517):cextensions.c: undefined reference to `_imp___Py_NoneStruct' |
| 12:01.30 | akm1 | build\temp.win32-2.5\Release\storm\cextensions.o(.text+0x51f):cextensions.c: undefined reference to `_imp___Py_NoneStruct' |
| 12:01.30 | akm1 | build\temp.win32-2.5\Release\storm\cextensions.o(.text+0x58e):cextensions.c: undefined reference to `_imp__PyExc_TypeError' |
| 12:01.32 | akm1 | build\temp.win32-2.5\Release\storm\cextensions.o(.text+0x61c):cextensions.c: undefined reference to `_imp___Py_NoneStruct' |
| 12:01.35 | akm1 | build\temp.win32-2.5\Release\storm\cextensions.o(.text+0x623):cextensions.c: undefined reference to `_imp___Py_NoneStruct' |
| 12:01.38 | akm1 | build\temp.win32-2.5\Release\storm\cextensions.o(.text+0x647):cextensions.c: undefined reference to `_imp___Py_NoneStruct' |
| 14:26.09 | *** join/#storm akm1 (n=akm1@122.166.39.10) |
| 14:43.05 | *** join/#storm shaunm (n=shaunm@proxyserver.wolfram.com) |
| 14:50.57 | *** join/#storm andrea-bs (n=andrea@ubuntu/member/beeseek.developer.andrea-bs) |
| 14:54.06 | jamesh | akm1: that's weird. Those should all be provided by Python. |
| 16:27.18 | *** join/#storm gord_ (n=gord@5acaca2b.bb.sky.com) |
| 16:38.57 | mzz | if I read that right that's a link-time error, it compiled successfully this time around. What's the linker commandline? |
| 16:41.26 | *** join/#storm sidnei (n=sidnei@189-30-60-24.cslce701.dsl.brasiltelecom.net.br) |
| 16:49.19 | *** join/#storm jukart (n=jukart@d91-128-122-124.cust.tele2.at) |
| 21:45.08 | *** join/#storm paranoidi (n=paranoid@av41.adsl.tnnet.fi) [NETSPLIT VICTIM] |
| 21:48.38 | *** join/#storm niemeyer (n=niemeyer@212.183.221.191.dyn.user.ono.com) |
| 22:03.00 | *** join/#storm thumper (n=tim@125-236-193-95.adsl.xtra.co.nz) |
| 22:22.56 | *** join/#storm vvinet (n=vince@bas2-sherbrooke40-1128555152.dsl.bell.ca) |