| 00:35.12 | m4v | ceronman: yes, looks that way |
| 00:58.08 | thumper | I have a quick question |
| 00:58.18 | thumper | given a field in a storm class |
| 00:58.18 | thumper | like a Reference |
| 00:58.32 | thumper | is there a public API to get the class the reference is from? |
| 00:58.34 | thumper | there is _cls |
| 00:58.43 | thumper | but I take it to be private |
| 00:58.55 | thumper | and I don't want it changing under me |
| 00:59.25 | lifeless | thumper: whatever you are doing is probably too clever ;) |
| 00:59.26 | thumper | I'm wanting to be clever(ish) and pass a single field to a method |
| 00:59.35 | thumper | which then builds the query |
| 00:59.40 | thumper | based on the table of the field |
| 01:00.06 | thumper | I could simply pass in the table and the field |
| 01:00.11 | thumper | but it seems like redundant information |
| 01:00.21 | thumper | as I should be able to get the table for a field |
| 01:07.20 | jkakar | ceronman: And and Or generate AND and OR statements. |
| 01:07.34 | jkakar | ceronman: There are no builtin operations for bitwise and and or, but they're easy to add. |
| 01:31.22 | ceronman | jkakar: I saw the following example: |
| 01:31.29 | ceronman | store.find(Foo, (Foo.name == u"bar") & (Foo.id == 5)) |
| 01:31.44 | jkakar | ceronman: Does that actually work? |
| 01:31.56 | ceronman | jkakar: I thought the bitwise operator was overloaded to match And |
| 01:32.09 | ceronman | jkakar: no idea, I just saw that in the manual: https://storm.canonical.com/Manual#AND |
| 01:32.29 | jkakar | ceronman: Hmm, interesting. |
| 01:32.37 | ceronman | jkakar: so, I was wondering if '|' will work as Or |
| 01:32.46 | jkakar | ceronman: I'd expect that you'd need to implement a BitwiseAnd (inheriting from CompoundOper) to make it work. |
| 01:33.26 | jkakar | ceronman: If you look in storm.expr and search for CompoundOper you can see examples... basically, I think, you'd need to implement a BitwiseAnd expression with 'oper = "&"' and a similar thing for BitwiseOr. |
| 01:34.21 | ceronman | jkakar: well, I'm not interested in implementing an operator, I was just wondering if it's already implemented |
| 01:34.28 | jkakar | ceronman: Anyway, if you want to build AND or OR expressions, I recommend using the And and Or expressions... they're more obvious than relying on operator overloading. |
| 01:34.37 | ceronman | if '&' is implemented for And, I thought that '|' was implemented for Or |
| 01:34.51 | jkakar | ceronman: Just like .union and .intersection are more obvious with sets than the relevant operators. |
| 01:35.54 | ceronman | jkakar: Yes, I don't want to overload anything. I just saw that in the manual and thought that was the prefered way to use AND operations |
| 01:36.15 | jkakar | ceronman: Ah, okay. The preferred way is to use And and Or directly. |
| 01:37.42 | ceronman | jkakar: Ah, ok, I see. Is just that the operator is the first example shown in the documentation. |
| 01:37.54 | jkakar | ceronman: Ah, interesting... I should delete that. ;) |
| 01:38.09 | ceronman | hehehe |
| 02:24.15 | *** part/#storm ceronman (~ceronman@cmodem190-5-198-109.emtel.net.co) |
| 02:42.00 | *** join/#storm fcorrea (~fcorrea@201-62-110-81-ma.static.vivax.com.br) |
| 05:54.49 | *** join/#storm sidnei (~sidnei@201-11-226-193.cslce701.dsl.brasiltelecom.net.br) |
| 06:14.10 | *** join/#storm medberry (~medberry@c-75-71-81-190.hsd1.co.comcast.net) |
| 06:34.50 | *** join/#storm medberry (~medberry@c-75-71-81-190.hsd1.co.comcast.net) |
| 08:00.45 | *** join/#storm medberry (~medberry@c-75-71-81-190.hsd1.co.comcast.net) |
| 08:07.51 | *** join/#storm medberry (~medberry@c-75-71-81-190.hsd1.co.comcast.net) |
| 08:57.48 | *** join/#storm zmijunkie (~sascha@89.244.144.168) |
| 09:12.12 | *** join/#storm bigjools (~quassel@82-71-93-254.dsl.in-addr.zen.co.uk) |
| 09:12.13 | *** join/#storm bigjools (~quassel@canonical/launchpad/bigjools) |
| 09:35.17 | *** join/#storm stub (~stub@canonical/launchpad/stub) |
| 09:45.19 | *** join/#storm fcorrea (~fcorrea@201-62-110-81-ma.static.vivax.com.br) |
| 10:41.31 | *** join/#storm esteve (~quassel@91.126.226.47) |
| 10:46.55 | *** join/#storm CrummyGummy (~meh@li185-28.members.linode.com) |
| 10:52.36 | *** join/#storm artista-frustrad (~artista_f@187.7.151.137) |
| 10:57.23 | *** join/#storm niemeyer (~niemeyer@201-66-179-18.pltce701.dsl.brasiltelecom.net.br) |
| 11:04.42 | *** join/#storm zmijunkie (~sascha@89.244.144.168) |
| 11:13.54 | *** join/#storm artista-frustrad (~artista_f@187.7.151.137) |
| 11:34.15 | *** join/#storm artista-frustrad (~artista_f@187.7.151.137) |
| 12:01.35 | *** join/#storm fcorrea (~fcorrea@201-62-110-81-ma.static.vivax.com.br) |
| 12:08.18 | *** join/#storm zmijunkie (~sascha@89.244.144.168) |
| 12:20.56 | *** join/#storm sidnei (~sidnei@186.195.17.2) |
| 13:07.07 | *** join/#storm fcorrea (~fcorrea@187.3.22.136) |
| 13:27.46 | *** join/#storm vvinet (~vince@modemcable121.135-59-74.mc.videotron.ca) |
| 14:36.06 | *** join/#storm sidnei (~sidnei@201-11-226-193.cslce701.dsl.brasiltelecom.net.br) |
| 15:00.49 | *** join/#storm magmatt (~Matt@74-92-245-181-Utah.hfc.comcastbusiness.net) |
| 15:00.52 | *** part/#storm magmatt (~Matt@74-92-245-181-Utah.hfc.comcastbusiness.net) |
| 16:39.11 | *** join/#storm _thumper_ (~thumper@125-236-193-95.adsl.xtra.co.nz) |
| 16:39.11 | *** join/#storm _thumper_ (~thumper@canonical/launchpad/thumper) |
| 18:54.01 | *** join/#storm magmatt (~Matt@74-92-245-181-Utah.hfc.comcastbusiness.net) |
| 18:54.20 | magmatt | I have an old table with no primary key; how should I model that in storm? |
| 18:55.05 | jkakar | magmatt: Heh, this question came up yesterday... |
| 18:55.30 | magmatt | shall I look at the logs :) ? |
| 18:55.34 | jkakar | magmatt: Basically, Storm needs a primary key to be present. It uses the primary key to key live objects in the cache. |
| 18:56.01 | jkakar | magmatt: Heh, nah... I can't remember who I was talking to, but they were going to try to use PostgreSQL OIDs as a primary key. That might work if you're using PostgreSQL. |
| 18:56.14 | jkakar | magmatt: Probably the best option is to add a primary key to the table. |
| 18:56.21 | magmatt | I use postgres for production and sqlite for testing |
| 18:56.30 | jkakar | magmatt: Ah, right. |
| 18:56.32 | magmatt | I could make an oid int in sqlite |
| 18:56.41 | jkakar | magmatt: Is adding a primary key not an option? |
| 18:56.49 | magmatt | nope |
| 18:57.03 | magmatt | not without the operations bureaucracy getting in the way |
| 18:57.27 | magmatt | I mean, it might take weeks or months |
| 18:57.36 | magmatt | is using postgres's oid recommended? |
| 18:57.41 | magmatt | any side effects? |
| 18:58.07 | jkakar | magmatt: Ouch. |
| 18:58.19 | jkakar | magmatt: I'm not really sure. I've never tried to use PostgreSQL's OID, but apparently its unique per row. |
| 18:58.31 | *** join/#storm vvinet (~vince@2001:470:8829:1000:227:eff:fe25:ee64) |
| 18:58.44 | magmatt | it is unique. . . I just don't know . I'll test it |
| 19:00.00 | jkakar | magmatt: Yeah, good idea. :) |
| 19:02.05 | magmatt | jkakar: sqlite has an oid, too |
| 19:02.14 | jkakar | magmatt: Ah, interesting. |
| 19:02.22 | magmatt | so I'll test with _oid = Int('oid', primary=True) |
| 19:02.25 | magmatt | as the property |
| 19:12.30 | magmatt | jkakar: looks like it will work (after minimal testing) I mostly need it for read-only access, and that seems to work |
| 19:12.46 | jkakar | magmatt: Nice! |
| 19:12.58 | magmatt | that'll be really helpful for a lot of our legacy tables |
| 19:13.02 | jkakar | magmatt: I'm going to be spending some time working on proper documentation for Storm. |
| 19:13.06 | magmatt | thanks |
| 19:13.12 | magmatt | oh! |
| 19:13.15 | magmatt | go docs! |
| 19:13.24 | jkakar | magmatt: If you have any suggestions about what to document or things that you stumbled over, please let me know. |
| 19:13.31 | magmatt | I will for sure |
| 19:13.36 | magmatt | we're starting to like storm here at work |
| 19:13.49 | magmatt | I'll ask around about deficiencies in the docs |
| 19:13.54 | jkakar | magmatt: Yeah, our documentation story has been bad for too long and nobody has fixed it. |
| 19:13.59 | jkakar | magmatt: Awesome, thanks! |
| 19:14.34 | jkakar | magmatt: I've just changed jobs and am introducing Storm... my new colleagues are accepting it, because they trust me, but it's proving to be hard for them to figure out how to use it. |
| 19:14.50 | magmatt | I've had the same experience |
| 19:14.56 | magmatt | I'm sending a department email right now |
| 19:16.58 | jkakar | magmatt: Sweet, thanks. :) |
| 22:03.59 | *** join/#storm fcorrea (~fcorrea@201-62-110-81-ma.static.vivax.com.br) |
| 22:41.02 | magmatt | jkakar: from my coworkers: |
| 22:41.03 | magmatt | The manual (https://storm.canonical.com/Manual) still says "under construction" and has a lot of blank pages. Filling in the blanks would be a good place to start. |
| 22:41.31 | magmatt | *one of my coworkers |
| 22:41.39 | magmatt | thanks |
| 22:41.42 | *** part/#storm magmatt (~Matt@74-92-245-181-Utah.hfc.comcastbusiness.net) |