| *** gokrokve has quit IRC | 00:17 | |
| *** sjmc7_ has joined #murano | 00:22 | |
| *** sjmc7 has quit IRC | 00:26 | |
| *** asalkeld has quit IRC | 00:44 | |
| *** asalkeld has joined #murano | 01:02 | |
| *** asalkeld has quit IRC | 01:07 | |
| *** ankurrr has quit IRC | 01:28 | |
| *** killer_prince is now known as lazy_prince | 02:42 | |
| *** lazy_prince is now known as killer_prince | 02:42 | |
| *** killer_prince is now known as lazy_prince | 02:43 | |
| *** lazy_prince is now known as killer_prince | 02:45 | |
| *** sjmc7_ has quit IRC | 03:55 | |
| *** killer_prince is now known as lazy_prince | 04:24 | |
| *** lazy_prince has quit IRC | 04:41 | |
| *** gokrokve has joined #murano | 04:44 | |
| *** lazy_prince has joined #murano | 05:09 | |
| *** gokrokve_ has joined #murano | 05:09 | |
| *** gokrokve has quit IRC | 05:11 | |
| *** lazy_prince is now known as killer_prince | 05:22 | |
| *** gokrokve_ has quit IRC | 06:08 | |
| *** killer_prince is now known as lazy_prince | 06:15 | |
| *** openstackgerrit has quit IRC | 07:10 | |
| *** gokrokve has joined #murano | 07:19 | |
| *** tnurlygayanov has quit IRC | 07:22 | |
| *** tnurlygayanov has joined #murano | 07:23 | |
| *** gokrokve has quit IRC | 07:23 | |
| *** sergmelikyan has quit IRC | 08:28 | |
| *** bogdando has quit IRC | 08:29 | |
| *** sergmelikyan has joined #murano | 08:31 | |
| *** openstackgerrit has joined #murano | 09:47 | |
| openstackgerrit | Ekaterina Chernova proposed a change to stackforge/python-muranoclient: Update package-import command https://review.openstack.org/102162 | 09:51 |
|---|---|---|
| openstackgerrit | Sergey Murashov proposed a change to stackforge/murano: Define asserts for python 2.6 https://review.openstack.org/101894 | 10:32 |
| openstackgerrit | Stan Lagun proposed a change to stackforge/murano: MuranoPL testing mini-framework https://review.openstack.org/101370 | 11:42 |
| *** stanlagun has joined #murano | 11:43 | |
| openstackgerrit | A change was merged to stackforge/murano-deployment: added new parameter for engine tests https://review.openstack.org/101506 | 12:39 |
| *** sergmelikyan has quit IRC | 12:40 | |
| *** sergmelikyan has joined #murano | 12:40 | |
| sergmelikyan | . | 12:41 |
| *** drupalmonkey has joined #murano | 12:42 | |
| openstackgerrit | A change was merged to stackforge/murano: fix db-sync execution in Murano setup script https://review.openstack.org/101916 | 12:56 |
| openstackgerrit | Sergey Murashov proposed a change to stackforge/murano: Add port checks after deployment success https://review.openstack.org/101190 | 13:13 |
| *** lazy_prince has quit IRC | 13:13 | |
| openstackgerrit | Sergey Murashov proposed a change to stackforge/murano: Define asserts for python 2.6 https://review.openstack.org/101894 | 13:13 |
| *** gokrokve has joined #murano | 14:01 | |
| openstackgerrit | Ekaterina Chernova proposed a change to stackforge/murano-dashboard: Introduces script for updating horizon config https://review.openstack.org/101578 | 14:11 |
| *** sjmc7 has joined #murano | 14:26 | |
| *** killer_prince has joined #murano | 14:33 | |
| *** killer_prince has quit IRC | 14:40 | |
| *** killer_prince has joined #murano | 14:45 | |
| openstackgerrit | Sergey Murashov proposed a change to stackforge/murano: Add port checks after deployment success https://review.openstack.org/101190 | 14:47 |
| *** gokrokve has quit IRC | 15:14 | |
| ruhe | folks, if anyone has something to discuss today on the meeting, please update https://wiki.openstack.org/wiki/Meetings/MuranoAgenda#Agenda | 15:30 |
| openstackgerrit | Ekaterina Chernova proposed a change to stackforge/murano-dashboard: Introduces script for updating horizon config https://review.openstack.org/101578 | 15:38 |
| *** gokrokve has joined #murano | 15:39 | |
| sjmc7 | i have done so, ruhe, and a blueprint | 15:39 |
| *** _ruhe_ has joined #murano | 16:02 | |
| *** gokrokve_ has joined #murano | 16:22 | |
| *** gokrokve has quit IRC | 16:25 | |
| openstackgerrit | Ekaterina Chernova proposed a change to stackforge/murano-dashboard: Introduces script for updating horizon config https://review.openstack.org/101578 | 16:29 |
| *** gokrokve_ has quit IRC | 16:32 | |
| *** _ruhe_ has quit IRC | 16:47 | |
| *** ankurrr has joined #murano | 16:57 | |
| *** akuznetsova_ has joined #murano | 17:02 | |
| *** akuznetsova_ has left #murano | 17:02 | |
| *** gokrokve has joined #murano | 17:10 | |
| *** iyozhikov is now known as IgorYozhikov | 17:16 | |
| *** IgorYozhikov is now known as iyozhikov | 17:16 | |
| openstackgerrit | Stan Lagun proposed a change to stackforge/murano: MuranoPL testing mini-framework https://review.openstack.org/101370 | 17:17 |
| *** openstackgerrit has quit IRC | 17:35 | |
| *** openstackgerrit has joined #murano | 17:36 | |
| sjmc7 | ruhe - currently if a call to Heat takes too long, eventually murano times out with an auth failure. is that what keystone trusts solve? | 18:02 |
| stanlagun | sjmc7m not sure trust is the right solution for that | 18:06 |
| sjmc7 | there's no way to refresh tokens i don't think. not without a lot more keystone privileges than is sensible | 18:09 |
| ruhe | there is a problem with trusts, they can't be chained | 18:09 |
| sjmc7 | i will rephrase my question - if heat takes a long time, the engine times out eventually | 18:10 |
| sjmc7 | does anyone have thoughts about handling it? | 18:10 |
| stanlagun | need to make a research first. Maybe something can be contributed to keystone | 18:10 |
| ruhe | i think, chained trusts would solve this problem | 18:11 |
| sjmc7 | this will be quite a big deal for large deployments | 18:11 |
| stanlagun | Is it possible to have different trusts for the same user simultaneously? Suppose there are 2 parallel deployments from different users with different privileges | 18:13 |
| sjmc7 | i'll file a BP | 18:15 |
| ruhe | solum folks had the same problem. sjmc7 you and me discussed oath approach with Angus on the summit. but from this thread https://www.mail-archive.com/openstack-dev@lists.openstack.org/msg25283.html it seems that oauth also wouldn't solve the problem | 18:21 |
| sjmc7 | yep, i remember that | 18:33 |
| ruhe | stanlagun: can you please take a look at https://review.openstack.org/#/c/100344/ ; it's been waiting for review for a long time | 18:33 |
| openstackgerrit | Ryan Peters proposed a change to stackforge/murano-dashboard: Adds test coverage for flavor filter https://review.openstack.org/102321 | 18:53 |
| sjmc7 | ruhe - stan didn't like this, i think | 18:56 |
| sjmc7 | i have been meeaning to try with multiple inheritance, but i fear it | 18:56 |
| ruhe | then we need to ask stan to help with the approach he's proposing. stanlagun will you help with that? | 19:06 |
| sjmc7 | well, he suggested making the heat instance inherit from Instance, and then I think a class to inherit from both Linux and HeatSWConfig to bring them back together | 19:07 |
| sjmc7 | my fear is that once two classes combined like this override the same method, you get weird results, and in general i avoid multiple inheritance for that reason | 19:08 |
| sjmc7 | i don't think mixins are really the answer though since these examples aren't adding functionality, but overriding it | 19:09 |
| sjmc7 | some input from stan would be good though if he's had a chance to think about it | 19:09 |
| sjmc7 | worst case i can bundle the class with the application package for now so it's not urgent | 19:09 |
| *** killer_prince has quit IRC | 19:24 | |
| stanlagun | I'm back | 19:29 |
| stanlagun | sjmc7, are there any problems with mixins/multiply inheritance that you've faced? | 19:35 |
| sjmc7 | not in this specific instance, no | 19:35 |
| sjmc7 | in my career, yes :) | 19:35 |
| stanlagun | It seems like all this Instanece* hierarchy need to be refactored. Maybe it not a good idea to have all possible Instance combination but rather refactor it into a well-designed mixins | 19:36 |
| sjmc7 | i think it would work in this case | 19:36 |
| sjmc7 | yeah, that's my impression too | 19:36 |
| openstackgerrit | Ryan Peters proposed a change to stackforge/murano-dashboard: Adds test coverage for flavor filter https://review.openstack.org/102321 | 19:37 |
| sjmc7 | the only problem is that the mixin functions need to override functions in Instance in some cases | 19:37 |
| stanlagun | sure | 19:37 |
| stanlagun | but maybe it is not a problem | 19:37 |
| stanlagun | I've made some improvements to how methods resolved with multiply inheritance. That may solve the problems. Or may not. If we don't try to make it right we will never know. Maybe it would be better to try to refactor it into mixins and then I something doesn't work improve MuranoPL. And if we fail we gat back to single inheritance and will know exactly that it is unavoidable | 19:40 |
| sjmc7 | ok. can a contract enforce multiple requirements? | 19:40 |
| sjmc7 | e.g. could an Application rely on an instance that was both a LinuxInstance and HeatSWConfigInstance? | 19:41 |
| stanlagun | honestly I've never tried. But probably yes. Or at least this can be done with small improvements to contracts code | 19:43 |
| stanlagun | But you will need to chose one type as a property interface and cast to a second type in workflow | 19:45 |
| stanlagun | $.class(HeatSWConfigInstance).check(isinstanceof($, LinuxInstance)) | 19:46 |
| stanlagun | just need to implement isinstnaceof() | 19:47 |
| sjmc7 | ok. well let's leave it for now | 19:47 |
| sjmc7 | i'll have a play around with it | 19:47 |
| stanlagun | Maybe we can implement something like $.class(class1, class2) | 19:53 |
| stanlagun | Let me also think about it | 19:53 |
| stanlagun | Can you create etherpad so we can work on this together? | 19:54 |
| sjmc7 | yeah | 19:54 |
| stanlagun | sjmc7, just checked Contract: $.class(ParentClass1).class(ParentClass2) works fine and you can invoke methods that are present only in one of those 2 classes | 20:22 |
| sjmc7 | splendid! ok | 20:22 |
| stanlagun | The easiest way to test some MuranoPL feature is to write UT using https://review.openstack.org/#/c/101370/ | 20:25 |
| stanlagun | This case added me one more test suite | 20:26 |
| stanlagun | *test case | 20:26 |
| sjmc7 | yup, agreed. that's why i've been reviewing it :) | 20:27 |
| stanlagun | :) thanks! | 20:29 |
| *** asalkeld has joined #murano | 20:51 | |
| ankurrr | had a question about unit tests for the murano api: If I want to add unit tests specifically for packages, should they go into murano/tests/api/v1/test_environments.py ? Or should they go into a new file at murano/tests/api/v1/test_packages.py ? | 20:55 |
| stanlagun | new file | 20:57 |
| ankurrr | ok | 20:58 |
| *** sergmelikyan has quit IRC | 21:07 | |
| *** asalkeld has quit IRC | 21:22 | |
| sjmc7 | stanlagun, you still there? | 21:49 |
| stanlagun | yep | 21:49 |
| sjmc7 | this is a really dumb question - if i want to cast an integer to a string, how do i do it? | 21:49 |
| stanlagun | str() | 21:49 |
| sjmc7 | in muranopl | 21:50 |
| sjmc7 | not python | 21:50 |
| stanlagun | yes | 21:50 |
| sjmc7 | ok. str($.value) ? | 21:50 |
| stanlagun | str($.property) | 21:50 |
| sjmc7 | thanks. i'd tried string() but i guess that's just an assertion | 21:51 |
| stanlagun | you can also do this via string concatenation or '{0}'.format($value) | 21:51 |
| sjmc7 | ah, ok | 21:51 |
| sjmc7 | where's that defined? | 21:52 |
| stanlagun | see yaql_funcions.py for a list of functions (there are 2 such files) | 21:52 |
| sjmc7 | so i don't need to ask stupid questions :) | 21:52 |
| stanlagun | + there are built-ins in yaql itself | 21:52 |
| sjmc7 | ah, ok | 21:53 |
| stanlagun | this is not a stupid question :) and this also need to be improved. I don't like having so many functions in global namespace | 21:53 |
| sjmc7 | yeah, it's a little confusing right now. maybe we need a sample that uses every built-in | 21:55 |
| stanlagun | UTs | 21:56 |
| stanlagun | I also think about introducing something like modules to group them | 21:57 |
| sjmc7 | as long as it's documented i don't think it matters too much | 21:57 |
| stanlagun | it will metter when we will have plugins with ability for plugin to introduce another set of functions | 21:59 |
| stanlagun | But this is not in J :) | 21:59 |
| sjmc7 | yeah. for now i just want to make it easier to write packages | 22:01 |
| *** drupalmonkey has quit IRC | 22:01 | |
| *** drupalmonkey has joined #murano | 22:02 | |
| *** btully has quit IRC | 22:06 | |
| stanlagun | thats cool! i hope you like MuranoPL :) | 22:07 |
| sjmc7 | i think it's very powerful; i do think though that it needs to be much easier to debug problems, and that's a problem all languages face | 22:09 |
| stanlagun | agree. Debuggability was greatly improved in J1 and it will be farther improved in J2. But the ultimate solution will be to have our own IDE (for example Eclipse-based or JetBrains MPS-based), interactive debugger and code analysis tools. btw Python have many of those problems either. Especially it had it in its youth | 22:14 |
| stanlagun | at least we have readable stack-traces with MuranoPL and Python frames mixed | 22:15 |
| sjmc7 | https://etherpad.openstack.org/p/murano-multiple-inheritance | 22:16 |
| stanlagun | is it linked to any BP? | 22:30 |
| sjmc7 | https://blueprints.launchpad.net/murano/+spec/decide-multiple-inheritance | 22:30 |
| openstackgerrit | Steve McLellan proposed a change to stackforge/python-muranoclient: Allow deletion of multiple environments https://review.openstack.org/102374 | 22:44 |
| stanlagun | Failed to delete '{}'. Isn't it supposed to be {0}? | 22:46 |
| stanlagun | answered to myself: {} does work, didn't new such syntax allowed | 22:49 |
| *** sjmc7_ has joined #murano | 22:51 | |
| sjmc7_ | if you don't put numbers it does them sequentially | 22:53 |
| sjmc7_ | it think mixing {0} and {} is frowned upon | 22:53 |
| *** sjmc7 has quit IRC | 22:55 | |
| *** drupalmonkey has quit IRC | 23:27 | |
| *** gokrokve has quit IRC | 23:47 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!