Thursday, 2016-08-11

*** uck has quit IRC00:01
*** trozet has quit IRC00:54
*** trozet has joined #openstack-heat-translator01:07
*** bobh has joined #openstack-heat-translator01:07
*** spzala has quit IRC01:21
*** spzala has joined #openstack-heat-translator02:22
*** spzala has quit IRC02:27
*** spzala has joined #openstack-heat-translator02:53
*** spzala has quit IRC02:59
*** spzala has joined #openstack-heat-translator03:01
*** spzala has quit IRC03:05
*** trozet has quit IRC03:32
*** sridhar_ram has quit IRC03:47
*** sridhar_ram has joined #openstack-heat-translator03:50
*** bobh has quit IRC03:51
*** skraynev has quit IRC03:56
*** spzala has joined #openstack-heat-translator04:02
*** spzala has quit IRC04:06
*** vishwanathj has quit IRC04:20
*** vishwanathj has joined #openstack-heat-translator04:21
*** spzala has joined #openstack-heat-translator05:01
*** spzala has quit IRC05:05
*** spzala has joined #openstack-heat-translator06:02
*** spzala has quit IRC06:06
openstackgerritSanthosh64 proposed openstack/heat-translator: Implemented Scaling policies in heat translator
openstackgerritSanthosh64 proposed openstack/heat-translator: Implemented Scaling policies in heat translator
*** vishwanathj has quit IRC06:15
*** vishwanathj has joined #openstack-heat-translator06:15
*** spzala has joined #openstack-heat-translator07:02
*** spzala has quit IRC07:07
openstackgerritSteve Martinelli proposed openstack/heat-translator: fix the tosca-parser job
*** spzala has joined #openstack-heat-translator08:03
*** spzala has quit IRC08:08
*** spzala has joined #openstack-heat-translator11:05
*** spzala has quit IRC11:10
*** spzala has joined #openstack-heat-translator12:06
*** spzala has quit IRC12:11
*** bobh has joined #openstack-heat-translator12:57
*** spzala has joined #openstack-heat-translator13:07
*** spzala has quit IRC13:12
*** bobh has quit IRC13:15
*** spzala has joined #openstack-heat-translator13:18
openstackgerritMerged openstack/heat-translator: fix the tosca-parser job
*** bobh has joined #openstack-heat-translator13:54
*** trozet has joined #openstack-heat-translator14:13
openstackgerritMerged openstack/heat-translator: Implemented Scaling policies in heat translator
*** uck has joined #openstack-heat-translator14:27
*** uck has quit IRC15:21
*** zaneb has quit IRC15:30
*** shangxdy_ has joined #openstack-heat-translator15:56
spzalaHi all topol: bobh: shangxdy_: tbh:16:00
shangxdy_HI,spzala and all:)16:00
shangxdy_I'am looking at the comments.16:01
spzalashangxdy_: cool and meanwhile we can wait for a minute or two if anyone else joining16:02
spzalaaite, we can start our short meeting16:04
spzalaannouncement: tosca-parser 0.6.0 is released16:04
bobho/  sorry late16:05
spzalaand seems like it's backward compatible this time and no issues with heat-translator or tacker16:05
spzalabobh: o/16:05
spzalabobh: np, glad that you made it :)16:05
shangxdy_great, i have updated the release locally.16:06
spzalaI just announced that tosca-parser 0.6.0 seems backward compatible and didn't create any translator or tacker issues,16:06
bobhspzala: cool16:06
spzalafirst time we didn't have to change requirements in translator16:06
spzalabobh: thanks again for your manual verification of new release with tacker16:07
bobhspzala: np16:07
spzalashangxdy: :-) Cool16:07
spzalathe second things - gate-heat-translator-tox-py27-tosca-parser-master-ubuntu-xenial16:07
spzalais been failing lately after the gate names were changed with xenial suffix16:07
spzalasince it's non-voting things were fine, but we wanted to make sure to fix it soon as possible and seems like the issue was with char length16:08
spzalaof the gate name .. possible fix was added by stevemar and it's merged .. hopefully this fixes the issue16:09
spzalathat's pretty much from my side16:09
spzalawe can move to substitution_mappings discussion
spzalashangxdy: we can have quick discussion here, since we don't have full test coverage to it and since the complete implementation is not yet done, the main thing here is to get agreed with our approach16:11
shangxdy_The gate name is Indeed too long:)16:11
spzalashangxdy: :) yup, it was long and got too long after the new suffix :-)16:11
shangxdy_Let's back to the substitution_mappings.16:12
spzalashangxdy: the patch looks great, and provides initial implementation nicely but as I said we just need to make sure our approach16:12
spzalato me the latest patch approach looks good, if not perfect, but we can discuss and change as needed16:13
shangxdy_I agree16:13
spzalashangxdy: and once we have initial patch merged (this patch), your new patches with test coverage will make it more clear16:14
spzalashangxdy: so let's not spend too much time on this one, if we like approach and you think it can be changed easily going forward as needed or as you add more patches then we should be fine16:15
spzalashangxdy: but it's your call :-)16:15
shangxdy_Can you consider the translation to HOT? If you have i agree the latest patch16:15
spzalashangxdy: you are the one doing hard work on it16:15
spzalashangxdy_: sorry I didn't get your question16:16
spzalaif you are asking if this patch will help with translation or not?16:16
spzalaI think it should allow translation nicely, again it's hard unless we start playing with translation but I agree that we need to keep translation in mind16:17
shangxdy_How to translate to hot if the mapped node template associated with sub mapping class16:17
shangxdy_I means the brief approach for translation.16:18
spzalawhen we translate node template, it will give associated sub_mapping and then we replace and translate that particular node template with call to sub_mapping16:18
spzalae.g. sub_mapping.capabilities() or .requirements() or .interfaces()16:19
spzalawhen I thought about it, node template with associated sub_mapping object makes it easy on translation side16:20
shangxdy_But in my initial opinion, there is not much information for translation in sub mapping class .16:20
spzalashangxdy: yes, I agree we definitely need to consider brief approach16:21
spzalathat's true, but we just need the data and replace the node template with sub_mapping data or node template16:21
spzalaand than continue translation as usual16:21
spzalait's similar to say NodeTemplate or other classes doesn't have  much information for translation but it gives a mean to iterate over data16:23
spzalalike properties, req, cap, etc and we use that during translation16:23
shangxdy_In my patch, sub mapping class is only used to validate and associated. and it doesn't replace any data.16:24
spzalatrue, we are not replacing any data no where in fact but16:25
spzalasub mapping class should provide the data like, submapping.type or etc16:25
spzalathe class shouldn't only do validation but provide all those data about itself16:26
shangxdy_I known your idea. but i think it's will be more complicated for translation.16:26
spzalathanks, I couldn't think about complication (again, you know it better than me as you are having a bigger picture with owing blueprints on both side)16:27
spzalaI guess during the translation, we have to do similar things as I was saying16:28
spzalawe can handle it similar ways like translation of node templates, groups, policy etc. .. the major difference is first we need to get sub_mapping object which will provide node template16:29
shangxdy_Sure, submapping.node_type submapping.capabilites and submapping.requirements will be provided in sub mapping class. But how to use it is different between us:)16:29
spzalaand then we use the data of that node template instead of the one in service template16:30
spzalaof course after validation16:30
shangxdy_Yes, that's the point between us.16:31
spzala:) well again I will respect your decision but if the diff between approach is not major then we can go ahead with the patch16:31
spzalaand modify as we need unless the current code doesn't set the base ..16:32
spzala:-) thanks, certain things we can only find out when we do actual translation but let's focus on the correct parsing and if you think we are doing it then during translation we may need to add some more code on translator side or even changes in parser16:34
shangxdy_We shall run to the same direction, and turn around if it doesn't work.16:34
spzalawe always need to remember that purpose of tosca-parser project (when I spun the code from translator) was to allow user to use parser outside translation need16:35
spzalashagnxdy_: absolutely, that's why I said if the base code are good then we go forward16:35
spzalaso definitely need to keep translation in mind and if we can assure here that we can correctly parse the template provided with sub_mappings then translation should be possible16:36
shangxdy_It sounds great:)16:36
spzalaevery patch that goes in parser impacts translation code one way or other16:37
spzala:-) thanks16:37
spzalaso I would say, please go over the patch one more time and if you think it looks good (if not great)16:38
spzalawe can merge it and then have your other patches come in with tests etc16:38
spzalaand as we do that, hopefully we will get a better picture within parser itself16:38
spzalai.e. before creating next pypi which will be used in translation16:38
spzalaand fix anything before next pypi16:39
shangxdy_Ok, a question, every patch needs a bp or bug in launchpad?16:39
spzalanot really, it's totally find to create patch without a bug or bp unless the patch is touching something that needs to be described for a better review of code changes16:40
shangxdy_I think we can complete the sub mapping function in parser and translator before next release.16:40
spzalaif it's straightforward change, no bug or bp needed16:40
spzalashangxdy_: i am sure and that sounds great.. once we have initial implementation working then we iterate over it with complex use cases as we go further in following release cycles16:41
spzalashangxdy: please feel free to ping or email me with any question .. sub_mapping is very important and thanks for working on it :-)16:45
spzalashangxdy: anything else from you? or even on sub mapping right now ?16:45
shangxdy_Next week I want to spend some time on OPNFV, but i'll alsowork for the patch16:47
spzalawell, and as we just talked the gate issue is fixed .. stevemar: thanks a bunch!16:47
spzalashangxdy: sure, if we can get this patch merged once you think it's alright we can merge this one .. but please make any changes as needed or we can discuss before this merge16:48
spzalashangxdy_: :-) yup16:48
shangxdy_Another question, we can discuss it next time: what's the different between abstract node template and substituted node template? and how to implement the tow node template?16:50
spzalasure, to me they are same16:51
spzalaabstract node is implemented by sub node16:51
spzalaI think that's the abstraction16:52
shangxdy_The tow node templates are different in spec. but the implementations are similar or same16:52
shangxdy_I think sub node is subset of abstract node.16:54
spzalaI see, not sure where exactly in spec you are mentioned but abstract is what is not implemented and sub mapping makes that abstraction possible16:54
spzala:-) I guess or other way16:55
shangxdy_The chapter 2.10 in spec.16:56
spzalabut to me that's the relationship between two of them16:56
spzalayup the chapter on sub mapping16:57
spzalayou agree with me on the relationship part right?16:58
spzalawould like to know if it's more than that16:58
*** bobh has quit IRC16:58
shangxdy_Agree, in fact, in my patch, i consider the tow node templates the same in function _is_sub_mapped_node.16:58
spzalashangxdy: yup, got to tell that you have understood the concept of sub_mapping well .. it's not easy to grasp honestly16:59
spzalaonce you understand it looks easy (well that's with everything)17:00
spzalashangxdy: we are over time on meeting ;) let's end?17:02
spzalashangxdy: we can continue discussion though until you feel sleepy :-) and I feel hungry :)17:03
shangxdy_I think template composition in HOT  is the best example of sub_mappings17:03
shangxdy_O, it's over time:)17:04
shangxdy_I'am sleepy once you said that:)17:05
spzalashangxdy: lol17:05
spzalashangxdy: wait, what did u mean?17:05
shangxdy_Have a good launch:)17:05
spzalashangxdy: OK, I see you mean in Heat itself?17:05
shangxdy_Yes, nested template solution in heat.17:06
spzalashangxdy_: yup, I am with you on that17:06
spzalashangxdy_: alright then, good night and I am heading to lunch17:07
shangxdy_good lunch.17:07
spzalashangxdy_: :) thanks17:08
shangxdy_By, spzala and all17:09
spzalashangxdy:_ bye17:10
*** shangxdy_ has quit IRC17:10
*** bobh has joined #openstack-heat-translator17:59
*** bobh has quit IRC18:03
*** bobh has joined #openstack-heat-translator18:35
*** bobh has quit IRC21:20
*** zaneb has joined #openstack-heat-translator21:30
*** zaneb has quit IRC21:30
*** zaneb has joined #openstack-heat-translator21:31
*** bobh has joined #openstack-heat-translator21:33
*** bobh has quit IRC21:37

Generated by 2.14.0 by Marius Gedminas - find it at!