Monday, 2015-08-17

*** kgoedecke has joined #openstack-sdks00:17
*** kgoedecke has left #openstack-sdks00:17
*** boris-42 has quit IRC01:10
*** boris-42 has joined #openstack-sdks01:38
*** Yanyanhu has joined #openstack-sdks01:47
*** gildub has joined #openstack-sdks02:58
openstackgerritSteve Baker proposed openstack/cliff: Implement a yaml formatter in cliff
*** britthouser has joined #openstack-sdks04:11
*** britthou_ has joined #openstack-sdks04:12
*** britthouser has quit IRC04:15
*** openstack has joined #openstack-sdks04:20
*** yonglihe has joined #openstack-sdks06:01
*** john_dev has quit IRC06:03
*** john_dev has joined #openstack-sdks06:03
*** john_dev is now known as john-dev06:04
-openstackstatus- NOTICE: Gerrit is currently under very high load and may be unresponsive. infra are looking into the issue.07:07
*** markus_z has joined #openstack-sdks07:14
*** aufi has joined #openstack-sdks07:35
*** gildub has quit IRC07:55
*** Guest47951 is now known as d0ugal09:49
*** d0ugal has quit IRC09:49
*** d0ugal has joined #openstack-sdks09:49
-openstackstatus- NOTICE: (aka gerrit) is going down for an emergency restart10:22
*** ChanServ changes topic to " (aka gerrit) is going down for an emergency restart"10:22
*** Yanyanhu has quit IRC10:34
*** terrylhowe has joined #openstack-sdks10:34
*** Qiming has joined #openstack-sdks10:46
*** ChanServ changes topic to "#openstack-sdks"10:49
-openstackstatus- NOTICE: Gerrit restart has resolved the issue and systems are back up and functioning10:49
*** Qiming has quit IRC10:59
*** boris-42 has quit IRC11:10
*** woodster_ has joined #openstack-sdks11:46
*** jose4183 has joined #openstack-sdks11:55
*** chlong has quit IRC11:59
*** Qiming has joined #openstack-sdks12:50
*** Qiming has quit IRC12:51
*** pm90_ has joined #openstack-sdks12:54
*** pm90__ has joined #openstack-sdks12:57
*** pm90_ has quit IRC12:59
*** britthouser has joined #openstack-sdks13:05
*** beagles has quit IRC13:26
*** keekz has quit IRC13:28
*** keekz has joined #openstack-sdks13:29
*** Zucan has quit IRC13:36
*** britthou_ has joined #openstack-sdks13:36
*** britthouser has quit IRC13:39
*** keekz has quit IRC13:41
*** keekz has joined #openstack-sdks13:55
*** petertr7_away is now known as petertr713:56
*** keekz has quit IRC13:56
*** Zucan has joined #openstack-sdks13:58
*** boris-42 has joined #openstack-sdks14:00
*** chlong has joined #openstack-sdks14:03
*** sigmavirus24_awa is now known as sigmavirus2414:15
*** keekz has joined #openstack-sdks14:31
*** britthou_ has quit IRC14:39
*** britthouser has joined #openstack-sdks14:39
*** morgan_503 is now known as morgan_254915:06
mordredShrews: if you get a sec, ... keystone is on board with it15:17
*** chlong has quit IRC15:18
Shrewsmordred:  will look15:19
*** liangbo has joined #openstack-sdks15:27
*** e0ne has joined #openstack-sdks15:27
Shrewsmordred: keystone is onboard, but what about osc?15:27
Shrewsmordred: i sort of feel dtroyer should weigh in first15:28
mordredShrews: osc shouldn't care - they have a different thing named token_endpoint which would remain token_endpoint15:28
mordredShrews: the issue is that ksc never published a plugin called token_endpoint, but osc did15:29
Shrewsmordred: so what's preventing this from breaking the world again?15:29
mordredShrews: the fact that the two different plugins will now have two different names15:30
Shrewsmordred: and that had been released?15:30
Shrewsin ksc15:30
mordredno - actually, once it's released in ksc we don't need this patch15:30
Shrewsor maybe i'm confusing myself15:30
mordredthis is just a patch to let people using occ access the ksc auth plugin using the name it will be published under before ksc releases with that patch applied15:31
mordredit should additionally not step on the osc plugin that does a similar thing15:31
*** chlong has joined #openstack-sdks15:31
*** etoews has joined #openstack-sdks15:32
Shrewsmordred: ok. let me try this question: before the first revert, we broke osc. is something in place (a requirements version change maybe?) that keeps us from breaking osc again?15:32
mordredShrews: gotcha - so the difference is that the patch is different15:33
mordredShrews: we broke osc before because we looked for the name 'token_endpoint' and loaded the ksc class directly if it was there15:34
mordredShrews: except that osc had a different thing called token_endpoint which takes a differently named parameter (url vs. endpoint)15:34
mordredShrews: in this patch, we look for a plugin name 'admin_token'15:34
Shrewsah ha! eureka... and what not15:34
Shrewsi see now15:34
* Shrews just doing his due diligence. thx15:35
openstackgerritMerged openstack/os-client-config: Revert "Revert "Use the correct auth_plugin for token authentication""
*** chlong has quit IRC15:38
*** trown is now known as trown|lunch15:40
*** chlong has joined #openstack-sdks15:40
*** liangbo has quit IRC15:41
*** aufi has quit IRC15:42
*** dguerri` is now known as dguerri15:46
*** e0ne has quit IRC16:09
briancurtinterrylhowe, etoews: terry - while writing up this blog post, i realized we now have both plugins and extensions inside the SDK, and they're both loaded up and worked with in teh same way. i think we should pick one name for these things, and i'm siding with plugin - thoughts?16:10
etoewsextensions is a dirty word in openstack-land. you'll immediately get people railing against extensions without even looking at what they are in the sdk. plugin++16:13
*** david-ly_ is now known as david-lyle16:15
*** stevemar has joined #openstack-sdks16:24
*** pm90__ has quit IRC16:28
*** trown|lunch is now known as trown16:44
*** dguerri is now known as dguerri`17:09
*** markus_z has quit IRC17:38
*** pm90_ has joined #openstack-sdks17:58
*** pm90__ has joined #openstack-sdks18:00
*** pm90_ has quit IRC18:03
*** stevemar has quit IRC18:08
*** stevemar has joined #openstack-sdks18:10
openstackgerritMerged openstack/cliff: Updated from global requirements
*** jose4183 has quit IRC18:14
*** jose4183 has joined #openstack-sdks18:14
*** jose4183 has quit IRC18:19
*** jose4183 has joined #openstack-sdks18:19
*** jose4183 has quit IRC18:23
*** jose4183 has joined #openstack-sdks18:23
openstackgerritMerged openstack/cliff: Improve help messages
*** e0ne has joined #openstack-sdks18:40
*** britthou_ has joined #openstack-sdks18:43
*** britthouser has quit IRC18:45
*** e0ne has quit IRC19:03
*** e0ne has joined #openstack-sdks19:09
*** petertr7 is now known as petertr7_away19:24
*** petertr7_away is now known as petertr719:29
*** e0ne has quit IRC19:45
*** e0ne has joined #openstack-sdks19:48
terrylhowebriancurtin: plugin is fine with me. Since ksc/ksa are going to just calling their plugins ‘auth_type’ I don’t think there will be much confusion.19:54
openstackLaunchpad bug 1474982 in openstack-ansible "auth_plugin to auth_type" [Medium,Confirmed]19:57
terrylhoweI’ve been kind of sitting on that one since the move to ksa will cover a lot of that19:58
*** Shrews has quit IRC20:47
openstackgerritTerry Howe proposed stackforge/python-openstacksdk: Remove connection CRUD methods
*** Shrews has joined #openstack-sdks20:53
*** trown is now known as trown|outttypeww20:53
*** Shrews has quit IRC20:53
etoewsbriancurtin: terrylhowe: where does module level function docs go?20:59
*** Shrews has joined #openstack-sdks20:59
etoewsi'm looking at
etoewsbut don't see any docs for from_config (which will be a pretty important function)21:00
terrylhoweHmm, etoews from_config is probably missing the right doc strings21:01
etoewsi was actually updating its docstring which is how i discovered this21:01
terrylhowelooks like automodule is set for the first section on that page21:02
etoewsi tried :members: underneath automodule:: but that didn't seem to work21:06
terrylhowestuff like this shows up:
etoewsokay. it actually did work.21:10
etoewsbut it comes right after the "Find or create" section and, without a section heading of its own, it's really difficult to pick out.21:10
terrylhoweMaybe you could ======= it21:13
*** britthou_ has quit IRC21:14
*** ameybhide has joined #openstack-sdks21:18
etoewsalright. i found something that looks decent.21:21
openstackgerritEverett Toews proposed stackforge/python-openstacksdk: Improve the from_config doc
openstackgerritSteve Baker proposed openstack/cliff: Implement a yaml formatter in cliff
openstackgerritSteve Baker proposed openstack/cliff: Implement a json formatter in cliff
*** pm90__ has quit IRC21:42
*** stevebaker has joined #openstack-sdks21:47
stevebakerdhellmann: are you about for a cliff question22:00
*** petertr7 is now known as petertr7_away22:01
openstackgerritTerry Howe proposed stackforge/python-openstacksdk: Move service filter out of auth
*** pm90_ has joined #openstack-sdks22:16
*** pm90__ has joined #openstack-sdks22:18
*** pm90_ has quit IRC22:21
*** jose4183 has quit IRC22:25
*** ameybhide has quit IRC22:28
dhellmannstevebaker: what's up?22:32
*** chlong has quit IRC22:34
* dhellmann goes in search of food22:34
stevebakerdhellmann: the output of many commands exceed the width of the console, making prettytable output very unpretty. I'm wondering if console width detection could be added to cliff, and the table formatter can be given rules on how and what to truncate to fit in the console22:36
*** ameybhide has joined #openstack-sdks22:40
*** ameybhide has quit IRC22:41
*** e0ne has quit IRC22:41
*** ameybhide has joined #openstack-sdks22:41
dhellmannstevebaker: maybe? are you offering? :-)23:05
stevebakerdhellmann: I am :)23:05
stevebakersince most heat output has data which exceeds screen width23:06
dhellmanndo you think it should be the default behavior, or an option? on narrow consoles I could see it making some of the output even worse23:06
dhellmannor at least not better, I guess23:07
stevebakerdhellmann: I'm thinking opt-in per command, so no truncation is done unless a developer specifies truncation rules. And those rules would be tailored to make output better not worse23:08
dhellmannah, I see23:08
stevebakerso take something like this
dhellmannso as a developer, how would that work? would I suggest leaving certain columns out of the output entirely? or reformat the data I'm giving in a column based on width?23:09
dhellmannor both?23:10
stevebakerRules could be the Value column gets truncated when short on space, and the links row has a truncation formatter which wraps using / as a delimiter23:10
*** gildub has joined #openstack-sdks23:11
dhellmannwe could do that already, but it would be interesting to see what a general API looks like23:11
stevebakerother truncation formatters could cut of the end (or the start) to fit available space23:11
dhellmannwhere would you tie in the rules? I guess just inside the table formatter?23:12
stevebakerand truncation would never be attempted if there is no interactive console, or if they specify the table formatter arg --dontdothatplease23:13
stevebakerdhellmann: haven't got that far yet :) first challange is how to add console width detection23:13
stevebakerwidth detection is different on py27/py3/linux/windows, so it might be better to add blessings as a dependency
dhellmannor blessings23:17
stevebakera better PrettyTable, even better23:18
stevebakerdhellmann: its not obvious whether terminaltables does width detection. I'll have a play23:21
dhellmannstevebaker: ++23:24
openstackgerritMerged stackforge/python-openstacksdk: Improve the from_config doc
*** britthouser has joined #openstack-sdks23:57
*** britthou_ has joined #openstack-sdks23:58

Generated by 2.14.0 by Marius Gedminas - find it at!