Monday, 2018-06-11

« Sunday, 2018-06-10 Index Tuesday, 2018-06-12 »
*** felipemonteiro has quit IRC00:24
*** r-daneel has quit IRC00:40
*** Dinesh_Bhor has joined #openstack-keystone00:58
*** zxy has joined #openstack-keystone01:05
*** lifeless has quit IRC01:06
*** lifeless has joined #openstack-keystone01:11
openstackgerritwangxiyuan proposed openstack/keystone master: Unified limit update APIs Refactor  https://review.openstack.org/55955201:31
*** felipemonteiro has joined #openstack-keystone01:39
*** felipemonteiro_ has joined #openstack-keystone01:40
*** felipemonteiro has quit IRC01:43
*** felipemonteiro has joined #openstack-keystone01:47
*** edmondsw has quit IRC01:47
*** lifeless has quit IRC01:47
*** lifeless has joined #openstack-keystone01:49
*** felipemonteiro_ has quit IRC01:50
*** r-daneel has joined #openstack-keystone02:01
*** annp has joined #openstack-keystone02:03
*** felipemonteiro has quit IRC02:16
*** lifeless has quit IRC02:19
*** lifeless has joined #openstack-keystone02:20
*** germs has quit IRC02:28
*** germs has joined #openstack-keystone02:29
*** felipemonteiro has joined #openstack-keystone02:41
*** edmondsw has joined #openstack-keystone02:54
*** edmondsw has quit IRC02:59
*** Dinesh_Bhor has quit IRC03:05
*** zxy has quit IRC03:28
*** zxy has joined #openstack-keystone03:37
openstackgerritwangxiyuan proposed openstack/keystone master: Unified limit update APIs Refactor  https://review.openstack.org/55955203:45
*** lifeless_ has joined #openstack-keystone03:46
*** lifeless has quit IRC03:46
*** Dinesh_Bhor has joined #openstack-keystone03:47
*** lifeless has joined #openstack-keystone03:52
*** lifeless_ has quit IRC03:52
*** annp has quit IRC04:13
*** links has joined #openstack-keystone04:14
*** annp has joined #openstack-keystone04:14
*** gagehugo has quit IRC04:24
*** dklyle has joined #openstack-keystone04:25
*** dklyle has quit IRC04:31
*** gagehugo has joined #openstack-keystone04:37
*** germs has quit IRC04:46
*** felipemonteiro has quit IRC04:52
*** Dinesh_Bhor has quit IRC05:00
*** itlinux has joined #openstack-keystone05:00
*** Dinesh_Bhor has joined #openstack-keystone05:23
*** mvk has joined #openstack-keystone05:39
*** AlexeyAbashkin has joined #openstack-keystone06:04
*** sheel has joined #openstack-keystone06:19
*** itlinux has quit IRC06:24
*** pcaruana has joined #openstack-keystone06:26
*** belmoreira has joined #openstack-keystone06:27
*** martinus__ has joined #openstack-keystone06:35
*** Dinesh_Bhor has quit IRC06:51
*** Dinesh__Bhor has joined #openstack-keystone06:51
openstackgerritwangxiyuan proposed openstack/keystone master: Add yaml-loaded filesystem catalog backend  https://review.openstack.org/48351406:58
*** jaosorior has joined #openstack-keystone07:10
*** rcernin has quit IRC07:10
*** pooja_jadhav has joined #openstack-keystone07:25
openstackgerritJuan Antonio Osorio Robles proposed openstack/keystone master: Ensure default roles created during bootstrap  https://review.openstack.org/57224307:30
*** belmoreira has quit IRC07:36
openstackgerritJuan Antonio Osorio Robles proposed openstack/keystone master: Ensure default roles created during bootstrap  https://review.openstack.org/57224307:37
openstackgerritwangxiyuan proposed openstack/keystone master: Add yaml-loaded filesystem catalog backend  https://review.openstack.org/48351407:48
*** belmoreira has joined #openstack-keystone07:48
*** links has quit IRC07:50
*** belmoreira has quit IRC07:53
*** lifeless_ has joined #openstack-keystone08:06
*** lifeless has quit IRC08:07
*** namnh has joined #openstack-keystone08:08
*** jmlowe has quit IRC08:10
*** ispp has joined #openstack-keystone08:17
*** d0ugal has joined #openstack-keystone08:28
*** belmoreira has joined #openstack-keystone08:33
openstackgerritSunYanbin proposed openstack/keystone master: a bug for test by my own Bug1776161  https://review.openstack.org/57416908:37
*** lifeless has joined #openstack-keystone08:46
*** lifeless_ has quit IRC08:47
openstackgerritwangxiyuan proposed openstack/keystone master: Add yaml-loaded filesystem catalog backend  https://review.openstack.org/48351409:58
*** Dinesh__Bhor has quit IRC09:58
openstackgerritAlessandro Nesta proposed openstack/oslo.limit master: fix tox python3 overrides  https://review.openstack.org/57395410:29
*** namnh has quit IRC10:34
*** nicolasbock has joined #openstack-keystone10:38
*** jrollen is now known as jroll10:44
*** zzzeek_ has quit IRC11:01
*** zzzeek has joined #openstack-keystone11:03
*** jroll has quit IRC11:18
*** jroll has joined #openstack-keystone11:18
*** annp has quit IRC11:19
*** lifeless has quit IRC11:45
*** lifeless_ has joined #openstack-keystone11:45
*** edmondsw has joined #openstack-keystone12:12
*** raildo has joined #openstack-keystone12:34
*** pcaruana has quit IRC12:39
*** odyssey4me has quit IRC12:39
*** odyssey4me has joined #openstack-keystone12:39
*** jistr is now known as jistr|mtg12:41
*** AlexeyAbashkin has quit IRC12:42
*** jistr|mtg is now known as jistr12:57
openstackgerritJuan Antonio Osorio Robles proposed openstack/keystone master: Ensure default roles created during bootstrap  https://review.openstack.org/57224313:02
jaosoriorhrybacki: did a tempest patch to get the implied roles tests to work with the default roles addition https://review.openstack.org/#/c/574149/13:03
jaosoriorFederico Ressi, are you around?13:05
hrybackijaosorior: what happens if someone wants to use tempest to test outside of the default roles?13:08
jaosoriorhrybacki: huh?13:09
hrybackilemme finish reviewing before I start asking questions :P /me shuts up13:09
jaosoriorhrybacki: so we have the deafult roles, and those are left untouched. Who knows, the deployer might even change them at some point, and that's cool13:10
jaosoriorhrybacki: the tempest test creates some test roles, with random names13:10
jaosoriorwhat my tempest patch does is filter out any other implied roles that might exist in the system, and only use the roles relevant to that test13:10
hrybackiah yes I see13:11
jaosoriorhrybacki: if it's not very understandable I can try to simplify it. Maybe I tried to be too clever :D readability is better13:12
*** jdennis has quit IRC13:27
*** jdennis has joined #openstack-keystone13:27
*** ispp has quit IRC13:29
*** ispp has joined #openstack-keystone13:31
*** pcaruana has joined #openstack-keystone13:32
*** belmorei_ has joined #openstack-keystone13:32
*** belmoreira has quit IRC13:34
*** sheel has quit IRC13:49
openstackgerritDavid Rabel proposed openstack/keystone master: Fix typo in docs  https://review.openstack.org/57426613:56
*** xinran__ has joined #openstack-keystone13:58
*** AlexeyAbashkin has joined #openstack-keystone14:04
*** Alexey_Abashkin has joined #openstack-keystone14:11
*** AlexeyAbashkin has quit IRC14:13
*** Alexey_Abashkin is now known as AlexeyAbashkin14:13
*** mchlumsky has joined #openstack-keystone14:28
*** jmlowe has joined #openstack-keystone14:31
cmurphylbragstad: thanks for the assist with the update, sorry I didn't get it out last week14:31
lbragstadcmurphy: no worries - i should have caught it last week too, but i was wrapped up in the enforcement model stuff :)14:32
*** dklyle has joined #openstack-keystone14:40
*** dklyle has quit IRC14:47
*** r-daneel has quit IRC14:50
openstackgerritLance Bragstad proposed openstack/keystone-specs master: Address follow-on comments in strict-two-level spec  https://review.openstack.org/57329214:50
*** ispp has quit IRC14:51
*** ispp has joined #openstack-keystone14:54
jaosoriorhrybacki: for some reason we're still getting for implied roles even if I tried to filter them out :/15:05
*** jmlowe has quit IRC15:08
knikollao/15:09
*** r-daneel has joined #openstack-keystone15:25
*** fiddletwix has joined #openstack-keystone15:25
*** felipemonteiro has joined #openstack-keystone15:25
*** r-daneel_ has joined #openstack-keystone15:28
*** r-daneel has quit IRC15:29
*** r-daneel_ is now known as r-daneel15:29
*** jmlowe has joined #openstack-keystone15:32
hrybackijaosorior: in meeting, 1 sec15:32
*** felipemonteiro has quit IRC15:34
*** itlinux has joined #openstack-keystone15:36
*** ispp has quit IRC15:40
*** r-daneel_ has joined #openstack-keystone15:42
*** belmorei_ has quit IRC15:42
*** r-daneel has quit IRC15:44
*** r-daneel_ is now known as r-daneel15:44
hrybackijaosorior: I saw that and was also confused. No obvious answer15:53
jaosoriorWell, in an hour I'm gonna end my day, do you have a chance to check that out?15:53
hrybackijaosorior: I've a few more meetings on my books today but will certainly try :)15:54
hrybackimonday/tuesday are always meeting slammed for me =/15:54
*** lifeless_ has quit IRC15:59
*** lifeless has joined #openstack-keystone16:00
*** dtruong_ has quit IRC16:05
*** AlexeyAbashkin has quit IRC16:11
*** AlexeyAbashkin has joined #openstack-keystone16:13
*** r-daneel has quit IRC16:13
*** felipemonteiro has joined #openstack-keystone16:19
*** agrebennikov has joined #openstack-keystone16:24
*** r-daneel has joined #openstack-keystone16:25
*** r-daneel has quit IRC16:29
kmalloco/16:30
*** r-daneel has joined #openstack-keystone16:30
*** felipemonteiro_ has joined #openstack-keystone16:33
*** felipemonteiro__ has joined #openstack-keystone16:34
*** Alexey_Abashkin has joined #openstack-keystone16:35
*** Kumar has joined #openstack-keystone16:36
*** AlexeyAbashkin has quit IRC16:36
*** Alexey_Abashkin is now known as AlexeyAbashkin16:36
*** felipemonteiro has quit IRC16:37
*** felipemonteiro_ has quit IRC16:37
*** Alexey_Abashkin has joined #openstack-keystone16:42
*** AlexeyAbashkin has quit IRC16:44
*** Alexey_Abashkin is now known as AlexeyAbashkin16:44
*** xinran__ has quit IRC16:58
openstackgerritMerged openstack/python-keystoneclient master: Fix python3 test compat  https://review.openstack.org/57377316:59
openstackgerritMerged openstack/python-keystoneclient master: fix tox python3 overrides  https://review.openstack.org/57333916:59
*** felipemonteiro__ is now known as felipemonteiro17:02
*** gyee has joined #openstack-keystone17:03
felipemonteirohrybacki lbragstad: is there anything else i need to do to https://review.openstack.org/#/c/464678/ or implement patrole-side prior to approval?17:03
*** germs has joined #openstack-keystone17:07
*** germs has quit IRC17:07
*** germs has joined #openstack-keystone17:07
*** felipemonteiro has quit IRC17:31
*** dklyle has joined #openstack-keystone17:32
*** felipemonteiro has joined #openstack-keystone17:35
*** AlexeyAbashkin has quit IRC17:35
*** felipemonteiro has quit IRC17:37
openstackgerritLance Bragstad proposed openstack/python-keystoneclient master: Add support for registered limits  https://review.openstack.org/53766817:39
*** felipemonteiro has joined #openstack-keystone17:39
lbragstadcurious what people think of ^17:39
lbragstadi based it on wxy's work for fixing up the unified limits api17:39
lbragstadbut the testing bits are a little strange just because the unified limits API it different from other API we support in keystone17:39
lbragstadand the assumptions some of the base CrudTests in python-keystoneclient make17:40
lbragstadit's clearly monday today, since my grammar in those last two sentences was way worse than it usually is17:42
*** felipemonteiro has quit IRC17:49
*** felipemonteiro has joined #openstack-keystone17:50
*** felipemonteiro_ has joined #openstack-keystone17:51
lbragstadkmalloc: should we update this? https://blueprints.launchpad.net/keystone/+spec/keystone-pecan17:53
*** felipemonteiro has quit IRC17:54
*** r-daneel has quit IRC18:02
*** ckonstanski has quit IRC18:06
*** r-daneel has joined #openstack-keystone18:09
*** Kumar has quit IRC18:09
*** dklyle has quit IRC18:17
*** dave-mcc_ has quit IRC18:25
*** dave-mccowan has joined #openstack-keystone18:27
*** dtruong has joined #openstack-keystone18:31
*** oikiki has joined #openstack-keystone18:34
*** felipemonteiro_ has quit IRC18:36
*** dklyle has joined #openstack-keystone18:48
*** r-daneel has quit IRC18:52
*** r-daneel has joined #openstack-keystone18:56
kmalloclbragstad: heh. probably18:58
*** felipemonteiro has joined #openstack-keystone18:59
lbragstaddo you know what the outcome was of pecan support in openstack?18:59
kmallocsome projects adopted it18:59
kmallocsome didn't18:59
lbragstadjamie's summary in that blueprint made it sound like it was a forsure thing18:59
kmallocsome did something else18:59
kmallocthat was back when wsme and pecan was a very large effort in openstack.18:59
kmalloci don't know the state of pecan and wsme, but it didn't work for us for some reasons19:00
kmallocand i don't remember why19:00
lbragstadhmm19:00
lbragstadintersting19:00
kmalloctbh it could probably be used now. but we had some very wonky things in the past19:02
*** jmlowe has quit IRC19:04
*** Neptu has joined #openstack-keystone19:04
jaosoriorhrybacki: so... The keystone tempest job ran referencing the wrong patchset (probably an error on my side), rechecked the job.19:16
hrybackijaosorior: ahhhh. I should have seen that19:18
hrybackiglad you did, I'm failing to get a instance to test against today w/ RDO Cloud19:18
jaosoriorhrybacki: I actually didn't see that :D had to ask in the infra channel and clarkb was kind enough to point that out19:19
hrybackithanks clarkb :)19:19
openstackgerritLance Bragstad proposed openstack/python-keystoneclient master: Add support for registered limits  https://review.openstack.org/53766819:25
openstackgerritLance Bragstad proposed openstack/python-keystoneclient master: Add support for project-specific limits  https://review.openstack.org/57439119:25
*** aeline has joined #openstack-keystone19:27
aelineHey, trying to modify keystone so that there is some more tables in the DB. Where does the the DB initialization actually happen?19:28
aelineSorry if this is the wrong place to ask19:29
lbragstadaeline: keystone provides a command line tool call `keystone-manage`19:34
lbragstadwhich offers a subcommand that is used to bootstrap keystone with information19:34
lbragstadhttps://docs.openstack.org/keystone/latest/cli/index.html#keystone-manage19:34
lbragstadhrybacki: i'm going through the trello board19:34
aelineIs adding classes under sql_model.py not the right way to go about this?19:35
* hrybacki listnes19:35
hrybackilistens*19:35
lbragstadhrybacki: a bunch of stuff had due dates of april 20th, what should we do with those, re-target them?19:35
lbragstadaeline: oh - i'm sorry, i misunderstood your question19:35
hrybackilbragstad: aye, likely. I'm running out the door here in a min to some appts. Have time for a review session together tomorrow morning?19:35
lbragstadaeline: keystone uses database migrations with sqlalchemy to make changes to the database19:36
hrybackigood time to catch you up on our policy work too19:36
lbragstadaeline: https://docs.openstack.org/keystone/latest/contributor/database-migrations.html19:36
lbragstad^ that has some information on the process19:36
lbragstadhrybacki: sure19:36
lbragstadjust ping me whenever19:36
aelineThanks! Appreciate the help.19:36
lbragstadaeline: yep19:37
lbragstadhrybacki: i don't have anything prior to the meeting19:37
*** mvk has quit IRC19:39
hrybackilbragstad: invite enroute19:40
kmallochrybacki: toss me an invite as well if you don't mind me sitting in on it19:41
kmalloclbragstad: ^19:41
lbragstadsure thing19:41
kmalloclbragstad: almost ready to post the initial "convert to flask" review(s)19:42
hrybackikmalloc: personal or rh email?19:42
kmallochrybacki: pref. @gmail, but honestly, it works either way19:42
hrybackiyour wish :)19:42
kmalloclbragstad: hmm. i am not liking the way this is shaping up. i think i need to re-work it. let me post what i'm up to and show you what i don't like.19:44
kmallochm. or. wonder if i can do the @url_for19:44
* kmalloc checks19:44
lbragstadok19:45
kmallocnop url_for wont fix it.19:45
openstackgerritMorgan Fainberg proposed openstack/keystone master: Add Flask-RESTful as a requirement  https://review.openstack.org/57441419:46
openstackgerritMorgan Fainberg proposed openstack/keystone master: Implement scaffolding for Flask-RESTful use  https://review.openstack.org/57441519:46
openstackgerritMorgan Fainberg proposed openstack/keystone master: [WIP] Convert Discovery (Version) subsystem to Flask  https://review.openstack.org/57441619:46
kmalloclbragstad: ok so that last one [WIP] one.19:46
kmallocif you look in keystone.subsystem.discovery.api19:47
kmallocyou'll see how each path gets a resource implemented for it19:47
kmallocand how it's added to the API object19:47
kmallocin api_obj.resource_mapping19:48
kmallocmaybe i should make the API object more of a soft-import time registry? i am trying to avoid cross-subsystem imports and/or where resources don't need to be explicitly added to the Resource_mapping attr19:48
lbragstadhmm19:49
kmallocso in the case of something that extends /USERs19:49
kmallocyou'd need the /USERS api object19:49
kmallocthen somehow add the resource19:49
kmallocot the resource_mapping19:49
kmalloc*or* you'd need to support cross-subsystem controller instantiation19:49
kmallocunless we want an "extends" api, so for example if it addes to user, a new API object is created with the deeper prefix19:50
kmalloce.g. /users/{user_id}/credentials/OS-EC219:51
kmallocno that doesn't work because of the {user_id} bit.19:52
kmallocugh.19:52
kmallocso 3 possible ways to approach this:19:53
kmalloc1) migrate the controller code to the resource objects19:53
kmalloc(most flask-like)19:53
kmalloc2) support cross subsystem imports and19:53
kmallocinstantiate controllers from the cross-system on the local resource19:54
kmallocor 3) support cross subsystem imports and allow the resources to register themselves in the API object's resource_mapping attribute19:54
kmallocin case #1, code would be always isolated to the prefix, so /users/{user_id}/credentials/OS-EC2 would live in identity, even though EC2 "owns" the resulting data19:55
kmallocthoughts?19:55
lbragstadsorry - just wrapping up a call quick and i'll read back19:55
lbragstadsans distractions :)19:56
kmallocsure19:56
-openstackstatus- NOTICE: Zuul was restarted for a software upgrade; changes uploaded or approved between 19:30 and 19:50 will need to be rechecked19:56
kmallocoh, one more option: 4 - don't use flask-Restful and directly craft routes -- still has some of the pitfalls because the blueprint(s) are local to the subsystem (e.g. /users in Identity)20:00
*** hoonetorg has joined #openstack-keystone20:05
*** oikiki has quit IRC20:06
*** oikiki has joined #openstack-keystone20:07
*** r-daneel has quit IRC20:11
*** ispp has joined #openstack-keystone20:12
*** rmascena has joined #openstack-keystone20:22
*** r-daneel has joined #openstack-keystone20:22
*** raildo has quit IRC20:23
lbragstadkmalloc: ok - reading scroll back20:25
lbragstadhmm20:28
kmalloclbragstad: and the other question is, is it worth moving anything but the API/controllers to subsystem, maybe instead subsystem is the non-routed bits [aka managers/backends] and keystone.api is the routed (controllers) bit[s]?20:28
lbragstadyeah - i can see where that would make sense20:28
kmallocok, so i'll aim for keystone.api, (and worry about cleanup to keystone.subsystem later)20:29
kmallocbut that doesn't answer the same questions re: controller import outlined above20:29
lbragstadso option #120:29
lbragstadwould be reworking everything in say keystone/identity/controller.py20:30
kmallocyeah20:30
kmallocwell, we're doing a chunk of the work anyway20:30
kmallocto move to flask_restful20:30
*** rmascena has quit IRC20:31
kmallocbrb dog potty20:31
*** AlexeyAbashkin has joined #openstack-keystone20:32
*** r-daneel has quit IRC20:33
kmalloclbragstad: though with keystone.api, it means we would be much more broken out: keystone.api.user keystone.api.group, etc so cross import isn't important20:35
lbragstadsorry - dumb question20:36
lbragstadcan you define cross import for me?20:36
kmallockeystone.subsystem.user needs ot be imported (in the current patch-style) by keystone.subsystem.ec220:36
kmallocbecause ec2 extends the keystone.user API20:36
kmallocvia: /users/{user_id}/credentials/OS-EC220:37
kmallocsince ec2 *owns* that path, right now we just say "hey mapper, route that"20:37
kmallocin flask, we're going to use blueprints which define the prefix, e.g. "users" which is owned by identity20:38
kmallocso if we did keystone.subsystem.identity is where everything from keystone.identity went, ec2 would either need to add it's resource to keystone.subsystem.identity.api.UserAPI or ... some other mechanism to get it loaded under /user path-prefix20:39
lbragstadhmm20:39
lbragstadactually20:39
lbragstadit would be nice to know that all router logic for a given path is in a single place20:40
kmallocright20:40
kmallocso, if we did keystone.api.user (for example)20:40
kmallocall router logic for "/user" would live there20:40
kmallocand keystone.identity.core/backends/etc stays in either keystone.identity or eventually moves to keystone.subsystem.identity <-- .subsystem = internal manager stuff20:41
kmallockeystone.api.* <--- Router/Controller manager20:41
kmallocthat seems to make more sense to me.20:41
kmallocand should be a lot cleaner.20:41
lbragstadyeah - i can see that20:42
lbragstadeven though it goes against the grouping we've had for *years*20:42
lbragstad:)20:42
kmallocright20:42
kmallocnow the question i have is...20:42
kmalloci can do: router-only type logic, where i just map the items from the BP to the route in keystone.api (raw flask)20:43
kmallocor i can do the resource-based style flask-restful does, e.g. User->GET/POST/DELETE/ETC is an object where /users/{user_id}/credentials/OS-EC2 would be a separate "resource" object20:43
kmallocthat implements (in this case) "get"20:43
kmallocso controller logic goes into resource objects.20:44
lbragstadand the resource object would call to managers?20:44
kmallocyep20:44
kmallocuser_list.get would implement "list users"20:44
kmallocuser.get would implement "get me user X"20:45
*** aeline has quit IRC20:45
kmallocuser.post would implement "create user"20:45
kmallocuser.delete would implemnet "delete user"20:45
kmallockeystone.api.user.ec2Resource would implement "get ec2 cred for user"20:45
kmalloceach "resource" can implement GEt/POST/PUT/DELETE/PATCH/etc20:46
kmallocand it's based upon the methods on the resource object20:46
kmallocGET/HEAD are explicitly the same unless defined separately20:46
*** Alexey_Abashkin has joined #openstack-keystone20:46
lbragstadis each resource object coupled to a specific path?20:46
kmallocyep, with substitutions for elements like {user_id}20:46
lbragstadok20:47
*** AlexeyAbashkin has quit IRC20:47
*** Alexey_Abashkin is now known as AlexeyAbashkin20:47
lbragstadso if we need to go deeper on the path, we need more resource objects20:47
lbragstad?20:47
kmallocyes.20:47
kmallocwhich is why i liked the user.ec2resource example20:47
lbragstadok - that makes sense20:47
kmalloc ->> /users/{user_id}/credentials/OS-EC2 is on a distinct resource object from the main user CRUD20:47
lbragstadbut still relative to the base user path/resource object20:48
kmallocand user list is a different distinct resource because GET /v3/user is different than /v3/user/20:48
lbragstadahhhh20:48
kmallocyes, it would still be all tied to the path, so /user would be in keystone.api.user20:48
kmallocyou'd then implement all resources that matter for the /user path there20:48
lbragstadso - i'm thinking about this partially as a developer new the project20:48
lbragstadand following an incoming request20:49
kmallocright, which is why i'm bouncing it :)20:49
kmallocinstead of just trying to implement, i *think* as a new dev, this makes a lot more sense20:49
kmallocbut, wanted a second brain20:49
lbragstadthat seems reasonable to me20:49
kmallocthe request would hit Flask_App, be dispatched based upon the URL to the blueprint/API, e.g. /USERS20:50
kmallocand from there, the resources would define the behavior20:50
lbragstadby calling into the business logic in various managers20:50
kmallocand all resources (think of resources as a controller with the router built-in)20:50
*** r-daneel has joined #openstack-keystone20:50
kmallocwould live relative to their path-prefix20:50
lbragstadok20:50
kmallocand call the managers directly20:50
lbragstadcool20:51
kmallocso routers/controllers -> keystone.api.<path>20:51
*** Alexey_Abashkin has joined #openstack-keystone20:51
kmalloci'll need to see how i can implement the @protected bits20:51
*** AlexeyAbashkin has quit IRC20:51
*** Alexey_Abashkin is now known as AlexeyAbashkin20:51
kmallocbut that might be a more direct call of @protected underlying code (easier to break out) in the process.20:52
kmallocso. let me go re-do these patches with all of this in mind.20:52
kmalloci *think* this just got a lot simpler to write the base API class and the resources20:52
lbragstadi can get behind it if it makes removing @protected easier20:52
kmalloc@protected will be a headache no matter what to deal with20:53
lbragstadand more structured organization is a plus20:53
kmallocyeah.20:53
*** felipemonteiro_ has joined #openstack-keystone20:55
*** AlexeyAbashkin has quit IRC20:56
* knikolla is going to read back this convo tonight. 21:01
*** felipemonteiro_ has quit IRC21:01
*** r-daneel has quit IRC21:08
*** AlexeyAbashkin has joined #openstack-keystone21:08
*** r-daneel has joined #openstack-keystone21:09
*** oikiki has quit IRC21:13
*** ispp has quit IRC21:15
*** oikiki has joined #openstack-keystone21:17
*** oikiki has quit IRC21:20
*** lifeless_ has joined #openstack-keystone21:28
*** lifeless has quit IRC21:28
*** r-daneel has quit IRC21:31
*** lifeless_ has quit IRC21:38
*** lifeless has joined #openstack-keystone21:44
*** martinus__ has quit IRC21:48
*** r-daneel has joined #openstack-keystone21:50
*** AlexeyAbashkin has quit IRC21:54
*** itlinux has quit IRC21:57
lbragstadif anyone here is client savvy and wants to give me some pointers, i'm all ears :)22:05
lbragstadhttps://review.openstack.org/#/c/574475/22:05
lbragstadhttps://review.openstack.org/#/c/574391/22:05
lbragstadhttps://review.openstack.org/#/c/537668/622:05
*** nicolasbock has quit IRC22:05
*** rcernin has joined #openstack-keystone22:16
*** r-daneel has quit IRC22:26
*** r-daneel has joined #openstack-keystone22:38
*** EmilienM|PTO is now known as EmilienM22:41
*** edmondsw has quit IRC22:47
*** r-daneel has quit IRC22:54
*** hoonetorg has quit IRC23:24
*** hoonetorg has joined #openstack-keystone23:36
*** felipemonteiro has quit IRC23:42
*** felipemonteiro has joined #openstack-keystone23:45
*** felipemonteiro has quit IRC23:55
kmalloclbragstad: more dead code removal coming soon [woo]23:56
« Sunday, 2018-06-10 Index Tuesday, 2018-06-12 »

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!