| *** sjmc7 has quit IRC | 00:11 | |
| *** jxstanford has joined #openstack-api | 00:13 | |
| *** Apoorva has quit IRC | 00:36 | |
| *** salv-orlando has joined #openstack-api | 00:48 | |
| *** salv-orlando has quit IRC | 00:55 | |
| *** Apoorva has joined #openstack-api | 01:30 | |
| *** Apoorva has quit IRC | 01:44 | |
| *** Apoorva has joined #openstack-api | 01:49 | |
| *** sjmc7 has joined #openstack-api | 02:00 | |
| *** sjmc7 has quit IRC | 02:04 | |
| *** salv-orlando has joined #openstack-api | 02:08 | |
| *** salv-orlando has quit IRC | 02:14 | |
| *** Apoorva has quit IRC | 02:19 | |
| *** annegentle has joined #openstack-api | 02:28 | |
| *** annegentle has quit IRC | 02:44 | |
| *** annegentle has joined #openstack-api | 02:51 | |
| *** annegentle has quit IRC | 02:51 | |
| *** jxstanfo_ has joined #openstack-api | 03:14 | |
| *** jxstanf__ has joined #openstack-api | 03:15 | |
| *** jxstanford has quit IRC | 03:18 | |
| *** jxstanfo_ has quit IRC | 03:19 | |
| *** jxstanf__ has quit IRC | 03:19 | |
| *** notmars has joined #openstack-api | 03:59 | |
| *** salv-orlando has joined #openstack-api | 04:10 | |
| *** notmars has quit IRC | 05:07 | |
| *** notmars has joined #openstack-api | 05:32 | |
| *** notmars has quit IRC | 05:32 | |
| *** sjmc7 has joined #openstack-api | 05:37 | |
| *** sjmc7 has quit IRC | 05:42 | |
| *** terrylhowe has quit IRC | 05:53 | |
| *** woodster_ has quit IRC | 06:00 | |
| openstackgerrit | Alex Xu proposed openstack/api-wg: Add guideline for api microversion https://review.openstack.org/187112 | 06:15 |
|---|---|---|
| *** fzdarsky has joined #openstack-api | 07:12 | |
| *** alex_klimov has joined #openstack-api | 07:19 | |
| openstackgerrit | Matthew Gilliard proposed openstack/api-wg: Adds a small update to tagging guidance https://review.openstack.org/187891 | 07:26 |
| *** sjmc7 has joined #openstack-api | 07:26 | |
| *** sjmc7 has quit IRC | 07:31 | |
| openstackgerrit | Alex Xu proposed openstack/api-wg: Add guideline for api microversion version https://review.openstack.org/187112 | 07:48 |
| openstackgerrit | Alex Xu proposed openstack/api-wg: Add guideline of Microverion bump https://review.openstack.org/187896 | 07:48 |
| openstackgerrit | Alex Xu proposed openstack/api-wg: Add guideline of api microverion bump https://review.openstack.org/187896 | 07:56 |
| openstackgerrit | Alex Xu proposed openstack/api-wg: Add guideline for api microversion specification https://review.openstack.org/187112 | 07:56 |
| *** lucasagomes has joined #openstack-api | 08:14 | |
| openstackgerrit | Alex Xu proposed openstack/api-wg: Add describe about extra query parameter https://review.openstack.org/187903 | 08:19 |
| *** e0ne has joined #openstack-api | 08:46 | |
| *** e0ne is now known as e0ne_ | 09:01 | |
| *** e0ne_ is now known as e0ne | 09:01 | |
| *** cdent has joined #openstack-api | 09:10 | |
| *** sjmc7 has joined #openstack-api | 09:15 | |
| *** sjmc7 has quit IRC | 09:20 | |
| *** e0ne is now known as e0ne_ | 09:55 | |
| *** e0ne_ is now known as e0ne | 09:57 | |
| *** sjmc7 has joined #openstack-api | 11:04 | |
| *** sjmc7 has quit IRC | 11:09 | |
| *** lucasagomes has quit IRC | 11:18 | |
| *** e0ne is now known as e0ne_ | 11:32 | |
| *** e0ne_ has quit IRC | 11:42 | |
| *** woodster_ has joined #openstack-api | 11:44 | |
| *** e0ne has joined #openstack-api | 11:58 | |
| *** lucasagomes has joined #openstack-api | 11:59 | |
| *** lucasagomes is now known as lucas-hungry | 12:05 | |
| *** flaper87 has joined #openstack-api | 12:13 | |
| *** sjmc7 has joined #openstack-api | 12:53 | |
| gilliard | cdent: I did a little facepalm when I saw your comment on https://review.openstack.org/#/c/180094/4 | 12:54 |
| cdent | :) | 12:54 |
| *** notmars has joined #openstack-api | 12:54 | |
| gilliard | While patches are sat in the approved-but-waiting-for-a-week phase, it seems a shame to make a change, though. | 12:54 |
| cdent | yeah, that's why I said only if there's another | 12:55 |
| gilliard | best to make a followup patch? | 12:55 |
| sdague | gilliard: yeh, why not just put a tidy up patch on top | 12:55 |
| cdent | yeah, as there's plenty of other "call" throughout the files | 12:55 |
| gilliard | It's a hard habit to break :( | 12:55 |
| cdent | tru | 12:55 |
| *** sjmc7 has quit IRC | 12:58 | |
| openstackgerrit | Matthew Gilliard proposed openstack/api-wg: s/call/request/ :: This isn't RPC https://review.openstack.org/187991 | 12:58 |
| *** lucas-hungry is now known as lucasagomes | 13:08 | |
| *** terrylhowe has joined #openstack-api | 13:15 | |
| *** e0ne is now known as e0ne_ | 13:41 | |
| *** e0ne_ is now known as e0ne | 13:41 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 13:54 | |
| *** sjmc7 has joined #openstack-api | 14:09 | |
| openstackgerrit | Ryan Brown proposed openstack/api-wg: Add section clarifying PUT vs PATCH semantics https://review.openstack.org/183945 | 14:35 |
| *** ameade has joined #openstack-api | 14:46 | |
| *** nikhil_k is now known as nikhil_k-brb | 14:52 | |
| openstackgerrit | Ryan Brown proposed openstack/api-wg: Add section clarifying PUT vs PATCH semantics https://review.openstack.org/183945 | 14:56 |
| cdent | ryansb: your response to my comment about custom headers implies that the api working group will be here forever like an immortal parent forever watching over the children creating their beautful creative works forever and ever amen | 14:57 |
| cdent | that would be a shame, the wg should have a natural finish | 14:57 |
| ryansb | cdent: ...rereading it that is true, I did imply that | 14:57 |
| ryansb | I guess I mean that a guideline of "don't add a header unless you really mean it" would be unlikely to solve the problem | 14:58 |
| cdent | yeah, that's definitely true | 14:58 |
| ryansb | I think it'd be better to figure out why people add headers we would say "aren't necessary" | 14:58 |
| ryansb | and add a guideline that says "if you have problem X or goal Y, do Z instead of mashing in MOAR X-OpenStack-HEADERS" | 14:59 |
| * cdent nods | 14:59 | |
| elmiko | cdent: agreed about natural finish, i can envision a maintenance mode at some point. | 14:59 |
| ryansb | so my question to you is; what are cases where people have added headers mistakenly | 15:00 |
| cdent | I'm insufficiently conversant with all the openstack apis thus far to really say, but my general opinion is: don't add headers | 15:01 |
| ryansb | I see. Well the biggest header user I know of is Swif | 15:01 |
| ryansb | *Swift | 15:01 |
| ryansb | trailed by Nova, I believe | 15:01 |
| cdent | yes, swift does all kinds of things in what I would consider completely broken ways | 15:02 |
| cdent | but it represents the reality that we live in | 15:02 |
| ryansb | swift has a reason though, since they basically ripped the S3 way of doing things | 15:02 |
| ryansb | (if that's a reason) | 15:02 |
| cdent | s3 pretty broken too :) | 15:03 |
| * cdent doesn't live in the real world when noodling about platonic ideals | 15:03 | |
| ryansb | cdent: whoa, you're about to turn us into the "AWS API Peanut Gallery" | 15:03 |
| cdent | I can come back to reality when required though | 15:03 |
| *** alex_klimov has quit IRC | 15:23 | |
| openstackgerrit | Chris Dent proposed openstack/api-wg: Add advice on when to use POST or PUT in create https://review.openstack.org/181912 | 15:26 |
| *** e0ne is now known as e0ne_ | 15:32 | |
| *** notmars has quit IRC | 15:32 | |
| *** e0ne_ is now known as e0ne | 15:35 | |
| *** notmars has joined #openstack-api | 15:38 | |
| *** notmars has quit IRC | 15:43 | |
| *** nikhil_k-brb is now known as nikhil_k | 15:50 | |
| *** notmars has joined #openstack-api | 15:55 | |
| *** fzdarsky has quit IRC | 16:22 | |
| *** Apoorva has joined #openstack-api | 16:36 | |
| *** lucasagomes is now known as lucas-dinner | 17:04 | |
| *** e0ne has quit IRC | 17:17 | |
| *** notmars has quit IRC | 17:20 | |
| *** fzdarsky has joined #openstack-api | 17:21 | |
| *** fzdarsky has quit IRC | 17:38 | |
| *** sjmc7 has quit IRC | 17:47 | |
| *** notmars has joined #openstack-api | 17:50 | |
| *** sjmc7 has joined #openstack-api | 17:51 | |
| *** annegentle has joined #openstack-api | 18:23 | |
| *** notmars has quit IRC | 18:29 | |
| *** dolphm has joined #openstack-api | 18:35 | |
| *** notmars has joined #openstack-api | 18:36 | |
| *** notmars has quit IRC | 18:39 | |
| *** notmars has joined #openstack-api | 18:40 | |
| *** notmars has quit IRC | 18:40 | |
| *** flaper87 has quit IRC | 18:44 | |
| *** cdent has quit IRC | 18:50 | |
| *** flaper87 has joined #openstack-api | 19:08 | |
| *** flaper87 has quit IRC | 19:09 | |
| *** flaper87 has joined #openstack-api | 19:09 | |
| *** cdent has joined #openstack-api | 19:09 | |
| *** e0ne has joined #openstack-api | 19:13 | |
| openstackgerrit | Chris Dent proposed openstack/api-wg: Add advice on when to use POST or PUT in create https://review.openstack.org/181912 | 19:17 |
| elmiko | cdent: are you ammenable to pr's on github for gabbi? | 19:19 |
| cdent | yes, for now that's the way to go | 19:19 |
| elmiko | k, i have a small doc-based suggestion | 19:20 |
| cdent | once I finish some additional test things and have decided that it is really 1.0 i'm going to move it under the qa program and make it all gerrit-ified etc | 19:20 |
| cdent | awesome! | 19:20 |
| elmiko | cool, i'll send it soon™ | 19:20 |
| elmiko | cdent: ok, sent your way | 19:30 |
| * cdent looks | 19:30 | |
| cdent | man, why's everybody always hatin on make | 19:33 |
| cdent | :) | 19:33 |
| elmiko | hehe, long history of abuse ;) | 19:34 |
| cdent | you want my comments/questions here or there? | 19:35 |
| elmiko | besides, it's not like i deleted the Makefile in that patch | 19:35 |
| elmiko | here is fine | 19:35 |
| cdent | they're basically "howcumzit ya did it like X" | 19:35 |
| cdent | a) what's the whitelist thing doing | 19:35 |
| ryansb | cdent: dontcha know, gotta switch to Gulp. It's like make, but in node.js so it's a zillion times better | 19:35 |
| elmiko | whitelist should reduce warnings for using commands outside of the tox virtualenv | 19:35 |
| ryansb | ;) | 19:35 |
| cdent | b) is the doc/build removal requirered or just tidy | 19:35 |
| elmiko | ryansb: LOL | 19:35 |
| elmiko | cdent: just wanted to be tidy | 19:36 |
| * cdent still can't manage to grunt | 19:36 | |
| elmiko | cdent: so, the tox env will create the doctrees/html/man for gabbi, i think the only worry i have about removing the whole build dir is if someone used the makefile to create a pdf. that might get stomped. | 19:38 |
| elmiko | i could make the rm command a little more specific if it helps? | 19:39 |
| cdent | naw, it's fine as is | 19:39 |
| cdent | my expectation is that the Makefile will eventually go away | 19:39 |
| cdent | it is there to facillitate me and my style and once the code is more public there's not much point | 19:39 |
| elmiko | makes sense, and i figured as much. i'm just digging deep into the code so i was curious about little openstack-esque type changes | 19:40 |
| cdent | my own personal style of making shit is not much in concert with the tox+pbr+god knows what else style in openstack-land | 19:40 |
| cdent | but it is the way of things, and it seems to work | 19:40 |
| elmiko | i didn't use tox before i started working on openstack, but i can see why people might like it. the virtualenv feature is nice for not messing with the system packages. | 19:42 |
| elmiko | like, when i ran the makefile i realized i don't have sphinx installed. lol | 19:43 |
| cdent | yeah, that's part of why I (historically) don't use tox because I intentionally want to make sure that I create chaos in my installed system packages | 19:44 |
| cdent | I also never cap versions | 19:44 |
| elmiko | hehe | 19:44 |
| cdent | and I'm not a fan of using requirements.txt to mean install_requires (it's not the same thing) | 19:44 |
| ryansb | anarchy, sir | 19:45 |
| elmiko | lol! | 19:45 |
| cdent | fecundity more like it: find more bugs sooner | 19:45 |
| ryansb | though reqs-as-install_requires is more a pbr-ism than a tox-ism | 19:45 |
| ryansb | (AIUI) | 19:45 |
| elmiko | i've gotten in the habit of trying to keep my root system very clean and relying more and more on virtualenvs, vms, and containers | 19:45 |
| ryansb | same. It never works though | 19:46 |
| cdent | yes, that's true, I expanded above with "tox+pbr+god knows what else" | 19:46 |
| ryansb | lol | 19:46 |
| elmiko | yea, makes perfect sense | 19:46 |
| ryansb | tox for multiple interpreter versions is a godsend though. I use it in my personal projects now | 19:46 |
| cdent | part of the issue is that recreating venvs, vms and containers is so freaking slow | 19:46 |
| cdent | yeah, the multiple interpreter aspect is what has converted me | 19:47 |
| ryansb | I have a local pip cache, and with pip7 I can reuse wheels | 19:47 |
| ryansb | so it's not too slow these days | 19:47 |
| ryansb | (vms are still slow) | 19:47 |
| cdent | in my experience if I have a problem with a python package, if I reinstall with --no-use-wheel it fixes it | 19:47 |
| ryansb | (containers are sorta slow) | 19:47 |
| cdent | so I have not been convinced about wheels | 19:47 |
| elmiko | havent' tried wheels | 19:47 |
| ryansb | works on my machine, YMMV | 19:47 |
| cdent | :) | 19:48 |
| cdent | experience suggests that my tolerance for slow is not in the normal part of the bell curve | 19:50 |
| elmiko | entirely fair | 19:50 |
| ryansb | yup. I just happen to like making coffee ;) | 19:50 |
| elmiko | nice, i could go for a beverage. brb | 19:51 |
| cdent | when I first showed up in openstack land and ran some tests locally I thought | 19:51 |
| cdent | huh, and this is considered okay? | 19:51 |
| elmiko | LOL | 19:51 |
| ryansb | yeeeaaah. | 19:51 |
| cdent | so yeah, while I run some tests, I'm going to go for a walk :) | 19:52 |
| cdent | bbl | 19:52 |
| ryansb | heh, enjoy | 19:52 |
| HenryG | elmiko: sorry I missed you in the neutron channel, but I read the backscroll. I'll also review the process guidelines update. | 19:56 |
| elmiko | HenryG: awesome, thanks for checking them out | 19:57 |
| *** flaper87 has quit IRC | 19:57 | |
| *** flaper87 has joined #openstack-api | 19:59 | |
| *** flaper87 has quit IRC | 20:03 | |
| *** salv-orlando has quit IRC | 20:09 | |
| *** salv-orlando has joined #openstack-api | 20:09 | |
| *** e0ne is now known as e0ne_ | 20:09 | |
| *** e0ne_ is now known as e0ne | 20:14 | |
| *** e0ne has quit IRC | 20:14 | |
| *** lucas-dinner has quit IRC | 20:15 | |
| *** flaper87 has joined #openstack-api | 20:16 | |
| elmiko | cdent: just wanted to give some kudos on the clear code in gabbi, i'm finding it relatively easy to get around and understand what's going on. | 20:32 |
| cdent | elmiko: thanks, I'm glad you're finding it readable | 20:37 |
| elmiko | definitely | 20:37 |
| cdent | I don't really think of it as up to snuff because I had to a lot of dirty nasty stuff to get it to behave in a way that makes the keepers of subunit and testr happy | 20:38 |
| elmiko | ah, ok. maybe i haven't tripped over those yet ;) | 20:38 |
| elmiko | i'm kinda doing some investigation about tools the security group might be able to use for fuzz testing apis | 20:39 |
| elmiko | not sure that gabbi is the best fit, but i'm still curious about using it | 20:39 |
| elmiko | (for other stuff) | 20:39 |
| cdent | i thikn for fuzz testing you'd either need to pre-process some yaml (from templates or whatever) or generate the tests through some other process (the reading yaml and translating into tests is just a small piece of the piece and not a required piece) | 20:41 |
| elmiko | that's what i'm thinking too | 20:41 |
| cdent | i'd be very happy to talk more about that or whatever else either in irc or email or whatever | 20:41 |
| cdent | but now I'm going to change puters, back in a while | 20:41 |
| elmiko | k, talk more later | 20:41 |
| *** Apoorva has quit IRC | 21:04 | |
| *** cdent_ has joined #openstack-api | 21:05 | |
| *** annegentle has quit IRC | 21:06 | |
| *** cdent has quit IRC | 21:06 | |
| *** cdent_ is now known as cdent | 21:06 | |
| cdent | you still around elmiko? | 21:12 |
| elmiko | yup | 21:13 |
| cdent | I was just curious what fuzz testing would actually look like, or rather what the mechanism is (in the conceptual sense). | 21:15 |
| elmiko | so, i have a couple thoughts about it. fuzzing the urls and fuzzing the input json | 21:16 |
| elmiko | in the gabbi sense, i think it might be interesting to have maybe a pluggable interface for creating test suites. | 21:16 |
| elmiko | so, by default yaml would work | 21:17 |
| elmiko | but | 21:17 |
| elmiko | maybe there could be a mode where a user could supply a small piece of python or something to describe an iterative test | 21:17 |
| elmiko | alternatively, | 21:17 |
| elmiko | maybe i could supply several urls that all line up with 1 expected output | 21:17 |
| elmiko | so i wouldn't need to make huge yaml files | 21:18 |
| elmiko | that's just url fuzzing, for input fuzzing it might be more complicated | 21:18 |
| cdent | have you looked at gabbi-runner: it can take yaml on stdin, so a program could produce it dynamically | 21:18 |
| elmiko | but having the ability to create a small piece of python that could be injested to create the test might be interesting | 21:18 |
| cdent | yes, that’s true | 21:18 |
| elmiko | interesting... i haven't looked at gabbi-runner yet | 21:19 |
| elmiko | i've been reading the test suite generator stuff from yaml to test case, so far | 21:19 |
| cdent | pluggable test case generation is on the maybe-roadmap | 21:19 |
| elmiko | cool | 21:19 |
| cdent | so if you come up with something workable that would be awesome | 21:20 |
| elmiko | would probably require some re-architecting, but could certainly work | 21:20 |
| cdent | or if I get around to it... | 21:20 |
| elmiko | i'll keep poking around, i also need to talk with the security group again to get more input on how/what we want to fuzz test | 21:20 |
| elmiko | so, one simple thought that occurred to me was to change gabbi.driver so that it could also discover .py files and then bring the code from that py file in to create a test | 21:21 |
| elmiko | but this is a very half baked idea | 21:21 |
| *** alex_klimov has joined #openstack-api | 21:22 | |
| elmiko | i'm more just hacking about at this point =) | 21:22 |
| *** sjmc7 has left #openstack-api | 21:22 | |
| cdent | what about passing a generator class to build_tests that defaults to the yaml reader? | 21:22 |
| elmiko | yea, exactly | 21:22 |
| elmiko | that would make nice room for a possible plugin at some point | 21:22 |
| elmiko | i imagine i'll have more ideas tomorrow after the sec meeting | 21:26 |
| elmiko | but, the generator i could see as being *very* useful | 21:26 |
| * cdent sets his back brain to think about that | 21:27 | |
| elmiko | nice =) | 21:29 |
| elmiko | i'm out for a few hours | 21:33 |
| *** Apoorva has joined #openstack-api | 21:45 | |
| *** Apoorva has quit IRC | 21:46 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:30 | |
| *** cdent has quit IRC | 22:40 | |
| *** Apoorva has joined #openstack-api | 22:47 | |
| *** Apoorva has quit IRC | 22:51 | |
| *** Apoorva has joined #openstack-api | 22:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!