| *** openstack has joined #openstack-marconi | 13:37 | |
| *** rongze has quit IRC | 13:37 | |
| *** cpallares has joined #openstack-marconi | 13:38 | |
| *** jmckind has joined #openstack-marconi | 13:52 | |
| *** oz_akan_ has joined #openstack-marconi | 13:58 | |
| *** rongze has joined #openstack-marconi | 14:32 | |
| *** rongze has quit IRC | 14:37 | |
| *** malini_afk is now known as malini | 14:50 | |
| *** oz_akan_ has quit IRC | 14:51 | |
| *** rongze has joined #openstack-marconi | 14:51 | |
| *** oz_akan_ has joined #openstack-marconi | 14:51 | |
| *** chandankumar has quit IRC | 14:52 | |
| *** balajiiyer has joined #openstack-marconi | 14:55 | |
| balajiiyer | Good morning | 14:55 |
|---|---|---|
| mpanetta | mornin | 14:55 |
| malini | Good Morning! | 14:57 |
| flaper87 | morning | 14:57 |
| flaper87 | morning | 14:57 |
| flaper87 | morning | 14:57 |
| flaper87 | morning | 14:57 |
| flaper87 | morning | 14:58 |
| flaper87 | morning | 14:58 |
| flaper87 | morning | 14:58 |
| flaper87 | morning | 14:58 |
| * flaper87 just had tons of gummy bears | 14:58 | |
| * flaper87 is happy | 14:58 | |
| balajiiyer | flaper87: too much coffee? | 14:58 |
| flaper87 | coffee is never enough | 15:00 |
| flaper87 | not for me | 15:00 |
| flaper87 | :D | 15:00 |
| *** liang has quit IRC | 15:11 | |
| *** rwsu has joined #openstack-marconi | 15:16 | |
| *** moijes12 has joined #openstack-marconi | 15:17 | |
| *** muralia has joined #openstack-marconi | 15:41 | |
| *** fifieldt has quit IRC | 16:04 | |
| *** alcabrera has joined #openstack-marconi | 16:07 | |
| alcabrera | Good morning! :D | 16:07 |
| *** ayoung_ has joined #openstack-marconi | 16:32 | |
| cpallares | good morning alcabrera! | 16:38 |
| alcabrera | cpallares: hey! How are you? :) | 16:39 |
| cpallares | alcabrera: I had a rough weekend. My hard drive crashed and I didn't back it up :( How are you doing? How was your weekend? | 16:40 |
| balajiiyer | cpallares: Sorry to hear that. | 16:40 |
| alcabrera | Yikes, sorry to hear it. Restoring a system is quite the time sink. :x | 16:41 |
| alcabrera | I'm doing great today. The weekend was full of reading and writing, so I feel all energized today. | 16:41 |
| *** ayoung_ is now known as ayoung | 16:43 | |
| cpallares | alcabrera: Yeah, it was too time consuming and that sounds like a good weekend, any good reads? | 16:43 |
| cpallares | hello balajiiyer, how are you? | 16:43 |
| alcabrera | yep! I finished reading Real World Haskell (again). It was much easier to absorb this time around. | 16:44 |
| alcabrera | cpallares: ^ | 16:44 |
| balajiiyer | cpallares: doing good today | 16:45 |
| alcabrera | be back later. Out for lunch. :) | 16:53 |
| *** alcabrera is now known as alcabrera|afk | 16:53 | |
| flaper87 | damn, what a day................ | 17:03 |
| flaper87 | :D | 17:04 |
| * flaper87 is still happy | 17:04 | |
| flaper87 | :P | 17:04 |
| *** ykaplan has quit IRC | 17:04 | |
| *** jcru has joined #openstack-marconi | 17:07 | |
| *** kgriffs_afk is now known as kgriffs | 17:17 | |
| flaper87 | alcabrera|afk: kgriffs could you guys help me debug this? https://review.openstack.org/#/c/65686/ | 17:22 |
| flaper87 | I mean, are you seeing the same failure ? | 17:22 |
| flaper87 | kgriffs: https://review.openstack.org/#/c/65708/ <- this one is really simple to review | 17:22 |
| flaper87 | :D | 17:22 |
| *** balajiiyer has quit IRC | 17:22 | |
| *** reed has joined #openstack-marconi | 17:28 | |
| moijes12 | which is the room for openstack-tempest ? | 17:35 |
| flaper87 | moijes12: try #openstack-tempest or #tempest | 17:36 |
| flaper87 | not sure which one | 17:36 |
| malini | moijes12: its #openstack-qa | 17:44 |
| malini | I dont think there is a separate channel for tempest | 17:44 |
| moijes12 | malini: ok. thanks | 17:44 |
| malini | moijes12: yw | 17:45 |
| flaper87 | malini: see, that's why you rock! | 17:47 |
| *** moijes12 has left #openstack-marconi | 17:48 | |
| *** balajiiyer has joined #openstack-marconi | 17:53 | |
| malini | flaper87: :D | 17:57 |
| *** balajiiyer has quit IRC | 17:57 | |
| kgriffs | flaper87: I am running tox on your testr patch now | 17:59 |
| kgriffs | I'll let you know what I find out | 17:59 |
| kgriffs | going to lunch now... | 17:59 |
| *** rongze has quit IRC | 18:05 | |
| *** jamieh has quit IRC | 18:06 | |
| *** alper_halbut has joined #openstack-marconi | 18:10 | |
| *** whenry has joined #openstack-marconi | 18:17 | |
| *** muralia has quit IRC | 18:22 | |
| *** alcabrera|afk is now known as alcabrera | 18:34 | |
| *** balajiiyer has joined #openstack-marconi | 18:40 | |
| *** alcabrera is now known as alcabrera|afk | 18:40 | |
| *** alper_halbut has quit IRC | 18:45 | |
| *** alper_halbut has joined #openstack-marconi | 18:52 | |
| *** jmckind has quit IRC | 18:53 | |
| *** jcru has quit IRC | 18:59 | |
| *** yassine has quit IRC | 19:08 | |
| *** yassine has joined #openstack-marconi | 19:08 | |
| *** yassine has quit IRC | 19:09 | |
| *** yassine has joined #openstack-marconi | 19:09 | |
| *** yassine has quit IRC | 19:10 | |
| *** yassine has joined #openstack-marconi | 19:12 | |
| *** yassine has quit IRC | 19:12 | |
| *** yassine has joined #openstack-marconi | 19:12 | |
| *** yassine has quit IRC | 19:13 | |
| *** jcru has joined #openstack-marconi | 19:17 | |
| *** kgriffs is now known as kgriffs_afk | 19:19 | |
| *** muralia has joined #openstack-marconi | 19:30 | |
| *** openstackgerrit has quit IRC | 19:31 | |
| *** openstackgerrit has joined #openstack-marconi | 19:31 | |
| *** whenry has quit IRC | 19:36 | |
| *** kgriffs_afk is now known as kgriffs | 19:36 | |
| *** alcabrera|afk is now known as alcabrera | 19:48 | |
| alcabrera | flaper87: I'll try to debug that. It could be a pypy version thing. I'll be running mine against pypy-2.2.1 | 19:51 |
| *** whenry has joined #openstack-marconi | 19:54 | |
| kgriffs | i'm having trouble getting virtualenv and pypy to play nice on my box lately | 20:08 |
| kgriffs | grrrr | 20:08 |
| kgriffs | once I fix that, I can help as well | 20:08 |
| *** jmckind has joined #openstack-marconi | 20:09 | |
| alcabrera | It's Babel failing to download/install for some reason. | 20:10 |
| flaper87 | alcabrera: thanks | 20:10 |
| flaper87 | I'm doing a test myself too | 20:10 |
| flaper87 | I got everything set up again | 20:10 |
| flaper87 | since my f19->f20 migration, I hadn't used pypy | 20:11 |
| alcabrera | I can't even get Babel to install using pip outside of tox. :/ | 20:12 |
| alcabrera | With pypy, that is. | 20:12 |
| alcabrera | pip==1.5 | 20:12 |
| kgriffs | ok, so I reinstalled latest version of pypy and virtualenv | 20:13 |
| kgriffs | now I am actually getting into tests | 20:13 |
| kgriffs | https://gist.github.com/anonymous/db14fa836b580dfba1cc | 20:13 |
| kgriffs | alcabrera, flaper87: ^^^ | 20:13 |
| kgriffs | that's my output from tox -e pypy | 20:14 |
| kgriffs | (if you couldn't tell) | 20:14 |
| alcabrera | that's better than mine. | 20:15 |
| alcabrera | My pypy fails to install any new package. | 20:15 |
| alcabrera | ConnectionError: HTTPSConnectionPool(host='pypi.python.org', port=443): Max retries exceeded with url: /packages/2.7/r/requests/requests-2.2.0-py2.py3-none-any.whl (Caused by <class '_socket.error'>: [Errno 9] Bad file descriptor) | 20:15 |
| alcabrera | :D | 20:15 |
| alcabrera | I likely need to patch that pypy-pip. | 20:15 |
| alcabrera | There, I fixed my silly issue. | 20:19 |
| alcabrera | I was using pypy-2.1 for pip installs, even though pypy-2.2.1 was first on my path. | 20:19 |
| alcabrera | retoxing~ | 20:19 |
| alcabrera | kgriffs: that error makes me think that pypy internally already differentiates between unicode and str, even in the 2.x series. | 20:21 |
| alcabrera | I got the same error after re-tox-ing. | 20:21 |
| alcabrera | brb | 20:24 |
| *** alcabrera is now known as alcabrera|afk | 20:24 | |
| *** malini is now known as malini_afk | 20:28 | |
| *** alcabrera|afk is now known as alcabrera | 20:40 | |
| alcabrera | Getting close - the tests pass and now I'm running into 'No module named gdbm' | 20:42 |
| alcabrera | kgriffs, flaper87: ^ | 20:42 |
| alcabrera | I just used six.text_type conversions to make the tests pass. | 20:42 |
| *** jergerber has joined #openstack-marconi | 20:44 | |
| *** jergerber has quit IRC | 20:44 | |
| flaper87 | mmh, wait | 20:45 |
| flaper87 | did you get the same error in that review? | 20:45 |
| kgriffs | figured it out | 20:45 |
| kgriffs | no, I did not get the same error | 20:45 |
| flaper87 | or it's just a different error that you're suddenly getting | 20:45 |
| flaper87 | oh, ok | 20:45 |
| flaper87 | I wonder what introduced that error | 20:45 |
| kgriffs | I think the jenkins error may be a symptom of the real error | 20:46 |
| kgriffs | flaper87: I figured it out | 20:46 |
| flaper87 | kgriffs: shoot | 20:46 |
| kgriffs | so, those tests relied on a side effect of json.dumps | 20:46 |
| kgriffs | in py27 if the object being serialized contained any unicode strings, then the result would be unicode | 20:46 |
| kgriffs | (with ensure_ascii=False) | 20:46 |
| kgriffs | however | 20:46 |
| kgriffs | in pypy this is no longer the case | 20:47 |
| kgriffs | it seems that the result with be a str if no chars had to be converted | 20:47 |
| kgriffs | s/with/will | 20:47 |
| alcabrera | cool, gtk. | 20:47 |
| kgriffs | I just verified with ipython, fwiw | 20:47 |
| kgriffs | so, if I cast the result of json.dumps then those errors go away | 20:48 |
| kgriffs | cast to six.text_type | 20:48 |
| kgriffs | s/cast/coerce :p | 20:48 |
| * kgriffs still thinks in terms of C on occasion | 20:48 | |
| kgriffs | so... | 20:49 |
| kgriffs | is there something besides io.StringIO we can use, or should we just convert or what? | 20:49 |
| kgriffs | thing is, if we use bytesio or something, then we have to normalize to bytearray/str | 20:50 |
| kgriffs | so... | 20:50 |
| kgriffs | maybe we have a helper json.dumps | 20:50 |
| kgriffs | that always returns six.text_type? | 20:50 |
| alcabrera | Hmm, bytes might be the way to go. | 20:51 |
| alcabrera | Considering that if we're communicating over the network, we should be sending bytes anyway. | 20:51 |
| flaper87 | mmh, I'd like to keep everything in undicode within marconi | 20:51 |
| flaper87 | to avoid weird erros, not necessary encoding process etc | 20:52 |
| kgriffs | so, filter_stream uses read_json(stream, len) | 20:52 |
| kgriffs | which does this: | 20:52 |
| kgriffs | return json.loads(stream.read(len), parse_int=_json_int) | 20:52 |
| kgriffs | seems that we are assuming a byte stream given by the WSGI server | 20:53 |
| alcabrera | hmmm | 20:54 |
| kgriffs | I'm pretty sure WSGI apps are supposed to assume binary because the body may not be text, after all | 20:55 |
| kgriffs | so, if we are trying to simulate a WSGI input stream (wsgi.input) we should be constructing a binary stream | 20:56 |
| kgriffs | "An input stream (file-like object) from which the HTTP request body bytes can be read" | 20:56 |
| kgriffs | (from PEP 3333) | 20:56 |
| alcabrera | as per pep 3333, we should be importing/exporting binary at the boundary. | 20:57 |
| alcabrera | Internally, it's a little trickier. Exception construction and/or 4xx/5xx level returns are the trickiest part, IMO. | 20:58 |
| flaper87 | phew https://github.com/FlaPer87/rust/commit/5c3acf197a64ee7b7ff49ce74c24559d1856ccb6 | 20:58 |
| flaper87 | mmh, interesting | 20:59 |
| flaper87 | IIRC, there's a way to get the unicode body out of the request | 21:00 |
| flaper87 | ah no, that's for responses | 21:00 |
| flaper87 | not requests | 21:00 |
| alcabrera | flaper87: fancy - llvm intrinsics. :D | 21:01 |
| kgriffs | A true test will be using a byte stream, not a text stream | 21:01 |
| kgriffs | true as in - closest to how that filter stream method is used in production with a real WSGI box | 21:02 |
| kgriffs | so, I propose a helper testing util | 21:02 |
| kgriffs | that serializes a dict or array to io.BytesIO | 21:03 |
| kgriffs | or something along those lines, anyway | 21:03 |
| * kgriffs sits down | 21:03 | |
| alcabrera | hmmm | 21:05 |
| alcabrera | sounds reasonable to me. I'm +1 for bytes-based testing. | 21:06 |
| flaper87 | kgriffs: thanks for digging into that. I had a busy crazy day today and I didn't have time | 21:09 |
| kgriffs | kk | 21:09 |
| kgriffs | well, that should give you some ideas for where you want to go next | 21:10 |
| flaper87 | but it's a scary bug, that sound's like will block all our future patches | 21:10 |
| kgriffs | agreed | 21:10 |
| alcabrera | I'm out for the day. We | 21:10 |
| alcabrera | Let's clear out that bug asap. | 21:11 |
| alcabrera | :D | 21:11 |
| alcabrera | Cheers, everyone. Have a good night! | 21:11 |
| *** alcabrera has quit IRC | 21:11 | |
| *** ayoung is now known as ayoung_dadmode | 22:11 | |
| *** whenry has quit IRC | 22:15 | |
| *** cpallares has quit IRC | 22:20 | |
| *** balajiiyer has quit IRC | 22:32 | |
| *** jmckind has quit IRC | 22:40 | |
| *** thomasem has quit IRC | 22:41 | |
| *** rwsu has quit IRC | 22:52 | |
| *** oz_akan_ has quit IRC | 22:59 | |
| *** tedross has quit IRC | 23:00 | |
| *** flwang has quit IRC | 23:03 | |
| *** mpanetta has quit IRC | 23:06 | |
| *** rwsu has joined #openstack-marconi | 23:08 | |
| *** kgriffs is now known as kgriffs_afk | 23:08 | |
| *** jmckind has joined #openstack-marconi | 23:14 | |
| *** jmckind has quit IRC | 23:14 | |
| *** openstackgerrit has quit IRC | 23:24 | |
| *** openstackgerrit has joined #openstack-marconi | 23:24 | |
| *** flaper87 is now known as flaper87|afk | 23:25 | |
| *** flwang has joined #openstack-marconi | 23:39 | |
| *** muralia has quit IRC | 23:40 | |
| *** balajiiyer has joined #openstack-marconi | 23:48 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!