Earlier  
Later
New Search
Posted Nick Remark
#openstack-sdks - 2018-03-20
16:52:05 kmalloc so, you're good with saying it's not part of the API?
16:52:17 mordred kmalloc: so maybe let's say "no" - and if people are finding places in which they need to look at exception text, then we nee more specific exception classes
16:52:23 kmalloc wfm
16:52:25 kmalloc ^5
16:52:34 mordred frickler: ^^
16:52:46 kmalloc exception text is useful for humans
16:52:51 kmalloc not for programatic use.
16:52:56 openstackgerrit Merged openstack/python-openstacksdk master: Fix 'block_store' aliases define error https://review.openstack.org/554403
16:53:48 mordred frickler: also - fwiw, maybe toss in an 'if isinstance(e, exceptions.DiscoveryFailure): raise else: raise exceptions.DiscoveryFailure(...) so that we don't lose the original discovery context just because we made an attempt to guess a versioned endpoint
17:28:16 openstackgerrit Chris Dent proposed openstack/microversion-parse master: Improve Version.matches() method https://review.openstack.org/550251
17:28:17 openstackgerrit Chris Dent proposed openstack/microversion-parse master: Add MicroversionMiddleware https://review.openstack.org/550252
17:28:17 openstackgerrit Chris Dent proposed openstack/microversion-parse master: Allow passing a json_error_formatter to the middleware https://review.openstack.org/550263
17:28:18 openstackgerrit Chris Dent proposed openstack/microversion-parse master: Clean up package version and python version support https://review.openstack.org/550264
17:28:18 openstackgerrit Chris Dent proposed openstack/microversion-parse master: Do not use redundant attr check for keys in fold_headers https://review.openstack.org/554547
20:07:17 openstackgerrit Merged openstack/python-openstacksdk master: Use defined version instead of service.version parameter https://review.openstack.org/554223
21:06:11 openstackgerrit Dean Troyer proposed openstack/python-openstackclient master: Fix additional output encoding issues https://review.openstack.org/554698
21:55:13 dhellmann dtroyer : regarding https://review.openstack.org/#/c/554698/1 I wonder if cliff should grow support for progress bars?
21:56:23 dhellmann https://pypi.python.org/pypi/progressbar was pretty easy to use in downpour
21:57:03 dhellmann oh, although I wonder if that's actively supported any more if it's on code.google.com
21:57:16 dhellmann ah, redirects to https://github.com/niltonvolpato/python-progressbar
21:59:18 dhellmann aha, I actually used https://pypi.python.org/pypi/progressbar2
21:59:56 dhellmann anyway, let me know if you think that would be a useful thing to be able to do via cliff
22:00:11 dtroyer dhellmann: maybe? I really didn't know what to do with that _show_progress() function. Putting it where I did made calling easy to get cliff's stdout but made testing harder than I wanted to finesse to get a ppinter to the nested function.
22:00:33 dhellmann yeah
22:01:08 dhellmann oh, you could have made it a method instead of a closure
22:01:24 dhellmann that would make it easier to mock it out or whatever
22:01:39 dtroyer but then it would require passing self around and changing the signature of the callback
22:01:53 dhellmann self is passed automatically to a bound method
22:02:14 dhellmann if the signature is "def _show_progress(self, progress)" and you pass self._show_progress the callback will expect 1 arg
22:02:21 dtroyer ok, so I didn't even try that because I assumed it needed a second argument.
22:02:30 dhellmann nah, it does the easy thing
22:02:36 dtroyer that is certainly a better solution
22:02:40 dhellmann what sorts of values are passed to that method?
22:02:59 dhellmann are they text like "working" or are they percentages?
22:05:02 dtroyer it is anything… if the resource being waited on has a progrss attribute, that is passed back
22:05:13 dtroyer *progress attribute
22:05:14 dhellmann so a progress bar might not really work
22:06:41 dhellmann oh, there's an "animated marker"
22:07:56 dhellmann eh, it's something to think about
22:08:58 dtroyer it isn't something I really thing we should be doing, as that kind of stuff often just pollutes logs, but some like it as a touchy-feedbacky thing, even if it is usually not very accurate
22:09:32 dhellmann that's true, i guess
22:09:50 dtroyer it could be tied to logging levels to address that, but I've never really cared enough to follow up
22:10:00 dhellmann I guess it's hard to tell if self.stdout is sys.stdout after we wrap it up
22:10:15 dhellmann it could be its own flag --progress or something
22:10:42 dhellmann I don't feel strongly enough about it to push and implement it; it just seemed like a thing to talk about in light of that patch
22:11:07 dhellmann I'll leave a wishlist ticket in lp in case someone else wants to do it
22:11:34 dtroyer sure, and I appreciate the reviews, this (encoding I/O) is an area I don't seem to be able to keep in my head and have to periodically e-learn it
22:12:20 dhellmann it's like a sandwich. the encoding/decoding steps are the bread and the application is the peanut butter
22:13:00 dtroyer :) it much be dinnertime in GA?
22:13:47 dhellmann https://bugs.launchpad.net/python-cliff/+bug/1757278
22:13:47 openstack Launchpad bug 1757278 in cliff "add support for progress bar/widgets" [Wishlist,New]
22:14:02 dhellmann haha, yeah, I think I'm hungry :-)
22:14:42 dhellmann speaking of eating, it' time to go exercise some
23:44:01 mordred dtroyer, dhellmann: fwiw, ianw has poked a little bit at the idea of having some of the calls in shade (so now obviously for sdk) to be able to have progress callbacks...
23:44:43 mordred it's a little different than what you're discussing - this is more of a "do something in the requests/adapter layer so that as we upload 12G images we can tell that it's not just stuck"
23:45:15 mordred but - I could see the existence of a progress bar facility and the existence of an image upload/download progress callback being complimentary
#openstack-sdks - 2018-03-21
01:44:30 openstackgerrit Matt Riedemann proposed openstack/python-openstackclient master: Default --nic to 'auto' if creating a server with >= 2.37 https://review.openstack.org/545946
09:21:25 openstackgerrit Jens Harbott (frickler) proposed openstack/keystoneauth master: Be more helpful when version discovery fails https://review.openstack.org/554044
10:10:22 frickler could someone take a look at this bug and tell me where it should be fixed, osc-lib or sdk? https://bugs.launchpad.net/python-openstackclient/+bug/1757391
10:10:24 openstack Launchpad bug 1757391 in python-openstackclient "Client raises AttributeError: 'module' object has no attribute 'OpenStackConfigException'" [Undecided,New]
12:23:36 openstackgerrit Jens Harbott (frickler) proposed openstack/python-openstackclient master: Display private flavors in server list https://review.openstack.org/540633
12:27:15 mnaser frickler: mordred proposed a fix for that already, let me find it
12:27:57 mnaser frickler: https://review.openstack.org/#/c/554615/
12:42:21 frickler mnaser: thanks for the pointer, added comments to crosslink those
12:48:18 mnaser np :)
12:48:34 openstackgerrit Merged openstack/service-types-authority master: Add qinling project https://review.openstack.org/550787
12:56:22 openstackgerrit Mohammed Naser proposed openstack/python-openstackclient master: Add support to list image members https://review.openstack.org/554590
12:56:35 mnaser dtroyer: addressed comments ^ :)
13:06:20 openstackgerrit OpenStack Proposal Bot proposed openstack/os-service-types master: Updated from OpenStack Service Type Authority https://review.openstack.org/554910
13:23:46 openstackgerrit Chris Dent proposed openstack/api-wg master: Update the errors guidance to use service-type for code https://review.openstack.org/554921
14:00:05 cdent elmiko, edleafe: if you could traverse this stack once again, now that I've fixed (simply thanks to edleafe) the Version issues, please and thank you: https://review.openstack.org/#/c/550251/
14:01:02 cdent with cherries and cookies and other nice things
14:03:10 edleafe cdent: Can you explain why we want version 0.0 to match?
14:03:34 cdent heh, my irc client turned that into a smiley
14:03:45 cdent a scared smiley
14:03:52 edleafe I can see that!
14:04:50 cdent because say you're experimenting with building your microversion service, and you just decide to make your first version a non-version, but you want the code to work, thus (0, 0)
14:06:31 edleafe so is that just something that you thought about, or is it some kind of microversion standard?
14:06:54 cdent the former
14:07:19 cdent or at least if it is a standard, I wasn't aware of it
14:08:12 edleafe well, I can see it being useful, but only if that behavior is, you know, mentioned somewhere
14:08:37 cdent perfection is the enemy of the done
14:08:44 cdent this isn't the end game, it is the start
14:08:52 cdent $cliche the $cliche
14:09:32 cdent I'd really just like to release early and often, and spread the audience
14:10:19 edleafe yeah, I wasn't talking about holding things up; I just wanted to share the love at some point.
14:10:35 elmiko cdent: ack, added to the queue
14:11:43 cdent thanks, both of you
14:12:07 elmiko and ++ to edleafe, i totally was not nearly snarky enough on those reviews XD
14:13:55 cdent /o\
14:15:40 openstackgerrit Monty Taylor proposed openstack/python-openstacksdk master: Provide OpenStackConfigException backwards compat https://review.openstack.org/554615
14:15:56 mordred frickler: ^^ added closes-bug footer if you have a second to re-approve
14:19:36 frickler mordred: sure, thx
14:23:10 frickler mordred: I'm seeing seemingly unrelated network test failures on this patch to which I only added a reno. looks like there may be a timing issue in the first test at least http://logs.openstack.org/33/540633/3/check/osc-functional-devstack/456e2b0/job-output.txt.gz#_2018-03-21_13_02_19_573915
14:23:46 frickler dtroyer: also related to your recent novaclient workarounding I think ^^
14:25:36 elmiko cdent: the test_version_zero_no_defaults test, am i reading that docstring wrong, it says "we want to it be able to match" but the test is assertFalse. i'm a little confused
14:26:05 cdent elmiko: it's a pasto, I'll fix it in a follow up if that's cool
14:26:10 cdent it's basically the wrong docstring
14:26:15 elmiko ahh, cool
14:26:18 elmiko thought so

Earlier   Later