*** tosky has quit IRC | 00:06 | |
*** iurygregory has quit IRC | 01:32 | |
*** vishalmanchanda has joined #openstack-requirements | 03:52 | |
*** johnsom has quit IRC | 04:46 | |
*** johnsom has joined #openstack-requirements | 04:49 | |
*** johnsom has quit IRC | 05:18 | |
*** johnsom has joined #openstack-requirements | 05:19 | |
*** evrardjp has quit IRC | 05:33 | |
*** evrardjp has joined #openstack-requirements | 05:33 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/requirements master: Updated from generate-constraints https://review.opendev.org/762296 | 06:23 |
---|---|---|
*** johnsom has quit IRC | 06:27 | |
*** johnsom has joined #openstack-requirements | 06:27 | |
*** sboyron has joined #openstack-requirements | 07:22 | |
*** rm_work has quit IRC | 07:26 | |
*** rm_work has joined #openstack-requirements | 07:28 | |
*** rpittau|afk is now known as rpittau | 08:05 | |
*** iurygregory has joined #openstack-requirements | 08:52 | |
*** tosky has joined #openstack-requirements | 08:56 | |
*** dtantsur|afk is now known as dtantsur | 09:04 | |
*** odyssey4me has quit IRC | 09:50 | |
*** odyssey4me has joined #openstack-requirements | 09:50 | |
rpittau | good morning everyone! Maybe someone here can help me finding some documentation on lower-constraints? Having a clear explanation or at least a "best practice" about its content would be great, I feel there's still some ambiguity on that left | 10:27 |
*** e0ne has joined #openstack-requirements | 11:08 | |
hberaud | rpittau: this doc seems a good starting point but I suppose you already seen it => https://docs.openstack.org/project-team-guide/dependency-management.html#solution | 11:51 |
hberaud | rpittau: also notice this tooling => https://docs.openstack.org/project-team-guide/dependency-management.html#build-lower-constraints it could be worth to dive deeper into it to learn things and grab the "best practices" | 12:06 |
hberaud | I think that the lack of related documentation about this topic is related to this tool that aim to abstract lower-constraints management | 12:07 |
hberaud | hopefully it help you | 12:08 |
*** Luzi has joined #openstack-requirements | 12:36 | |
rpittau | hberaud: thanks! I saw both already, I was more interested in some sort of final clarification on the content of lower-constraints. | 13:02 |
rpittau | What I know is that it should have at least the min versions of all the requirements for a project, excluding test/doc requirements | 13:02 |
rpittau | what stays in the limbo is how to treat the hidden dependencies, so all the libs that the requirements rely on | 13:02 |
hberaud | I see | 13:03 |
hberaud | rpittau: AFAIK they could be managed by using `pip -c <constraints-file.txt>` however I think that if you want to manage hidden dependencies then using "lower-constraints.txt" isn't the right place as it will be tested/compared with global requirements, so maybe a solution could be to define your own hidden-constraints.txt to check reqs that are not managed by openstack/requirements | 13:08 |
hberaud | not sure to be clear :) | 13:08 |
rpittau | hberaud: it's clear, I think I was not clear! :D | 13:12 |
rpittau | what I'm looking for is an understanding on what needs to be or would best to have in lower-constraints, should we also include a requirement that is not a direct requirement of a project but a requirement of one of its requirements? | 13:12 |
rpittau | For example, ironic doesn't depend on six, but some of its requirements do, should we include six in lower-constraints? | 13:12 |
hberaud | rpittau: The first thought that comes to me while reading your comment is "explicit is better than implicit". It could allow you to manage your reqs with finer granularity, however it could be also overkill to redefine everything | 13:17 |
hberaud | rpittau: example of more or less similar topic => http://lists.openstack.org/pipermail/openstack-discuss/2020-October/017740.html | 13:18 |
rpittau | hberaud: mmm yeah, kind of the same thought | 13:19 |
rpittau | hberaud: thanks for the exchange :) | 13:19 |
hberaud | if it works well for you since a while without explicit definition then I think you can continue like that | 13:20 |
hberaud | (with six by example) | 13:20 |
hberaud | I think it's depends on the kind of requirement, by example on a messaging POV if you rely on kombu you could need to define explicitelly a version of py-amqp too to pull a specific version for some specific technical needs | 13:22 |
hberaud | rpittau: you're welcome | 13:23 |
rpittau | hberaud: yeah, I guess we probably need to differentiate case by case | 13:23 |
hberaud | exactly | 13:24 |
*** hberaud has quit IRC | 13:44 | |
*** hberaud has joined #openstack-requirements | 13:49 | |
*** Luzi has quit IRC | 14:18 | |
*** lbragstad__ has quit IRC | 14:37 | |
*** lbragstad has joined #openstack-requirements | 14:40 | |
smcginnis | rpittau: The intention is that lower-constraints contains all dependencies that get pulled in, not just direct dependencies. | 14:45 |
smcginnis | So anything that ends up being in a pip freeze output after doing an install, so it can constrain the entire set of dependencies. | 14:45 |
rpittau | smcginnis: yes, but that intention is nowhere to be found in the docs, as far as I can see :) | 14:46 |
smcginnis | rpittau: Yeah, I can't recall what is all in the docs at this point. The original ML post from Doug Hellmann explained the intention much more thoroughly. Should probably copy paste all of that into the docs. :D | 14:49 |
rpittau | :) | 14:49 |
hberaud | could be worth | 14:52 |
*** bnemec has quit IRC | 16:16 | |
*** iurygregory has quit IRC | 16:29 | |
*** bnemec has joined #openstack-requirements | 16:50 | |
*** iurygregory has joined #openstack-requirements | 17:19 | |
*** e0ne has quit IRC | 17:22 | |
*** rpittau is now known as rpittau|afk | 17:27 | |
*** dtantsur is now known as dtantsur|afk | 20:17 | |
*** smcginnis has quit IRC | 20:24 | |
*** sboyron has quit IRC | 21:58 | |
*** smcginnis has joined #openstack-requirements | 21:59 | |
*** hberaud has quit IRC | 23:36 | |
*** hberaud has joined #openstack-requirements | 23:40 | |
*** tosky has quit IRC | 23:58 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!