Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We鈥檒l occasionally send you account related emails.

Already on GitHub? Sign in to your account

馃彈 Integrate open PRs 馃毀 #486

Closed
sbernard31 opened this issue Nov 3, 2020 · 32 comments
Closed

馃彈 Integrate open PRs 馃毀 #486

sbernard31 opened this issue Nov 3, 2020 · 32 comments

Comments

@sbernard31
Copy link
Contributor

There is a lot of open PR, this issue aims to centralize discussion about integrating this PR in master to put the project back on track.

Following @sbertin-telular, the first one should be : #423.

@sbernard31 sbernard31 changed the title Integrate open PRs 馃彈 Integrate open PRs 馃毀 Nov 3, 2020
@sbernard31 sbernard31 pinned this issue Nov 3, 2020
@rettichschnidi
Copy link
Contributor

rettichschnidi commented Nov 4, 2020

@sbernard31 Proposal:

@sbernard31
Copy link
Contributor Author

sbernard31 commented Nov 4, 2020

Now that PR #489 got merged, all PRs should get rebased on current master in order to run this new check. See PR #471 for an example. This would also have the nice side-effect of knowing for which PRs the creator is still interested in/willing to invest time into.

@rettichschnidi, It sounds a good idea, but should we integrate "our" PR first ? I mean PR written by you and your teammates and @sbertin-telular. The idea is to avoid that too many people resolve conflict at the same time and so by the way generate new one ? (I'm not sure I'm clear 馃)

While being at it, asking the PR creator to squash the commits sounds worthwhile to me too (but maybe a bit too much to ask as for this project there seems to be no such requirement/recommendation for PRs as of today)

We can ask to politely squash commit and also say that if they can not we will do it (once PR is rebased without conflict, it's easy to squash.)

About recommendation, I will create an issue to discuss about creating/adding all "classic" community documentation.

@sbertin-telular
Copy link
Contributor

I was going to suggest #427 after #423 is merged, but if we're going to do "ours" first that would be #430.

@sbernard31
Copy link
Contributor Author

sbernard31 commented Nov 4, 2020

@sbertin-telular, @rettichschnidi, Do you want to review your PR each other before I integrate it in master ? or for those old PRs I can merge it without review ?

@sbertin-telular
Copy link
Contributor

I think it would be good for them to be reviewed. I'm not sure if they've had much of a review before now.

@sbernard31
Copy link
Contributor Author

#423 is integrated. Let's go for the next one.

@sbernard31
Copy link
Contributor Author

#430 is integrated. Let's go for the next one.

@sbertin-telular
Copy link
Contributor

#432 is next.

@sbernard31
Copy link
Contributor Author

sbernard31 commented Nov 13, 2020

#432 is integrated. Let's go for the next one. (I'm a kind of github robot 馃)

@sbertin-telular
Copy link
Contributor

I think the next couple are #435 and #446. Maybe #461 after that if it is ready. I also have many changes I haven't created pull requests for yet.

@tuve
Copy link
Contributor

tuve commented Nov 16, 2020

I noticed i have to rebase #461 . There has also been an internal code review from some of the embedded developers in a different department for their patch set from an older version of wakaama. Not all is related to #461 but other parts could possible be reported as bugs I guess? The person who has done this as part of his assignment will not be available to report here as he is leaving for another assignment next week, but I will make sure we get it in to the code review, either by me or some colleague of mine.

@sbernard31
Copy link
Contributor Author

#435 is integrated. Let's go for the next one.

@sbernard31
Copy link
Contributor Author

#437 and #396 was closed because it should be fixed by #435.

@sbernard31
Copy link
Contributor Author

#446 is integrated. Let's go for the next one.

@sbertin-telular
Copy link
Contributor

Most of my remaining changes for LWM2M 1.1 started from the assumption that #431 would be merged first. I don't have PRs open for those changes yet. Do we want to skip #431 for now and I'll adjust my changes or get it merged? I do think it helps with understanding the structure of the code vs having everything under core.

@sbernard31
Copy link
Contributor Author

I just look at #431, it is a code reorg/refactoring. Currently this PR is not in mergeable state (lot of conflicts).

As this is a lot of file moves, I suspect resolving conflicts will be pretty much the same effort than rewrite it. Anyway in all case I think maybe you should create a new PR (then cherry-picked david's commit or rewrite it on your own).

Maybe before to do that we should see if there is not PR (not specially yours) which could be easier to integrate before ?

WDYT ?

@sbertin-telular
Copy link
Contributor

Probably best to get other open PRs merged first. That will make it easier for the contributors. I can create a new PR to do the reorg later.

@tuve are you close on being ready to merge #461, or should we work to get other PRs merged first?

@tuve
Copy link
Contributor

tuve commented Dec 6, 2020

I think @qleisan can finish up the comments otherwise I should be able to prioritize #461 later half of the upcoming week.

@qleisan
Copy link

qleisan commented Dec 7, 2020

@tuve I'm looking into it already. Found some issues and would like to convince myself that it works as intended.

@sbertin-telular
Copy link
Contributor

#463, #483, and #485 look good to me.

@sbernard31
Copy link
Contributor Author

#463, #483 are integrated in master. (it seems that #485 was already fixed).

@sbernard31
Copy link
Contributor Author

5 opened PR left. We are on good track. 馃憤

@qleisan
Copy link

qleisan commented Dec 8, 2020

It is good with clarity on what issues that must be fixed together with block transfer #461 and what can be fixed after it has been merged

@qleisan
Copy link

qleisan commented Jan 25, 2021

@sbertin-telular @sbernard31 what is remaining to implement in order to have lwm2m 1.1 functionality once #461 and "your branch" is delivered? Can we make a plan for a 1.1 release (content and testing). An official release (with good quality) will send a strong signal to the community.

@sbernard31
Copy link
Contributor Author

@qleisan, I suppose this is the roadmap item(6.) from #488.
Do not hesitate to create a dedicated issue and you can discuss what is desired for a next release, what is already implemented and so what is missing.

Once you agree on it. Either we keep the issue as roadmap or you can create wiki dedicated page(s).

Just as an example at Leshan side we have this :

@sbernard31
Copy link
Contributor Author

By the way as soon as #461 is merged, I understand that #431 should be the next one. (@sbertin-telular is on it if I'm not wrong)

@sbernard31
Copy link
Contributor Author

#461 is integrated. @sbertin-telular #431 is still the next one in the plan ?

@sbertin-telular
Copy link
Contributor

I just created #528 to replace #431.

@sbertin-telular
Copy link
Contributor

It has been a long time since I've seen no open pull requests. Good to see it. I'll start working to create more with my LWM2M 1.1 changes.

@sbernard31
Copy link
Contributor Author

Just to let you know that currently, I integrate PR as soon as I get 1 approval.
Feel free to review after integration and comment PR anyway.

If you think I should do differently, maybe waiting 1 day(or more) after approval to let a chance to get more review, let me know 馃檹 !

@sbernard31
Copy link
Contributor Author

Do we still need this issue or can I close it now ?
I guess we are back on track about PR integration and don't need it anymore ? 馃

@qleisan
Copy link

qleisan commented Mar 31, 2021

I think it is ok to close this issue, the major PR:s have been coordinated and merged

@sbernard31 sbernard31 unpinned this issue Apr 1, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

5 participants