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’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Copy the CI tooling from the RFC repo #62

Closed
bkchr opened this issue Jun 4, 2024 · 13 comments · Fixed by #66 or #68
Closed

Copy the CI tooling from the RFC repo #62

bkchr opened this issue Jun 4, 2024 · 13 comments · Fixed by #66 or #68

Comments

@bkchr
Copy link
Contributor

bkchr commented Jun 4, 2024

I think we want to have the same automation as on the RFC repo? If yes, we can just copy it from there.

@rzadp are you interested in helping with this?

@rzadp
Copy link

rzadp commented Jun 4, 2024

@rzadp are you interested in helping with this?

Sure, I'll try to help. Is the process analogical to the RFC one - voting on a remark in the collectives parachain?

BTW, did you see this one?

@acatangiu
Copy link
Contributor

hmm, I was thinking to join the xcm RFC process with the general RFC repo - aka, drop this XCM-specific RFC and use the one in https://github.com/polkadot-fellows/RFCs

This repo would only hold the spec.

@bkchr
Copy link
Contributor Author

bkchr commented Jun 4, 2024

hmm, I was thinking to join the xcm RFC process with the general RFC repo - aka, drop this XCM-specific RFC and use the one in https://github.com/polkadot-fellows/RFCs

Ahh fine as well! How did you imagined the merge rights for this repo?

@bkchr
Copy link
Contributor Author

bkchr commented Jun 4, 2024

BTW, did you see this one?

Yeah I have seen this one. We can probably merge it, but the future move to use subsquare/polkassembly directly will probably render it useless.

@acatangiu
Copy link
Contributor

Ahh fine as well! How did you imagined the merge rights for this repo?

I don't have a strong preference. We can work backwards from the flow:

  1. someone proposes RFC to https://github.com/polkadot-fellows/RFCs, RFC is approved and merged (using the process there)
  2. someone (else) opens PR to https://github.com/polkadot-fellows/xcm-format for updating spec based on approved RFC - this can be done by anyone, but it is ultimately fellows responsibility to maintain the spec
  3. someone (else) opens PR for updating implementation in https://github.com/polkadot-fellows/runtimes based on the approved RFC and updated spec - this can be done by anyone, but it is ultimately fellows responsibility to maintain the system runtimes

(for step (2) we can automate opening "PR - update spec according to RFCxxx" then some fellow or community member can bring it over the finish line, but this is nice to have)

based on above I would use the same merge rights for https://github.com/polkadot-fellows/xcm-format as for https://github.com/polkadot-fellows/runtimes and we can reuse the process and automation/scripts from there.
The "owners" are the same, the "level/points" required could be same as changes to relay chain runtimes directories for example.

wdyt?

@bkchr
Copy link
Contributor Author

bkchr commented Jun 4, 2024

Sounds fine to me. I think at some point someone proposed to directly modify the spec. This would merge the RFC with modifying the spec. However, it may makes it more complicated to review the actual changes.

@franciscoaguirre
Copy link
Contributor

The RFCs make it better to review the actual changes, I'd put them in the already existing process and then update the spec

@acatangiu
Copy link
Contributor

use the same merge rights for https://github.com/polkadot-fellows/xcm-format as for https://github.com/polkadot-fellows/runtimes and we can reuse the process and automation/scripts from there

@rzadp @alvicsam @Bullrich who can help with this?

@rzadp
Copy link

rzadp commented Jun 6, 2024

The point of re-using the polkadot-fellows/RFCs for these XCM RFCs makes total sense to me.
But I have some questions about what happens next. I didn't have much to do with XCM so might be missing something.


@acatangiu If we want to use same access rights, automations, etc as polkadot-fellows/runtimes here in polkadot-fellows/xcm-format, then why don't we place the XCM format into the runtimes repo?
Since it's the same owner, same group responsible, and Polkadot&Kusama are supposed to be implementing it.

This xcm-format repo could be a read-only automated copy, if need be.

If we seperate the "staging" XCM spec from the one that is currently implemented in Polkadot/Kusama, we would have the advantage of easily correlating the XCM format that is the basis of any given release (or any given commit) in the runtimes repo.
Right now, if I look at any old release of Polkadot/Kusama, do I have a way of finding the exact version of XCM format used to implement it? Other than browsing timestamps.


So I would propose the following modified steps:

  1. An xcm-format.md file exists in runtimes, and the implementation in the repo is based on it.
  2. [no changes] someone proposes RFC to https://github.com/polkadot-fellows/RFCs, RFC is approved and merged (using the process there)
  3. someone (else) opens PR to https://github.com/polkadot-fellows/runtimes for updating spec based on approved RFC - creating a new xcm-format-pending.md
  4. someone (else) opens PR for updating implementation in https://github.com/polkadot-fellows/runtimes based on the approved RFC and pending spec. In the same PR, mv xcm-format-pending.md > xcm-format.md is done.

What do you all think? Of course it falls apart if the XCM format MUST be detached from the runtimes.

@acatangiu
Copy link
Contributor

then why don't we place the XCM format into the runtimes repo?

The XCM specification and the system chain runtimes are totally different things. It just happens that the fellowship is responsible for maintaining both. That's why they both live under the polkadot-fellows GH organization.

The Polkadot Protocol spec is also under the purview of the fellowship, but it is, same as XCM, larger scope than system chain runtimes so they're not lumped together.

As such, I would not move the spec under https://github.com/polkadot-fellows/runtimes, it deserves its own repo with its own audience (albeit same maintainers as the runtimes repo).

(Also, a clarification missing from my original post: Actually the XCM spec is also not really implemented in https://github.com/polkadot-fellows/runtimes, but in https://github.com/paritytech/polkadot-sdk, and eventually integrated into https://github.com/polkadot-fellows/runtimes)

@rzadp
Copy link

rzadp commented Jun 11, 2024

Ok, fair enough.

So my takeaway is that:

  1. We do NOT want to copy the CI tooling from the RFC repo - but rather move with the xcm RFCs over there.
  2. We want to duplicate the merging rights/automations in this xcm-format repository from runtimes - mainly we're talking about the review-bot, if I understand correctly.

@acatangiu
Copy link
Contributor

@rzadp yes, correct!

  1. will be handled by the RFC authors - they will have to migrate their RFCs from here to https://github.com/polkadot-fellows/RFCs
  2. this is where we need help from you 🙏

@Bullrich
Copy link
Contributor

I created a ticket:

Once this is completed, I can install Review-Bot.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
5 participants