Support Git as only VCS

[qwerty287]

Only support git as vcs and remove the corresponding env var CI_REPO_SCM.

[6543]

e.g. #2355 integration

and more could come, so it does not add much maintinance to keep it as generic as it is now

[6543]

why was this initially proposed at all? e.g. what's the reason behind?

[6543]

I see the current architecture that we can support more than git even as far as an "selling" over other cicd systems

[xoxys]

e.g. #2355 integration

How are other git forges that were implemented by an external plugin related to removing none-git vsc systems?

I see the current architecture that we can support more than git even as far as an "selling" over other cicd systems

I don't see it as a selling point especially not as e can't even handle all existing bugs for the existing feature set already.

why was this initially proposed at all?!? e.g. what's the reason behind?

I would like to kindly ask you again not to communicate with so many punctuation marks. It's hard to read and a somewhat “aggressive” tone.

[6543]

PS: still thanks for thinking how to slim down and remove legacy code :)

[qwerty287]

I don't see it as a selling point especially not as e can't even handle all existing bugs for the existing feature set already.

That's also my point. Radicle has a git "bridge" so it should work.

And we can't have everything as selling point. At some point, we should have some limitations, we can't build the "egg-laying woolly milk sow" (don't know if there's a proper English saying 😅).

[6543]

we can't build the "egg-laying woolly milk sow"

agree but this is realy not much code and losing it is a shame ...

[6543]

I as an owner wana veto this change

[pat-s]

Without any opinion: Should this PR be closed then?

[xoxys]

If a single owner has the power to block PRs at any time, yes.

[qwerty287]

Well, @anbraten approved it...

[woodpecker-bot]

Deployment of preview was torn down

[6543]

well the argument stands: either merge it, in this case I will also merge things, that got denyed from me, because an owner is for it ... or block it untill veto is removed in this case untill I'm no owner anymore. next election is in 1 months btw. ...

so yes our community gudelines dont have a clear statement of when the owners dissagree - wich would not be a problem if we would have $owner.mod(2) = 0 (and if there is something stated as majority of owners agree ..), .. witch we currently dont.

so I personally interpret it as veto. ?

[xoxys]

I just dont understand your point. Right now there is not a single none-git forge supported in woodpecker core right? So this code is unused and useless. To keep it just because we might or might not support none-git forges doesnt make sense to me. Even if we would like to support such forges we could also do it by an official extension in the same way we do it for plugins.

[pat-s]

As this is now more a political issue than a code-one, it would be great if you @6543 @anbraten could discuss the general governance part how to proceed in such situations.

@qwerty287 I think there's not much value in further branch updated until this is resolved?

[qwerty287]

Tbh I personally think owners should only exist because they're necessary, but they shouldn't have additional permissions like a veto. All maintainers should have the same possibilities. If the maintainers can't come to a conclusion, it's fine that an owner's vote counts more or something like this. You still need owners of course because we have servers and online accounts etc. And thus I also think the name "owners" is not the best, "admins" would be better.

[qwerty287]

Extensions, LOL. Even config service has not been working since start of the year.

I don't get your point. Addon forges should work and for cloning you can use anything as I said already.

[pat-s]

@qwerty287 I am not really sure what this PR is waiting on. I see you rebase frequently. So, we are stuck here somehow and nobody wants to act I guess?

[qwerty287]

Yes well I kinda still wait for a decision. I just update my branch to not get too many conflicts, but that's no big task...
We should discuss this owner-veto thing or do a vote among maintainers about this change.

[xoxys]

@woodpecker-ci/maintainers please vote on this post for this PR. 👍 merge 👎 close

[pat-s]

The tendency seems to be in clear favor or removal judging from the last comments and overall discussion. To move on here and get this into 3.0 as it is breaking, I'll merge this now.

[qwerty287]

@6543 requested changes so it can't be merged until the review is dismissed...

[xoxys]

Well then dismiss it?

[pat-s]

I guess only owners can?

[xoxys]

No

As discussed

[xoxys]

Done now.

[codecov]

Codecov Report

Attention: Patch coverage is 72.72727% with 3 lines in your changes missing coverage. Please review.

Project coverage is 28.24%. Comparing base (b4055e8) to head (f3b794f).
Report is 4 commits behind head on main.

Files with missing lines	Patch %	Lines
...r/store/datastore/migration/023_remove_repo_scm.go	70.00%	2 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4346      +/-   ##
==========================================
- Coverage   28.25%   28.24%   -0.01%     
==========================================
  Files         399      400       +1     
  Lines       28238    28232       -6     
==========================================
- Hits         7978     7974       -4     
+ Misses      19555    19552       -3     
- Partials      705      706       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[6543]

well I'm still against this cange but It looks like i have to creat a pull and rebase it untill all others see it worth adding it again ... :/